System

Abstract

A system is provided.SOLUTION: Means for collecting past power consumption data, weather data, and power price data, means for cleansing the collected data and complementing missing data, means for performing supply and demand prediction of power consumption using a machine learning model, means for predicting a power price based on a result of the supply and demand prediction, means for generating a plurality of power procurement plans based on the supply and demand prediction and the price prediction, means for presenting the generated plans to a user in an interactive format, and means for receiving a selection and a desired condition of the user, the system includes a means for formulating a demand and supply plan and a control plan on the basis of a selected plan, a means for adjusting demand and supply in real time in order to secure compliance with planned value balancing, and a means for automatically executing various operations required for power procurement.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] For modern businesses and households, procuring optimal, green electricity is an important management issue, but achieving this requires extensive specialized knowledge and human resources. Managing price fluctuations and supply-demand balances in the electricity market is difficult, and this complexity is increasing, especially in today's world where renewable energy is becoming more and more popular. For this reason, businesses and households without specialized knowledge have a hard time procuring optimal electricity that balances affordability and greenness. [Means for solving the problem]

[0005] In order to solve the above problems, the present invention provides the following means. First, a means for collecting past power consumption data, weather data, and power price data is provided. Next, a means for cleansing the collected data and supplementing missing data is provided. Next, a means for predicting power consumption supply and demand using a machine learning model is provided. Also, a means for predicting power prices based on the results of the supply and demand prediction is provided. Furthermore, a means for generating multiple power procurement plans based on the supply and demand prediction and the price prediction is provided. Finally, a means for presenting the generated plans to a user in an interactive format is provided. Also, a means for receiving the user's selection and desired conditions and formulating a supply and demand plan and a control plan based on the selected plan is provided. Furthermore, a means for adjusting supply and demand in real time to ensure compliance with planned values ​​is provided. Finally, a means for automatically executing various operations required for power procurement is provided. In this way, it is possible for businesses and households to achieve efficient and effective power procurement without specialized knowledge.

[0006] "Power consumption data" refers to historical data on the power consumed by businesses and households, and includes information such as date, consumption amount, and time.

[0007] "Weather Data" means past and forecast weather information obtained from the Japan Meteorological Agency or weather forecasting services, including date, time, temperature, precipitation, wind speed, etc.

[0008] "Electricity price data" refers to historical data of trading prices in the electricity market, and includes the date and time of the transaction, the unit price, the amount of supply, and the like.

[0009] "Collection Methods" refers to the processes and tools for collecting historical electricity consumption data, weather data, and electricity price data.

[0010] "Data cleansing" refers to the process of removing outliers and duplicates from raw data and preparing it for analysis.

[0011] "Missing data imputation" refers to the process of filling in gaps in a dataset using statistical methods and predictive models.

[0012] "Supply and demand forecasting" refers to the process of predicting the future balance between electricity demand and supply based on past data.

[0013] A "machine learning model" refers to an algorithm or statistical model that analyzes large amounts of data to generate new knowledge and make predictions.

[0014] "Electricity price forecasting" refers to the process of predicting future electricity market prices based on the results of supply and demand forecasts.

[0015] "Power Procurement Plan" refers to a specific power procurement strategy, including the proportion of renewable energy and price conditions, prepared based on power consumption forecasts and price forecasts.

[0016] "Interactive presentation" refers to a method of presenting information through an interactive interface in a way that is easy for users to understand and operate.

[0017] "Adherence to planned values ​​at the same time and in the same amount" refers to managing the balance between supply and demand in real time and matching the electricity supply plan with actual consumption.

[0018] "Supply and demand adjustment" refers to the process of adjusting the difference between supply and demand in real time, thereby aiming to prevent the occurrence of imbalance charges.

[0019] "Automatic execution of operations" refers to a system that automatically executes various procedures and operations related to electricity procurement.

[0020] "Renewable energy" refers to energy sources such as wind, solar, hydroelectric, and biomass that reduce environmental impact in a sustainable manner.

[0021] "Selection and desired conditions" refers to the elements that a user selects from among the electricity procurement plans and conditions based on their own preferences and needs.

[0022] "Supply and demand plans and control plans" refer to plans that establish specific supply methods for future electricity demand and the operational procedures required for supply. [Brief explanation of the drawings]

[0023] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0024] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0025] First, the terms used in the following description will be explained.

[0026] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0027] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0028] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0029] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0030] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0031] [First embodiment]

[0032] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0033] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0034] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0035] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0036] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0037] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0038] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0039] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0040] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0041] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0042] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0043] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0044] This invention is a system that enables businesses and households to procure optimal and green electricity without requiring specialized knowledge. This system consists of a server, terminals, and users, and collects, analyzes, and predicts past electricity consumption data, weather data, and electricity price data, and then generates, presents, and executes an appropriate electricity procurement plan.

[0045] Main components and their operation

[0046] 1. Server Operation

[0047] The server collects historical electricity consumption data, weather data, and electricity price data, cleanses this data, and fills in missing data. To do this, the server obtains the necessary data through APIs and database connections.

[0048] Using the collected and pre-processed data, the server applies machine learning models to generate supply and demand forecasts and price predictions, using time series forecasting algorithms such as LSTM and ARIMA models.

[0049] Based on the forecast results, multiple power procurement plans (e.g., a 100% renewable energy plan, a price-focused plan, etc.) are generated. These plans are optimized based on evaluation criteria such as cost, environmental impact, and stability.

[0050] 2. Device Operation

[0051] The server generates multiple electricity procurement plans and sends them to the terminal, which then presents them to the user in an interactive format. The details of each plan are displayed in graphs and text, providing the user with an interface that is easy to understand and operate.

[0052] The terminal receives the user's selection and desired conditions and transmits them to the server, where the user can enter specific conditions (such as maximum cost or minimum required percentage of renewable energy).

[0053] 3. User Operation

[0054] Users can review multiple electricity procurement plans offered through their terminal and select the plan that best suits their needs.

[0055] Based on the plan selected by the user, the user inputs the necessary adjustments and specific plan details. The information entered by the user is sent via the terminal to the server, which then uses that information to formulate supply and demand plans and control plans.

[0056] Explanation of program processing

[0057] Data collection and preprocessing

[0058] The server collects historical electricity consumption data, weather data, and electricity price data from smart meters, the Japan Meteorological Agency's API, market databases, etc.

[0059] The collected data is cleansed and missing data is filled using linear imputation and statistical methods.

[0060] Supply and demand forecasts and price forecasts

[0061] The server uses machine learning models to forecast supply and demand, including LSTM and ARIMA, to predict future electricity consumption, taking seasonal fluctuations and trends into account.

[0062] Based on the results of the supply and demand forecast, the server also predicts electricity prices, calculating predicted price fluctuations according to the supply and demand balance using ARIMA models and other methods.

[0063] Generate electricity procurement plans

[0064] The server generates multiple procurement plans based on supply and demand forecasts and price forecasts, and the plans are optimized taking into account factors such as the proportion of renewable energy, cost, and stability of supply.

[0065] Plan presentation and selection

[0066] The terminal presents the plan sent from the server to the user and provides information in an interactive format. The interface is designed to meet the user's desired conditions and needs.

[0067] The user selects the most suitable plan and enters their desired conditions. The device then sends this information to the server.

[0068] Supply and demand planning

[0069] Based on the user's selections and desired conditions, the server creates detailed supply and demand plans and control plans, including supply schedules and load balancing plans.

[0070] Automatic execution of operations

[0071] The server ensures compliance with planned values ​​and adjusts supply and demand in real time. All necessary procedures and operations are automatically performed, optimizing power procurement without user intervention.

[0072] Specific examples

[0073] For businesses

[0074] 1. Data Collection:

[0075] The server collects electricity consumption data for the past year for the company, local weather data, and electricity market price data.

[0076] 2. Supply and demand forecast:

[0077] The server uses the LSTM model to predict supply and demand for the next year, including increased consumption in the winter.

[0078] 3. Price Prediction:

[0079] The server predicts future electricity prices using an ARIMA model based on supply and demand forecasts.

[0080] 4. Plan generation and presentation:

[0081] The server generates three plans: a 100% renewable energy plan, a 70% renewable energy plan, and a price-conscious plan, and the terminal presents them to the company's energy management officer.

[0082] 5. Select a plan:

[0083] The user (energy manager) selects Plan B (70% renewable energy plan) and enters specific desired conditions.

[0084] 6. Supply and demand planning:

[0085] The server will create a detailed supply and demand plan and control plan based on Plan B.

[0086] For ordinary households

[0087] 1. Data Collection:

[0088] The server collects the household's electricity consumption data and weather forecast data for the past six months.

[0089] 2. Supply and demand forecast:

[0090] The server uses a time series prediction model to predict power supply and demand up to one month in advance.

[0091] 3. Price Prediction:

[0092] The server uses the ARIMA model to predict electricity prices.

[0093] 4. Plan generation and presentation:

[0094] The server generates plans that use only renewable energy and have low night-time rates, and the terminal presents them to the household.

[0095] 5. Select a plan:

[0096] The user selects Plan C (a plan with a cheaper nighttime rate) and transmits the selection to the server via the terminal.

[0097] 6. Supply and demand planning:

[0098] The server develops a power consumption control plan based on Plan C and submits it to the power company.

[0099] This will provide a system that enables businesses and households to procure electricity effectively without the need for specialized knowledge.

[0100] The processing flow will be explained below.

[0101] Step 1:

[0102] The server collects historical electricity consumption data, weather data, and electricity price data, including data from smart meters, the Japan Meteorological Agency's API, market databases, etc.

[0103] Step 2:

[0104] The server cleanses the collected data, removing outliers and duplicate data to make it more reliable, and then fills in missing data using statistical methods and linear interpolation.

[0105] Step 3:

[0106] The server uses machine learning models to forecast supply and demand. Specifically, it uses LSTM and ARIMA models to input past electricity consumption data and weather data to predict future electricity consumption.

[0107] Step 4:

[0108] The server predicts electricity prices based on the results of the supply and demand forecast, and calculates future electricity prices using ARIMA models and other methods, taking into account the supply and demand balance.

[0109] Step 5:

[0110] The server generates multiple electricity procurement plans based on supply-demand and price forecast data, including factors such as the proportion of renewable energy, price, and stability.

[0111] Step 6:

[0112] The server sends the generated electricity procurement plans to the terminal, which provides the user with an interactive interface that displays the details of each plan in graphs and text.

[0113] Step 7:

[0114] The user compares multiple electricity procurement plans offered through the terminal and selects the plan that best suits their needs. The user then inputs the selected plan and desired conditions into the terminal.

[0115] Step 8:

[0116] The terminal sends the user's selection and desired conditions to the server, which uses the selection and conditions to formulate a supply and demand plan in the next step.

[0117] Step 9:

[0118] The server creates detailed supply and demand plans and control plans based on the user's selections. The supply and demand plans include supply schedules and load balancing plans, and are designed to ensure compliance with planned values.

[0119] Step 10:

[0120] The server adjusts supply and demand in real time to ensure compliance with the planned simultaneous balance. If an imbalance occurs, adjustment measures are automatically implemented.

[0121] Step 11:

[0122] The server automatically executes the various operations required for power procurement, including submitting plans to the power company and making the necessary reports after procurement is confirmed.

[0123] Step 12:

[0124] The electricity procurement is officially executed based on the plan selected by the user. The server monitors the actual electricity consumption data and collects and analyzes feedback to improve the accuracy of the next forecast.

[0125] In this way, the entire system works together to achieve optimal power procurement.

[0126] Example 1

[0127] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0128] Conventional power procurement systems require specialized knowledge, making it difficult for many businesses and households to achieve optimal power procurement. Furthermore, the accuracy of supply and demand forecasts and power price forecasts is low, and environmentally friendly renewable energy use is often not sufficiently considered. Real-time supply and demand adjustments and the provision of information in a format that is easy for users to understand are also insufficient.

[0129] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0130] In this invention, the server includes means for collecting past electricity consumption data, weather data, and electricity price data, means for cleansing the collected data and filling in missing data, means for forecasting electricity consumption supply and demand using a machine learning model, means for forecasting electricity prices based on the supply and demand forecast results, means for generating multiple electricity procurement plans based on the supply and demand forecast and price forecast, means for interactively presenting the generated plans to a user, means for receiving the user's selection and desired conditions and formulating a supply and demand plan and a control plan based on the selected plan, means for adjusting supply and demand in real time to ensure compliance with planned values, means for automatically executing various operations required for electricity procurement, means for providing graphs and text information in a user-understandable format, and means for performing long-term and short-term data analysis for forecasting and planning. This enables users without specialized knowledge to efficiently procure optimal and green electricity.

[0131] "Past power consumption data" is data showing the history of power consumption, and is a record of power usage and consumption patterns over a certain period of time.

[0132] "Weather Data" means data that includes meteorological information such as temperature, humidity, wind speed, and precipitation, and indicates weather conditions for a particular region and period of time.

[0133] "Electricity price data" refers to data showing fluctuations in electricity prices and past price history in the electricity market.

[0134] "Cleansing" refers to the process of organizing collected data and correcting or removing erroneous or incomplete information.

[0135] "Missing data imputation" refers to filling in missing information in an incomplete data set using estimates or other information.

[0136] A "machine learning model" is an algorithm that can learn patterns and rules from data and make predictions and classifications.

[0137] "Demand and supply forecasting" is the process of predicting the future balance of electricity consumption and supply.

[0138] "Electricity price forecasting" refers to predicting future electricity market prices based on the results of supply and demand forecasts, etc.

[0139] An "electricity procurement plan" is a plan that shows how electricity will be procured based on predicted electricity supply and demand and prices.

[0140] "Interactive" refers to a format in which a user interacts with a system to exchange information.

[0141] "Planning based on a selected plan" means creating a detailed supply and demand plan and control plan based on the power procurement plan selected by the user.

[0142] "Adherence to planned values" means making adjustments to match the balance of electricity supply and demand in real time.

[0143] "Automatic execution of operations" means that the system automatically performs the operations and procedures required for power procurement.

[0144] "Providing graphs and text information" means visualizing and presenting information in a way that is easy for users to understand.

[0145] "Long-term and short-term data analysis" is the process of analyzing long-term and short-term data to aid in forecasting and planning.

[0146] This invention is a system that enables businesses and households to procure electricity in an optimal and environmentally friendly manner without requiring specialized knowledge. This system consists of a server, terminals, and users, and collects, analyzes, and predicts past electricity consumption data, weather data, and electricity price data, and then generates, presents, and executes an appropriate electricity procurement plan.

[0147] Server Operations

[0148] Data collection

[0149] The server collects historical electricity consumption data, weather data, and electricity price data using smart meters, the Japan Meteorological Agency's API, and market databases. For example, when obtaining electricity consumption data from smart meters, the server collects the data using a communication protocol.

[0150] Data Cleansing and Imputation

[0151] The server cleanses the collected data and fills in missing data using linear imputation and statistical methods. For example, the cleansing process removes outliers and ensures consistency.

[0152] Supply and demand forecast

[0153] The server uses software such as TensorFlow to build LSTM and ARIMA models, inputs the collected data, and predicts future electricity supply and demand, for example, taking into account peak electricity consumption in the summer.

[0154] Price Prediction

[0155] The server uses the ARIMA model to predict electricity prices based on the results of the supply and demand forecast, using past price data and predicted supply and demand data.

[0156] Generate electricity procurement plans

[0157] The server generates multiple electricity procurement plans based on the results of supply and demand forecasts and price forecasts. The plans include the proportion of renewable energy, cost, stability of supply, etc. For example, it generates a 100% renewable energy plan, a 70% renewable energy plan, a price-focused plan, etc.

[0158] Device operation

[0159] Plan presentation and selection

[0160] The terminal interactively presents the electricity procurement plans sent from the server to the user. The details of each plan are displayed to the user in graphs and text, providing an interface that is easy to understand and operate. For example, a visual graph showing that Plan A has a high proportion of renewable energy use is displayed.

[0161] Collecting user preferences

[0162] The terminal receives the user's desired conditions and sends them to the server. The user can input specific conditions (e.g., maximum cost or minimum required percentage of renewable energy).

[0163] User operations

[0164] Review and select a plan

[0165] Users can review multiple electricity procurement plans offered through their devices and select the plan that best suits their needs. For example, an energy manager might select a plan that uses 70% renewable energy.

[0166] Enter detailed information

[0167] The user inputs the necessary adjustments and specific plan details based on the selected plan. The input information is sent via the terminal to the server, which then uses that information to create a supply-demand plan and a control plan.

[0168] Example prompts for generative AI models

[0169] "Generate a 100% renewable energy electricity procurement plan based on the past year's electricity consumption data, weather data, and electricity price data."

[0170] "Predict electricity consumption for the next year using an LSTM model and present a price-focused electricity procurement plan."

[0171] This invention provides businesses and households with effective and sustainable power supply options tailored to their specific requirements and preferences, enabling them to procure power efficiently without specialized knowledge.

[0172] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0173] Step 1: Data collection

[0174] The server collects historical electricity consumption data, weather data, and electricity price data from smart meters, the Japan Meteorological Agency's API, and the electricity market database. As a specific example, it sends a "GET / electricity / consumption" request from the smart meter to obtain consumption data for the past year and obtains the latest weather data from the Japan Meteorological Agency's API. As a result, it obtains a set of electricity consumption data, weather data, and electricity price data.

[0175] Step 2: Data cleansing and impregnation

[0176] The server cleanses the collected data and imputes missing data. It uses the collected data set as input, and as part of the data cleansing, removes outliers and produces clean data. Missing data is imputed using linear imputation and statistical methods. For example, statistical methods can be used to impute missing values ​​with the mean value, resulting in a complete and consistent data set.

[0177] Step 3: Supply and demand forecast

[0178] The server uses the cleansed and complemented data to input it into an LSTM model built using TensorFlow. The data is then processed by converting the electricity consumption data and weather data into a time series format and applying it to the LSTM model. The LSTM model then predicts future electricity consumption and outputs the supply and demand pattern for the next fiscal year as a result of the prediction.

[0179] Step 4: Price prediction

[0180] The server inputs the results of the supply and demand forecast and uses the ARIMA model to predict electricity prices. It analyzes and processes the supply and demand forecast results and past price data, and predicts future electricity market prices based on this. The output is a predicted future electricity price dataset.

[0181] Step 5: Generate a power procurement plan

[0182] The server generates an electricity procurement plan based on the results of supply and demand forecasts and price forecasts. For example, it generates a 100% renewable energy plan, a 70% renewable energy plan, and a price-focused plan. Supply and demand forecast data and price forecast data are used as input, and optimization is performed when generating each plan, taking into account factors such as the renewable energy ratio, cost, and supply stability. Detailed data for each generated plan is obtained as output.

[0183] Step 6: View and select a plan

[0184] The terminal interactively presents the electricity procurement plans sent from the server to the user. As input, it uses multiple procurement plan data received from the server. Specifically, it displays the details of each plan in an easy-to-understand graph and text format, allowing the user to easily compare and consider them. As output, it generates an information display screen to be presented to the user.

[0185] Step 7: Collect user preferences

[0186] The user inputs desired conditions (for example, maximum cost or minimum percentage of renewable energy) through a terminal. As input, the user's desired condition data is entered into the terminal. The terminal sends this to the server and requests the server to design a plan based on the conditions. As output, the user's desired condition data sent to the server is obtained.

[0187] Step 8: Develop supply and demand plans

[0188] The server formulates detailed supply and demand plans and control plans based on the user's selections and desired conditions. The inputs include the user's desired condition data and detailed data of the selected power procurement plan. To formulate the supply and demand plan, it creates a supply schedule and load balancing plan and issues instructions to various power supply devices and systems. The output generates detailed supply and demand plan data and control plan data.

[0189] Step 9: Automate operations

[0190] The server automatically executes power procurement and supply-demand adjustments based on the formulated plan. Detailed supply-demand and control plan data is used as input. Specifically, it monitors the supply-demand balance in real time and adjusts the procurement plan as needed. The output is an optimized power procurement operation.

[0191] (Application example 1)

[0192] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0193] In modern society, households and businesses lack the knowledge and time to procure optimal, green energy, making efficient and environmentally friendly energy management difficult. In particular, there is a need for a system that can automatically generate plans that take into account the proportion and cost of renewable energy and propose them in a format that is easy for users to understand. Furthermore, it is necessary to adjust supply and demand in real time based on daily electricity supply and demand forecasts and price forecasts, but this is difficult to do without specialized knowledge.

[0194] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0195] In this invention, the server includes: means for collecting past electricity consumption data, weather data, and electricity price data; means for cleansing the collected data and filling in missing data; means for forecasting electricity supply and demand using a machine learning model; means for forecasting electricity prices based on the supply and demand forecast results; means for generating multiple electricity procurement plans based on the supply and demand forecast and price forecast; means for interactively presenting the generated plans to a user; means for receiving the user's selection and desired conditions and formulating a supply and demand plan and a control plan based on the selected plan; means for adjusting supply and demand in real time to ensure compliance with planned values; means for automatically executing various operations required for electricity procurement; means for operating on a smartphone; and data processing means for automatically generating multiple plans with renewable energy usage rates and cost-focused features based on the forecast data. This enables households and businesses to achieve optimal and green electricity procurement and efficient energy management without specialized knowledge.

[0196] "Historical power consumption data" is data that records the amount of power used during a specific period of time.

[0197] "Weather data" refers to data that records meteorological conditions such as temperature, humidity, precipitation, and wind speed.

[0198] "Electricity price data" is data indicating the price of electricity in the market within a certain period of time.

[0199] "Data cleansing methods" are processing methods for removing inaccurate information from collected data and appropriately filling in missing values.

[0200] A "machine learning model" is an algorithm that learns patterns in data and predicts future data.

[0201] "Means for forecasting supply and demand" refers to a method for predicting the balance between electricity demand and supply in advance.

[0202] A "means for forecasting electricity prices" is a method for forecasting future electricity prices based on the results of supply and demand forecasts.

[0203] An "electricity procurement plan" is a specific plan regarding the method and conditions for purchasing electricity.

[0204] "Means of presenting information to the user in an interactive format" refers to a method of displaying information in a way that is easy for the user to understand and providing appropriate options through dialogue.

[0205] "Means for formulating supply and demand plans and control plans" refers to methods for determining specific electricity usage plans and adjustment methods based on the predicted supply and demand balance and price.

[0206] "Means for adjusting supply and demand in real time" refers to a method for monitoring the balance of supply and demand for electricity in real time and making adjustments as necessary.

[0207] "Means for automatic execution" refers to a method for automatically performing various operations related to power procurement without user intervention.

[0208] "Means operating on a smartphone" refers to applications and systems that operate on a smartphone and are available for use by the user.

[0209] "Renewable energy usage rate" refers to the percentage of renewable energy in the total electricity used.

[0210] "Cost-sensitive features" are plan characteristics that emphasize minimizing costs.

[0211] "Data processing means" refers to the method used to process collected data and convert it into the form required for forecasting and plan generation.

[0212] This invention is a system that enables homes and businesses to procure optimal and green electricity without requiring specialized knowledge. This system is primarily composed of a server, terminals, and users, and the interaction between these enables efficient power management.

[0213] Server Operations

[0214] The server collects historical energy consumption, weather, and electricity price data, cleansing it, and filling in missing data using APIs and database connections, such as from smart meters and weather information services.

[0215] Based on the collected and preprocessed data, the server applies machine learning models to forecast electricity supply and demand and prices. This uses the SARIMAX model, a time-series forecasting algorithm. The SARIMAX model analyzes past electricity consumption and weather data to predict future electricity supply and demand.

[0216] Based on the forecast results, multiple power procurement plans are generated. These plans include plans that prioritize renewable energy usage rates and costs, and are optimized according to the user's needs. The generated plans are presented to the user via a smartphone application.

[0217] Device operation

[0218] The device, specifically a smartphone, interactively presents the user with multiple electricity procurement plans sent from the server. The user can then review the details of each plan through the smartphone app's intuitive interface. For example, graphs and text descriptions clearly show each plan's cost, environmental impact, and supply stability.

[0219] Users input their desired conditions (e.g., maximum cost or minimum renewable energy usage rate) and select the optimal plan. The device sends the user's selection and desired conditions to the server, which then uses that information to create detailed supply and demand plans and control plans.

[0220] User operations

[0221] Users can review multiple electricity procurement plans offered through a smartphone application, select the plan that best suits their needs, and enter specific conditions (e.g., maximum cost, minimum required percentage of renewable energy) based on the plan they select.

[0222] After selecting a plan, the server creates a detailed supply and demand plan based on the information entered by the user. This plan includes a supply schedule and load balancing plan. Supply and demand are also adjusted in real time to ensure compliance with the planned balance. This allows users to achieve optimal power procurement without the need for specialized knowledge or effort.

[0223] Specific examples

[0224] For example, a server collects electricity consumption data, weather data, and electricity price data from a household for the past six months. An LSTM model is used to predict electricity supply and demand for up to one month in advance, and an ARIMA model is used to predict future electricity prices based on the supply and demand forecast. The generated plan is checked on a smartphone app, and when the user selects a plan with a 70% renewable energy usage ratio, the server uses that information to create a detailed supply and demand plan.

[0225] Prompt Sentence Examples

[0226] For example, give the generative AI model the following prompt:

[0227] "I want to develop an application that uses a home's electricity consumption, weather data, and electricity price data to propose the optimal electricity procurement plan. Can you predict future electricity consumption and prices based on past data, generate plans that include features such as the proportion of renewable energy used and focus on cost, and provide prompts to present them to the user?"

[0228] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0229] Step 1:

[0230] The server collects past electricity consumption data, weather data, and electricity price data. It obtains data from various APIs and databases as input and obtains raw data as output. Specifically, it obtains electricity consumption data from smart meters, weather data from weather information services, and electricity price data from market databases.

[0231] Step 2:

[0232] The server cleanses the collected data and fills in missing data. It uses the raw data obtained in step 1 as input and obtains cleansed data as output. Specifically, it converts the data into a data frame, removes inaccurate information, and interpolates missing values.

[0233] Step 3:

[0234] The server uses a machine learning model to predict power consumption. It uses cleansed data as input and obtains future power demand and supply forecast data as output. Specifically, it analyzes past data using the SARIMAX model and predicts future consumption.

[0235] Step 4:

[0236] The server predicts electricity prices based on the results of the supply and demand forecast. Supply and demand forecast data is used as input, and future electricity price forecast data is obtained as output. Specifically, the server uses an ARIMA model based on the balance between electricity supply and demand to predict fluctuations in electricity prices.

[0237] Step 5:

[0238] The server generates multiple electricity procurement plans based on supply and demand forecasts and price forecasts. It uses supply and demand forecast data and price forecast data as input, and obtains multiple electricity procurement plans as output. Specifically, it processes data to generate plans that prioritize renewable energy usage rates and costs.

[0239] Step 6:

[0240] The server sends the generated plan to a device (smartphone), which then presents it to the user in an interactive format. The generated electricity procurement plan is used as input, and the details of the plan are displayed as output through an interface that is easy for the user to understand. Specifically, information such as the plan's cost, environmental impact, and stability is provided through graphs and text explanations.

[0241] Step 7:

[0242] Users can check the details of each plan through a smartphone app, input their desired conditions, and then select the most suitable plan. The system uses the user's desired conditions (e.g., maximum cost or minimum required renewable energy share) as input, and the user's selection is sent to the server as output. Specific actions include clicking options and inputting desired conditions.

[0243] Step 8:

[0244] The server formulates supply and demand plans and control plans based on the user's selections and desired conditions. It uses the user's selections and desired conditions as input and obtains a detailed supply and demand plan as output. Specific operations include formulating a supply schedule and load balancing plan that is optimal for the selected plan.

[0245] Step 9:

[0246] The server adjusts supply and demand in real time to ensure compliance with the planned balance and automatically executes various necessary operations. It uses the formulated supply and demand plan as input and performs adjustments to maintain an optimal supply and demand balance as output. Specific operations include monitoring and adjusting power usage in real time.

[0247] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0248] This invention is a system that enables businesses and households to procure optimal, green electricity without requiring specialized knowledge. This system consists of a server, terminals, and users. It collects, analyzes, and predicts past electricity consumption data, weather data, and electricity price data, and then generates, presents, and executes an appropriate electricity procurement plan. Furthermore, it combines an emotion engine that recognizes the user's emotions to improve the user experience.

[0249] Main components and their operation

[0250] 1. Server Operation

[0251] The server collects historical electricity consumption data, weather data, and electricity price data, cleans the data, and fills in missing data. It retrieves the necessary data through APIs and database connections, removes outliers and duplicates, and fills in missing data using linear imputation and statistical methods.

[0252] Using the collected and pre-processed data, the server applies machine learning models (e.g., LSTM and ARIMA models) to make supply and demand forecasts and price predictions.

[0253] Based on the forecast results, multiple power procurement plans are generated, which are optimized based on criteria such as renewable energy share, cost, environmental impact, and stability.

[0254] 2. Device Operation

[0255] The server generates multiple electricity procurement plans and sends them to the terminal, which then presents them to the user in an interactive format. The details of each plan are displayed in graphs and text, providing the user with an interface that is easy to understand and operate.

[0256] The device is equipped with an emotion engine that recognizes the user's emotions and estimates their emotional state from their facial expressions, voice, input actions, etc. This allows the device to adjust the way the plan is presented to them, taking into account the user's emotions.

[0257] The device receives the user's selection and desired conditions and sends that information to the server, where the user can enter specific conditions (e.g., maximum cost or minimum required percentage of renewable energy).

[0258] 3. User Operation

[0259] Users can review multiple electricity procurement plans offered through their devices and select the plan that best suits their needs. The emotion engine monitors users' emotions in real time and assists in the plan selection process.

[0260] Based on the plan selected by the user, the user inputs the necessary adjustments and specific plan details. The information entered by the user is sent via the terminal to the server, which then uses that information to formulate supply and demand plans and control plans.

[0261] Explanation of program processing

[0262] Data collection and preprocessing

[0263] The server collects historical electricity consumption data, weather data, and electricity price data from smart meters, the Japan Meteorological Agency's API, market databases, etc.

[0264] The collected data is cleansed and missing data is filled using linear imputation and statistical methods.

[0265] Supply and demand forecasts and price forecasts

[0266] The server uses machine learning models to forecast supply and demand, including LSTM and ARIMA, to predict future electricity consumption, taking seasonal fluctuations and trends into account.

[0267] Based on the results of the supply and demand forecast, the server also predicts electricity prices, calculating predicted price fluctuations according to the supply and demand balance using ARIMA models and other methods.

[0268] Generate electricity procurement plans

[0269] The server generates multiple procurement plans based on supply and demand forecasts and price forecasts, and the plans are optimized taking into account factors such as the proportion of renewable energy, price, and stability of supply.

[0270] Presenting a plan and utilizing the emotion engine

[0271] The terminal presents the plan sent from the server to the user and provides information in an interactive format. The interface is designed to meet the user's desired conditions and needs.

[0272] The device's emotion engine recognizes the user's facial and vocal expressions and adjusts the way the plan is presented based on this. For example, if the user looks confused, it will provide a simpler explanation.

[0273] Plan selection and supply and demand planning

[0274] The user selects the most suitable plan and enters their desired conditions. The device then sends this information to the server.

[0275] The server creates detailed supply and demand plans and control plans based on the user's selections. The supply and demand plans include supply schedules and load balancing plans, and are designed to ensure compliance with planned values.

[0276] Automatic execution of operations

[0277] The server ensures compliance with planned values ​​and adjusts supply and demand in real time. All necessary procedures and operations are automatically performed, optimizing power procurement without user intervention.

[0278] Specific examples

[0279] For businesses

[0280] 1. Data Collection:

[0281] The server collects electricity consumption data for the past year for the company, local weather data, and electricity market price data.

[0282] 2. Supply and demand forecast:

[0283] The server uses the LSTM model to predict supply and demand for the next year, including increased consumption in the winter.

[0284] 3. Price Prediction:

[0285] The server predicts future electricity prices using an ARIMA model based on supply and demand forecasts.

[0286] 4. Plan generation and presentation:

[0287] The server generates three plans: a 100% renewable energy plan, a 70% renewable energy plan, and a price-conscious plan, and the terminal presents them to the company's energy management officer.

[0288] 5. Plan Selection and Emotion Engine:

[0289] When the energy manager (user) selects Plan B (70% renewable energy plan), the emotion engine analyzes the manager's facial expressions and voice and adjusts the explanation of the plan as necessary.

[0290] 6. Formulating a supply plan:

[0291] The server will create a detailed supply and demand plan and control plan based on Plan B.

[0292] For ordinary households

[0293] 1. Data Collection:

[0294] The server collects the household's electricity consumption data and weather forecast data for the past six months.

[0295] 2. Supply and demand forecast:

[0296] The server uses a time series prediction model to predict power supply and demand up to one month in advance.

[0297] 3. Price Prediction:

[0298] The server uses the ARIMA model to predict electricity prices.

[0299] 4. Plan generation and presentation:

[0300] The server generates plans that use only renewable energy and have low night-time rates, and the terminal presents them to the household.

[0301] 5. Plan Selection and Emotion Engine:

[0302] When a user selects Plan C (a plan with lower nighttime rates), the emotion engine analyzes the user's facial expressions and voice and provides appropriate information.

[0303] 6. Formulating a supply plan:

[0304] The server develops a power consumption control plan based on Plan C and submits it to the power company.

[0305] In this way, the entire system works together to achieve optimal power procurement, and by combining it with an emotion engine, it improves the user experience.

[0306] The processing flow will be explained below.

[0307] Step 1:

[0308] The server collects historical electricity consumption data, weather data, and electricity price data. Specifically, it obtains consumption data from smart meters, weather data from the Japan Meteorological Agency's API, and electricity price data from a market database.

[0309] Step 2:

[0310] The server cleanses the collected data, removing outliers and duplicate data to make it more reliable, and then fills in missing data using linear interpolation and statistical methods.

[0311] Step 3:

[0312] The server uses machine learning models to forecast supply and demand. Specifically, it uses LSTM and ARIMA models to input past electricity consumption data and weather data to predict future electricity consumption.

[0313] Step 4:

[0314] The server predicts electricity prices based on the results of the supply and demand forecast. It uses the ARIMA model to calculate future electricity prices while taking into account the supply and demand balance.

[0315] Step 5:

[0316] The server generates multiple electricity procurement plans based on supply and demand forecasts and price forecasts, and optimizes the plans by taking into account factors such as the proportion of renewable energy, prices, and supply stability.

[0317] Step 6:

[0318] The server sends the generated electricity procurement plans to the terminal, which provides the user with an interactive interface that displays the details of each plan in graphs and text.

[0319] Step 7:

[0320] The device's emotion engine analyzes the user's facial expressions and voice input in real time while the plan is being presented, recognizing their emotional state. If the user shows confusion or anxiety, the interface will provide a simpler explanation or additional information.

[0321] Step 8:

[0322] The user compares multiple electricity procurement plans offered through the terminal and selects the plan that best suits their needs. The user then inputs the selected plan and desired conditions into the terminal.

[0323] Step 9:

[0324] The terminal transmits the user's selections and desired conditions to the server, which uses the selections and conditions to develop supply and demand plans and control plans.

[0325] Step 10:

[0326] The server creates detailed supply and demand plans and control plans based on user selections, including factors such as supply schedules, load balancing plans, and renewable energy utilization rates.

[0327] Step 11:

[0328] The server adjusts supply and demand in real time to ensure compliance with the planned value. If the supply and demand balance is disrupted, necessary adjustments are automatically made.

[0329] Step 12:

[0330] The server automatically executes the various operations required for power procurement, including submitting plans to the power company and making the necessary reports after procurement is confirmed.

[0331] Step 13:

[0332] The electricity procurement is officially executed based on the plan selected by the user. The server monitors the actual electricity consumption data and collects and analyzes feedback to improve the accuracy of the next forecast.

[0333] Example: In the case of a company

[0334] Data collection and preprocessing

[0335] Step 1:

[0336] The server collects electricity consumption data from the past year, local weather data, and electricity market price data.

[0337] Step 2:

[0338] The server cleanses the collected data and fills in any missing parts. A reliable data set is created through a series of data cleansing processes.

[0339] Supply and demand forecasts and price forecasts

[0340] Step 3:

[0341] The server uses the LSTM model to predict supply and demand for the next year, including increased consumption in the winter.

[0342] Step 4:

[0343] The server uses an ARIMA model to predict future electricity prices based on supply and demand forecasts.

[0344] Plan generation and presentation

[0345] Step 5:

[0346] The server generates three plans: a 100% renewable energy plan, a 70% renewable energy plan, and a price-conscious plan.

[0347] Step 6:

[0348] The server sends the generated plan to the terminal, which then presents it to the company's energy management officer.

[0349] Plan selection and emotional engine utilization

[0350] Step 7:

[0351] The device's emotion engine analyzes the facial expressions and voice of the energy manager to recognize their emotional state in real time.

[0352] Step 8:

[0353] The energy manager (user) selects Plan B (70% renewable energy plan) and enters specific desired conditions.

[0354] Formulating and implementing a supply plan

[0355] Step 9:

[0356] The terminal sends the selection of the person in charge and the desired conditions to the server.

[0357] Step 10:

[0358] The server will create a detailed supply and demand plan and control plan based on Plan B.

[0359] Step 11:

[0360] The server ensures compliance with planned values ​​and adjusts supply and demand in real time.

[0361] Step 12:

[0362] The server will automatically handle the necessary plan submissions and procedures.

[0363] Step 13:

[0364] Power procurement is officially carried out according to the plan selected by the energy manager (user).

[0365] Example: For an average household

[0366] Data collection and preprocessing

[0367] Step 1:

[0368] The server collects the household's electricity consumption data and weather forecast data for the past six months.

[0369] Step 2:

[0370] The server cleanses the collected data and fills in any missing parts.

[0371] Supply and demand forecasts and price forecasts

[0372] Step 3:

[0373] The server uses a time series prediction model to predict power supply and demand up to one month in advance.

[0374] Step 4:

[0375] The server predicts the electricity price based on the supply and demand forecast.

[0376] Plan generation and presentation

[0377] Step 5:

[0378] The server generates renewable energy only plans and plans with cheaper nighttime rates.

[0379] Step 6:

[0380] The server sends the generated plan to the terminal, which then presents it to the household.

[0381] Plan selection and emotional engine utilization

[0382] Step 7:

[0383] The device's emotion engine analyzes the user's facial expressions and voice to recognize their emotional state in real time.

[0384] Step 8:

[0385] The user selects Plan C (a plan with a cheaper nighttime rate) and transmits the selection to the server via the terminal.

[0386] Formulating and implementing a supply plan

[0387] Step 9:

[0388] The terminal transmits the user's selection and desired conditions to the server.

[0389] Step 10:

[0390] The server will create a detailed supply and demand plan and control plan based on Plan C.

[0391] Step 11:

[0392] The server ensures compliance with planned values ​​and adjusts supply and demand in real time.

[0393] Step 12:

[0394] The server automatically handles the necessary plan submissions and procedures.

[0395] Step 13:

[0396] Power procurement is officially carried out according to the plan selected by the user.

[0397] In this way, the entire system works together to achieve optimal power procurement, and by combining it with an emotion engine, it improves the user experience.

[0398] Example 2

[0399] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0400] Conventional energy management systems are inefficient in predicting power consumption and optimizing procurement plans, making it difficult to flexibly respond to user emotions and needs. This requires advanced expertise for businesses and households to procure power in an optimal and environmentally friendly manner. Furthermore, supply-demand adjustments and price forecasts are often inaccurate, leading to insufficient adherence to planned simultaneous balancing. It is necessary to resolve these issues and improve the user experience.

[0401] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0402] In this invention, the server includes means for collecting past electricity consumption data, weather data, and electricity price data, means for cleansing the collected data and filling in missing data, means for forecasting electricity consumption supply and demand using a machine learning model, means for forecasting electricity prices based on the supply and demand forecast results, means for generating multiple electricity procurement plans based on the supply and demand forecast and the price forecast, means for interactively presenting the generated plans to a user, means for recognizing the user's emotions and adjusting the method of presenting the plans depending on the user's emotional state, means for receiving the user's selection and desired conditions and formulating a supply and demand plan and a control plan based on the selected plan, means for adjusting supply and demand in real time to ensure compliance with planned values, and means for automatically executing various operations required for electricity procurement. This enables users to procure electricity in an optimal and environmentally friendly manner without specialized knowledge, and the emotion engine can be used to improve the user experience.

[0403] "Historical electricity consumption data" refers to historical information about electricity usage over a specific period of time, and is digital data obtained from smart meters and passive measuring devices.

[0404] "Weather data" refers to information about past and current weather conditions, including temperature, humidity, precipitation, wind speed, and other factors.

[0405] "Electricity Price Data" means data that includes historical and current information about electricity unit prices and trading prices in a particular market or region.

[0406] "Cleansing" is a data processing method for removing outliers and duplicate data from collected data to improve its quality.

[0407] "Missing data imputation" is the process of filling in gaps in a dataset using statistical or linear interpolation techniques to restore the data to a complete state.

[0408] A "machine learning model" is an algorithm that learns patterns and regularities from data and performs predictions and classifications. Examples include LSTM and ARIMA.

[0409] "Supply and demand forecasting" refers to predicting future electricity demand and supply, a process carried out using machine learning models.

[0410] "Electricity price forecasting" is the process of predicting future fluctuations in electricity prices based on supply and demand forecasts.

[0411] An "electricity procurement plan" is a specific plan for how to procure electricity, and includes factors such as the proportion of renewable energy used, costs, and stability of supply.

[0412] The "means for presenting to the user in an interactive format" refers to a method for interactively displaying the generated electricity procurement plan to the user using graphs and text.

[0413] "Means for recognizing emotions and adjusting the way plans are presented according to the emotional state" refers to technology that has the function of analyzing emotions from the user's facial expressions, voice, etc., and changing the way information is presented according to those emotions.

[0414] "Adherence to planned values ​​and quantities" refers to making real-time adjustments to ensure that planned power supply and consumption match.

[0415] "Means for adjusting supply and demand in real time" refers to a system for instantly adjusting the balance between supply and demand.

[0416] "Means for automatically executing various operations required for power procurement" refers to technology that allows the system to automatically perform operations and procedures for power procurement and management without user intervention.

[0417] This invention is a system that enables businesses and households to procure electricity in an optimal and environmentally friendly manner without requiring specialized knowledge. This system consists of a server, terminals, and users, and performs data collection, data analysis, forecasting, plan generation, plan presentation, emotion recognition, supply and demand planning, and operation automation.

[0418] System configuration and processing overview

[0419] Server-based data collection and cleansing

[0420] The server collects historical electricity consumption data, weather data, and electricity price data from smart meters, the Japan Meteorological Agency's API, market databases, etc.

[0421] Cleanse the collected data and fill in missing data, for example by removing outliers and using linear interpolation and statistical methods to fill in missing data.

[0422] Supply and demand forecasts and price forecasts

[0423] The server uses machine learning models (e.g., LSTM and ARIMA) to predict supply and demand based on past data, taking into account seasonal fluctuations and trends to predict future electricity consumption.

[0424] Based on the results of the supply and demand forecast, the server also predicts electricity prices, predicting price fluctuations according to the supply and demand balance.

[0425] Generate electricity procurement plans

[0426] The server generates multiple electricity procurement plans based on supply-demand and price forecasts, and these plans are optimized taking into account factors such as the proportion of renewable energy, cost, and stability of supply.

[0427] Presenting plans and utilizing emotion recognition

[0428] The terminal presents the electricity procurement plans sent from the server to the user, and details of each plan are displayed in graphs and text, providing an interface that is easy to understand and operate.

[0429] The device is equipped with an emotion engine that recognizes the user's emotions and estimates their emotional state from their facial expressions, voice, input actions, etc. This allows the device to adjust the way the plan is presented to them, taking into account the user's emotions.

[0430] User-Server Interaction

[0431] Users can review the various electricity procurement plans offered through their devices and select the plan that best suits their needs, for example, the "70% renewable energy plan," and specify the maximum monthly cost.

[0432] The terminal receives the user's selection and desired conditions and sends this information to the server, which then uses this information to formulate supply and demand plans and control plans.

[0433] The server formulates a supply and demand plan and a control plan based on the plan selected by the user, and adjusts supply and demand in real time based on this.

[0434] Example: In the case of a company

[0435] 1. Data Collection:

[0436] The server collects a company's electricity consumption data from the past year, local weather data, and electricity market price data.

[0437] 2. Supply and demand forecast:

[0438] The server uses the LSTM model to predict supply and demand for the next year, including increased consumption in the winter.

[0439] 3. Price Prediction:

[0440] The server predicts future electricity prices using an ARIMA model based on supply and demand forecasts.

[0441] 4. Plan generation and presentation:

[0442] The server generates three plans: a 100% renewable energy plan, a 70% renewable energy plan, and a price-conscious plan, and the terminal presents them to the company's energy management officer.

[0443] 5. Plan Selection and Emotion Engine:

[0444] When the energy manager (user) selects Plan B (70% renewable energy plan), the emotion engine analyzes the manager's facial expressions and voice and adjusts the explanation of the plan as necessary.

[0445] 6. Developing supply and demand planning and control plans:

[0446] The server will create a detailed supply and demand plan and control plan based on Plan B.

[0447] Example: For an average household

[0448] 1. Data Collection:

[0449] The server collects the household's electricity consumption data and weather forecast data for the past six months.

[0450] 2. Supply and demand forecast:

[0451] The server uses a time series prediction model to predict power supply and demand up to one month in advance.

[0452] 3. Price Prediction:

[0453] The server uses the ARIMA model to predict electricity prices.

[0454] 4. Plan generation and presentation:

[0455] The server generates plans that use only renewable energy and have low night-time rates, and the terminal presents them to the household.

[0456] 5. Plan Selection and Emotion Engine:

[0457] When a user selects Plan C (a plan with lower nighttime rates), the emotion engine analyzes the user's facial expressions and voice and provides appropriate information.

[0458] 6. Supply and demand planning:

[0459] The server develops a power consumption control plan based on Plan C and submits it to the power company.

[0460] Prompt Sentence Examples

[0461] Below are some example prompts to explain each element or process of the system to the generative AI model:

[0462] User: How do we forecast our company's electricity consumption for next year?

[0463] Terminal: The server collects historical electricity consumption data and weather data, and uses an LSTM model to predict supply and demand, then uses an ARIMA model to predict prices.

[0464] In this way, the entire system works together to achieve optimal power procurement, and the emotional engine is combined to improve the user experience.

[0465] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0466] Program processing steps

[0467] Step 1: Data collection

[0468] The server collects historical electricity consumption data, weather data, and electricity price data.

[0469] Input: Data sources (smart meter API, Japan Meteorological Agency API, market database)

[0470] Specific operations: Retrieve historical electricity consumption data from the smart meter API and store it in a database. Retrieve local weather data using the Japan Meteorological Agency API and store it in a database. Query and retrieve historical electricity price data from the market database.

[0471] Output: Raw dataset for cleansing and imputation

[0472] Step 2: Data cleansing and imputation

[0473] The server cleanses the collected data and completes any missing data.

[0474] Input: raw dataset

[0475] What it does: Detect and remove outliers from datasets, remove duplicates, and impute missing data using linear interpolation and statistical methods.

[0476] Output: Clean dataset

[0477] Step 3: Supply and demand forecast

[0478] The server uses a machine learning model (e.g., LSTM) to predict supply and demand.

[0479] Input: Clean dataset

[0480] What it does: Train an LSTM model and use the clean data as input to predict future electricity demand, for example, predicting electricity consumption for the next year in 30-minute increments.

[0481] Output: Power supply and demand forecast data

[0482] Step 4: Forecasting electricity prices

[0483] The server predicts the electricity price based on the results of the supply and demand forecast.

[0484] Input: Power supply and demand forecast data

[0485] Specific operation: Uses an ARIMA model to predict future electricity prices based on supply and demand forecast data. For example, predicts electricity prices for the next year in 30-minute increments.

[0486] Output: Power price forecast data

[0487] Step 5: Generate a power procurement plan

[0488] The server generates a plurality of power procurement plans based on the supply and demand forecast and the price forecast.

[0489] Input: Power supply and demand forecast data, power price forecast data

[0490] Specific actions: Taking into account the proportion of renewable energy, cost, and stability of supply, three procurement plans will be created: a 100% renewable energy plan, a 70% renewable energy plan, and a price-focused plan.

[0491] Output: Power Procurement Plan

[0492] Step 6: Present your plan

[0493] The terminal presents the power procurement plan transmitted from the server to the user.

[0494] Input: Power Procurement Plan

[0495] Specific operation: Through the GUI, details of each plan (such as trends in electricity usage and the percentage of renewable energy used) are displayed to the user in graphs and text.

[0496] Output: The plan presented to the user

[0497] Step 7: Emotion recognition and plan adjustment

[0498] The terminal recognizes the user's emotions and adjusts the way it presents plans.

[0499] Input: User's facial expression data, voice data

[0500] Specific operation: Analyzes the user's facial expressions and voice through a camera and microphone, and if the user is confused, provides a more concise explanation and adds diagrams and charts.

[0501] Output: Optimized presentation of information to the user

[0502] Step 8: Select a plan and enter your desired conditions

[0503] The user selects the desired plan and sets specific conditions.

[0504] Input: Offered plan, user's desired conditions

[0505] Specific operation: The user selects the desired plan through the device and enters the maximum monthly cost and minimum required renewable energy percentage.

[0506] Output: User's selections and preferences

[0507] Step 9: Develop supply and demand plans and control plans

[0508] The server formulates specific supply and demand plans and control plans based on the user's selections.

[0509] Input: User selections and preferences

[0510] Specific operation: Creates a supply schedule and load balancing plan based on the conditions of the selected plan and saves it in the database.

[0511] Output: Supply and demand planning and control planning

[0512] Step 10: Real-time supply and demand adjustment and operational automation

[0513] The server ensures compliance with planned values, adjusts supply and demand in real time, and automatically executes various necessary operations.

[0514] Input: Supply and demand plan and control plan

[0515] Specific operations: Monitor the supply-demand balance and take action such as shifting the operation of specific devices in the event of an oversupply. Operations are automated and executed.

[0516] Output: Optimized power supply and demand balance

[0517] (Application example 2)

[0518] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0519] Conventional power procurement systems have made it difficult to optimize power consumption in factories and homes, particularly in promoting the use of renewable energy. Furthermore, the information required for users to select an energy plan is often diverse and difficult to understand. This makes it difficult to select the optimal power procurement plan, resulting in sluggish progress in improving power consumption efficiency and reducing costs. Furthermore, these systems have not been able to utilize emotion recognition to improve the user experience.

[0520] The specification process by the specification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting past power consumption data, weather data, and power price data, means for cleansing the collected data and supplementing missing data, and means for predicting power consumption supply and demand using a machine learning model. This makes it possible to provide optimal and green power procurement plans for power consumption in factories and homes.

[0521] The server further includes means for predicting electricity prices based on the results of the supply and demand forecast, means for generating multiple electricity procurement plans based on the supply and demand forecast and the price forecast, means for interactively presenting the generated plans to the user, means for receiving the user's selection and desired conditions and formulating a supply and demand plan and a control plan based on the selected plan, means for adjusting supply and demand in real time to ensure compliance with planned values, means for automatically executing various operations required for electricity procurement, and means for recognizing the user's emotions and adjusting the method of presenting the plans based on the emotions. This makes it possible to select the optimal electricity procurement plan according to the user's emotions, thereby achieving efficient electricity consumption and cost reduction.

[0522] "Past electricity consumption data" refers to information about the history of electricity used by consumers such as businesses and households over a certain period of time.

[0523] "Weather data" is information indicating weather conditions in a specific region or period, and includes temperature, humidity, rainfall, wind speed, and the like.

[0524] "Electricity Price Data" means information about the price of electricity in a particular market or period.

[0525] "Means for collection" refers to means for acquiring electricity consumption data, weather data, and electricity price data using devices such as servers and sensors.

[0526] "Cleansing and missing data imputation methods" are methods for removing outliers and duplicate data from collected data and filling in missing data using linear imputation and statistical methods.

[0527] "Method for forecasting supply and demand of electricity consumption using machine learning models" refers to a method for forecasting future electricity consumption using machine learning algorithms such as LSTM (Long Short-Term Memory) and ARIMA (AutoRegressive Integrated Moving Average).

[0528] "Means for forecasting electricity prices based on the results of supply and demand forecasts" refers to means for using algorithms or models to forecast future electricity prices using the results of supply and demand forecasts as input.

[0529] The "means for generating multiple electricity procurement plans" refers to a means for creating multiple optimal electricity procurement plans based on predicted supply and demand and price data, taking into consideration factors such as the proportion of renewable energy, costs, and stability of supply.

[0530] The "means for presenting the generated plan to the user in an interactive format" refers to a means for visually or audibly presenting the power procurement plan to the user through an interface, thereby enabling a dialogue with the user.

[0531] "Means for receiving a user's selection and desired conditions and formulating a supply and demand plan and a control plan based on the selected plan" refers to means for formulating detailed power consumption distribution and control schedules in accordance with the power procurement plan selected by the user.

[0532] "Means for adjusting supply and demand in real time to ensure compliance with planned values" refers to means for monitoring the power supply and demand balance in real time and making adjustments as necessary.

[0533] "Means for automatically executing various operations necessary for power procurement" refers to means for automating and executing operations such as power purchasing, load balancing, and power redistribution.

[0534] "Means for recognizing the user's emotions and adjusting the way the plan is presented based on those emotions" refers to means for analyzing the user's facial expressions and voice to identify their emotions and appropriately changing the way the plan is presented depending on their emotional state.

[0535] The present invention provides a system in which a server, a terminal, and a user work together to optimize power consumption and utilize renewable energy.

[0536] First, the server collects historical electricity consumption data, weather data, and electricity price data. This data collection is performed using various APIs and database connections. For example, electricity consumption data is obtained from smart meters, weather data is obtained from the Japan Meteorological Agency's API, and price data is obtained from the electricity market database. This collected data is then cleansed and missing data is filled by removing outliers and duplicates, and linear interpolation and statistical methods are used.

[0537] Next, the server uses a machine learning model to predict power consumption supply and demand. For this purpose, models such as LSTM (Long Short-Term Memory) and ARIMA (AutoRegressive Integrated Moving Average) are used. The server also predicts power prices based on the results of the supply and demand forecast. This makes it possible to predict price fluctuations according to the supply and demand balance. Based on the forecast data, multiple power procurement plans are generated, taking into account factors such as the proportion of renewable energy, price, and supply stability.

[0538] The generated electricity procurement plan is sent from the server to the device, which then presents it to the user in an interactive format. The device displays the details of each plan using graphs and text, providing an interface that is easy for the user to understand. Furthermore, the device is equipped with an emotion engine that estimates the user's emotional state from facial expressions, voice, and input actions. This allows the device to adjust the way information is presented based on the user's emotions, for example, by providing a simple explanation if the user is confused.

[0539] Users can review the multiple power procurement plans presented and select the one that best suits their needs. When the user enters the plan they have selected and their desired conditions into the terminal, the information is sent to the server, which then uses this information to formulate supply and demand plans and control plans. The supply and demand plans include supply schedules and load balancing plans, which ensure compliance with planned values ​​for simultaneous and equal amounts. The server adjusts supply and demand in real time and automatically performs various operations required for power procurement.

[0540] As a specific example, if a factory were to use this invention, the server would collect the past year's worth of power consumption data, local weather data, and power market price data, and use this information to make supply and demand forecasts and price predictions. The server would then generate three plans: a 100% renewable energy plan, a 70% renewable energy plan, and a price-focused plan, and present these to the factory's energy manager via a terminal. When the manager selects a plan, the emotion engine would analyze the manager's facial expressions and voice to provide an explanation appropriate to the situation.

[0541] An example prompt is:

[0542] "Please create a program that will forecast supply and demand and electricity prices based on factory electricity consumption data, weather data, and electricity price data, generate an optimal electricity procurement plan, and make suggestions based on the user's emotional state."

[0543] In an embodiment of the present invention, a server, a terminal, and a user work together to optimize power procurement and improve the user experience. This system promotes the use of renewable energy, thereby achieving cost reduction and improved consumption efficiency.

[0544] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0545] Step 1:

[0546] The server collects historical electricity consumption data, weather data, and electricity price data. Specifically, it obtains data from smart meters, the Japan Meteorological Agency's API, market databases, etc. It receives historical data as input and obtains the collected data set as output. This collected data is stored in a database after converting the responses from each data source.

[0547] Step 2:

[0548] The server cleanses the collected data and fills in missing data. Specifically, it removes outliers and duplicates, and fills in missing data using linear interpolation and statistical methods. It takes the collected data as input and produces a cleansed dataset as output. The cleansing process includes filtering and replacing data.

[0549] Step 3:

[0550] The server uses the cleansed data to forecast electricity consumption. This uses machine learning models such as LSTM and ARIMA. The cleansed dataset is received as input, and future supply and demand forecast data is obtained as output. Specifically, time series data is input into the model to forecast consumption for a certain period of time.

[0551] Step 4:

[0552] The server predicts the electricity price based on the results of the supply and demand forecast. To do this, it uses the ARIMA model to predict price fluctuations. It receives supply and demand forecast data as input and obtains electricity price forecast data as output. The model calculates the price according to the supply and demand balance.

[0553] Step 5:

[0554] The server generates multiple electricity procurement plans based on supply and demand forecasts and electricity price forecast data. The plans are optimized taking into account the proportion of renewable energy, costs, and supply stability. Supply and demand forecast data and electricity price forecast data are received as input, and an electricity procurement plan is obtained as output. Specifically, conditions for each plan are set and scenarios are generated based on them.

[0555] Step 6:

[0556] The server sends the generated electricity procurement plan to the terminal, which then presents it to the user in an interactive format. The terminal receives the electricity procurement plan as input and obtains the information to be presented to the user as output. The terminal displays each plan in graphs and text, providing an easy-to-understand presentation to the user.

[0557] Step 7:

[0558] The device recognizes the user's emotions and adjusts the way the plan is presented based on those emotions. It receives the user's facial expressions, voice, and input actions as input, and outputs the emotion estimation results and an appropriate presentation method based on those emotions. The emotion engine analyzes the user's emotional state in real time, and provides simplified explanations if the user is confused, for example.

[0559] Step 8:

[0560] The user reviews the multiple electricity procurement plans provided and selects the plan that best suits their needs. The system receives the electricity procurement plans as input and obtains the selected plan as output. The user also inputs the selected plan and desired conditions into the terminal.

[0561] Step 9:

[0562] The terminal sends the user's selections and desired conditions to the server, which then formulates supply and demand plans and control plans based on them. The server receives the user's selection information as input and obtains detailed supply and demand plans and control plans as output. The supply and demand plans include supply schedules and load balancing plans.

[0563] Step 10:

[0564] The server ensures compliance with planned values ​​and adjusts supply and demand in real time. It receives real-time consumption data and planned data as input, and obtains adjusted supply and demand balance data as output. The server automatically issues control signals to execute various operations.

[0565] Step 11:

[0566] The server automatically executes various operations required for power procurement. It receives supply and demand planning data as input and obtains data on the operations that have been executed as output. Specifically, this includes operations such as power purchasing, load balancing, and power redistribution.

[0567] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0568] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0569] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0570] [Second embodiment]

[0571] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0572] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0573] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0574] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0575] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0576] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0577] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0578] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0579] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0580] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0581] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0582] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0583] This invention is a system that enables businesses and households to procure optimal and green electricity without requiring specialized knowledge. This system consists of a server, terminals, and users, and collects, analyzes, and predicts past electricity consumption data, weather data, and electricity price data, and then generates, presents, and executes an appropriate electricity procurement plan.

[0584] Main components and their operation

[0585] 1. Server Operation

[0586] The server collects historical electricity consumption data, weather data, and electricity price data, cleanses this data, and fills in missing data. To do this, the server obtains the necessary data through APIs and database connections.

[0587] Using the collected and pre-processed data, the server applies machine learning models to generate supply and demand forecasts and price predictions, using time series forecasting algorithms such as LSTM and ARIMA models.

[0588] Based on the forecast results, multiple power procurement plans (e.g., a 100% renewable energy plan, a price-focused plan, etc.) are generated. These plans are optimized based on evaluation criteria such as cost, environmental impact, and stability.

[0589] 2. Device Operation

[0590] The server generates multiple electricity procurement plans and sends them to the terminal, which then presents them to the user in an interactive format. The details of each plan are displayed in graphs and text, providing the user with an interface that is easy to understand and operate.

[0591] The terminal receives the user's selection and desired conditions and transmits them to the server, where the user can enter specific conditions (such as maximum cost or minimum required percentage of renewable energy).

[0592] 3. User Operation

[0593] Users can review multiple electricity procurement plans offered through their terminal and select the plan that best suits their needs.

[0594] Based on the plan selected by the user, the user inputs the necessary adjustments and specific plan details. The information entered by the user is sent via the terminal to the server, which then uses that information to formulate supply and demand plans and control plans.

[0595] Explanation of program processing

[0596] Data collection and preprocessing

[0597] The server collects historical electricity consumption data, weather data, and electricity price data from smart meters, the Japan Meteorological Agency's API, market databases, etc.

[0598] The collected data is cleansed and missing data is filled using linear imputation and statistical methods.

[0599] Supply and demand forecasts and price forecasts

[0600] The server uses machine learning models to forecast supply and demand, including LSTM and ARIMA, to predict future electricity consumption, taking seasonal fluctuations and trends into account.

[0601] Based on the results of the supply and demand forecast, the server also predicts electricity prices, calculating predicted price fluctuations according to the supply and demand balance using ARIMA models and other methods.

[0602] Generate electricity procurement plans

[0603] The server generates multiple procurement plans based on supply and demand forecasts and price forecasts, and the plans are optimized taking into account factors such as the proportion of renewable energy, cost, and stability of supply.

[0604] Plan presentation and selection

[0605] The terminal presents the plan sent from the server to the user and provides information in an interactive format. The interface is designed to meet the user's desired conditions and needs.

[0606] The user selects the most suitable plan and enters their desired conditions. The device then sends this information to the server.

[0607] Supply and demand planning

[0608] Based on the user's selections and desired conditions, the server creates detailed supply and demand plans and control plans, including supply schedules and load balancing plans.

[0609] Automatic execution of operations

[0610] The server ensures compliance with planned values ​​and adjusts supply and demand in real time. All necessary procedures and operations are automatically performed, optimizing power procurement without user intervention.

[0611] Specific examples

[0612] For businesses

[0613] 1. Data Collection:

[0614] The server collects electricity consumption data for the past year for the company, local weather data, and electricity market price data.

[0615] 2. Supply and demand forecast:

[0616] The server uses the LSTM model to predict supply and demand for the next year, including increased consumption in the winter.

[0617] 3. Price Prediction:

[0618] The server predicts future electricity prices using an ARIMA model based on supply and demand forecasts.

[0619] 4. Plan generation and presentation:

[0620] The server generates three plans: a 100% renewable energy plan, a 70% renewable energy plan, and a price-conscious plan, and the terminal presents them to the company's energy management officer.

[0621] 5. Select a plan:

[0622] The user (energy manager) selects Plan B (70% renewable energy plan) and enters specific desired conditions.

[0623] 6. Supply and demand planning:

[0624] The server will create a detailed supply and demand plan and control plan based on Plan B.

[0625] For ordinary households

[0626] 1. Data Collection:

[0627] The server collects the household's electricity consumption data and weather forecast data for the past six months.

[0628] 2. Supply and demand forecast:

[0629] The server uses a time series prediction model to predict power supply and demand up to one month in advance.

[0630] 3. Price Prediction:

[0631] The server uses the ARIMA model to predict electricity prices.

[0632] 4. Plan generation and presentation:

[0633] The server generates plans that use only renewable energy and have low night-time rates, and the terminal presents them to the household.

[0634] 5. Select a plan:

[0635] The user selects Plan C (a plan with a cheaper nighttime rate) and transmits the selection to the server via the terminal.

[0636] 6. Supply and demand planning:

[0637] The server develops a power consumption control plan based on Plan C and submits it to the power company.

[0638] This will provide a system that enables businesses and households to procure electricity effectively without the need for specialized knowledge.

[0639] The processing flow will be explained below.

[0640] Step 1:

[0641] The server collects historical electricity consumption data, weather data, and electricity price data, including data from smart meters, the Japan Meteorological Agency's API, market databases, etc.

[0642] Step 2:

[0643] The server cleanses the collected data, removing outliers and duplicate data to make it more reliable, and then fills in missing data using statistical methods and linear interpolation.

[0644] Step 3:

[0645] The server uses machine learning models to forecast supply and demand. Specifically, it uses LSTM and ARIMA models to input past electricity consumption data and weather data to predict future electricity consumption.

[0646] Step 4:

[0647] The server predicts electricity prices based on the results of the supply and demand forecast, and calculates future electricity prices using ARIMA models and other methods, taking into account the supply and demand balance.

[0648] Step 5:

[0649] The server generates multiple electricity procurement plans based on supply-demand and price forecast data, including factors such as the proportion of renewable energy, price, and stability.

[0650] Step 6:

[0651] The server sends the generated electricity procurement plans to the terminal, which provides the user with an interactive interface that displays the details of each plan in graphs and text.

[0652] Step 7:

[0653] The user compares multiple electricity procurement plans offered through the terminal and selects the plan that best suits their needs. The user then inputs the selected plan and desired conditions into the terminal.

[0654] Step 8:

[0655] The terminal sends the user's selection and desired conditions to the server, which uses the selection and conditions to formulate a supply and demand plan in the next step.

[0656] Step 9:

[0657] The server creates detailed supply and demand plans and control plans based on the user's selections. The supply and demand plans include supply schedules and load balancing plans, and are designed to ensure compliance with planned values.

[0658] Step 10:

[0659] The server adjusts supply and demand in real time to ensure compliance with the planned simultaneous balance. If an imbalance occurs, adjustment measures are automatically implemented.

[0660] Step 11:

[0661] The server automatically performs the various operations required for power procurement, including submitting plans to the power company and making the necessary reports after procurement is confirmed.

[0662] Step 12:

[0663] The electricity procurement is officially executed based on the plan selected by the user. The server monitors the actual electricity consumption data and collects and analyzes feedback to improve the accuracy of the next forecast.

[0664] In this way, the entire system works together to achieve optimal power procurement.

[0665] Example 1

[0666] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0667] Conventional power procurement systems require specialized knowledge, making it difficult for many businesses and households to achieve optimal power procurement. Furthermore, the accuracy of supply and demand forecasts and power price forecasts is low, and environmentally friendly renewable energy use is often not sufficiently considered. Real-time supply and demand adjustments and the provision of information in a format that is easy for users to understand are also insufficient.

[0668] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0669] In this invention, the server includes means for collecting past electricity consumption data, weather data, and electricity price data, means for cleansing the collected data and filling in missing data, means for forecasting electricity consumption supply and demand using a machine learning model, means for forecasting electricity prices based on the supply and demand forecast results, means for generating multiple electricity procurement plans based on the supply and demand forecast and price forecast, means for interactively presenting the generated plans to a user, means for receiving the user's selection and desired conditions and formulating a supply and demand plan and a control plan based on the selected plan, means for adjusting supply and demand in real time to ensure compliance with planned values, means for automatically executing various operations required for electricity procurement, means for providing graphs and text information in a user-understandable format, and means for performing long-term and short-term data analysis for forecasting and planning. This enables users without specialized knowledge to efficiently procure optimal and green electricity.

[0670] "Past power consumption data" is data showing the history of power consumption, and is a record of power usage and consumption patterns over a certain period of time.

[0671] "Weather Data" means data that includes meteorological information such as temperature, humidity, wind speed, and precipitation, and indicates weather conditions for a particular region and period of time.

[0672] "Electricity price data" refers to data showing fluctuations in electricity prices and past price history in the electricity market.

[0673] "Cleansing" refers to the process of organizing collected data and correcting or removing erroneous or incomplete information.

[0674] "Missing data imputation" refers to filling in missing information in an incomplete data set using estimates or other information.

[0675] A "machine learning model" is an algorithm that can learn patterns and rules from data and make predictions and classifications.

[0676] "Demand and supply forecasting" is the process of predicting the future balance of electricity consumption and supply.

[0677] "Electricity price forecasting" refers to predicting future electricity market prices based on the results of supply and demand forecasts, etc.

[0678] An "electricity procurement plan" is a plan that shows how electricity will be procured based on predicted electricity supply and demand and prices.

[0679] "Interactive" refers to a format in which a user interacts with a system to exchange information.

[0680] "Planning based on a selected plan" means creating a detailed supply and demand plan and control plan based on the power procurement plan selected by the user.

[0681] "Adherence to planned values" means making adjustments to match the balance of electricity supply and demand in real time.

[0682] "Automatic execution of operations" means that the system automatically performs the operations and procedures required for power procurement.

[0683] "Providing graphs and text information" means visualizing and presenting information in a way that is easy for users to understand.

[0684] "Long-term and short-term data analysis" is the process of analyzing long-term and short-term data to aid in forecasting and planning.

[0685] This invention is a system that enables businesses and households to procure electricity in an optimal and environmentally friendly manner without requiring specialized knowledge. This system consists of a server, terminals, and users, and collects, analyzes, and predicts past electricity consumption data, weather data, and electricity price data, and then generates, presents, and executes an appropriate electricity procurement plan.

[0686] Server Operations

[0687] Data collection

[0688] The server collects historical electricity consumption data, weather data, and electricity price data using smart meters, the Japan Meteorological Agency's API, and market databases. For example, when obtaining electricity consumption data from smart meters, the server collects the data using a communication protocol.

[0689] Data Cleansing and Imputation

[0690] The server cleanses the collected data and fills in missing data using linear imputation and statistical methods. For example, the cleansing process removes outliers and ensures consistency.

[0691] Supply and demand forecast

[0692] The server uses software such as TensorFlow to build LSTM and ARIMA models, inputs the collected data, and predicts future electricity supply and demand, for example, taking into account peak electricity consumption in the summer.

[0693] Price Prediction

[0694] The server uses the ARIMA model to predict electricity prices based on the results of the supply and demand forecast, using past price data and predicted supply and demand data.

[0695] Generate electricity procurement plans

[0696] The server generates multiple electricity procurement plans based on the results of supply and demand forecasts and price forecasts. The plans include the proportion of renewable energy, cost, stability of supply, etc. For example, it generates a 100% renewable energy plan, a 70% renewable energy plan, a price-focused plan, etc.

[0697] Device operation

[0698] Plan presentation and selection

[0699] The terminal interactively presents the electricity procurement plans sent from the server to the user. The details of each plan are displayed to the user in graphs and text, providing an interface that is easy to understand and operate. For example, a visual graph showing that Plan A has a high proportion of renewable energy use is displayed.

[0700] Collecting user preferences

[0701] The terminal receives the user's desired conditions and sends them to the server. The user can input specific conditions (e.g., maximum cost or minimum required percentage of renewable energy).

[0702] User operations

[0703] Review and select a plan

[0704] Users can review multiple electricity procurement plans offered through their devices and select the plan that best suits their needs. For example, an energy manager might select a plan that uses 70% renewable energy.

[0705] Enter detailed information

[0706] The user inputs the necessary adjustments and specific plan details based on the selected plan. The input information is sent via the terminal to the server, which then uses that information to create a supply-demand plan and a control plan.

[0707] Example prompts for generative AI models

[0708] "Generate a 100% renewable energy electricity procurement plan based on the past year's electricity consumption data, weather data, and electricity price data."

[0709] "Predict electricity consumption for the next year using an LSTM model and present a price-focused electricity procurement plan."

[0710] This invention provides businesses and households with effective and sustainable power supply options tailored to their specific requirements and preferences, enabling them to procure power efficiently without specialized knowledge.

[0711] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0712] Step 1: Data collection

[0713] The server collects historical electricity consumption data, weather data, and electricity price data from smart meters, the Japan Meteorological Agency's API, and the electricity market database. As a specific example, it sends a "GET / electricity / consumption" request from the smart meter to obtain consumption data for the past year and obtains the latest weather data from the Japan Meteorological Agency's API. As a result, it obtains a set of electricity consumption data, weather data, and electricity price data.

[0714] Step 2: Data cleansing and impregnation

[0715] The server cleanses the collected data and imputes missing data. It uses the collected data set as input, and as part of the data cleansing, removes outliers and produces clean data. Missing data is imputed using linear imputation and statistical methods. For example, statistical methods can be used to impute missing values ​​with the mean value, resulting in a complete and consistent data set.

[0716] Step 3: Supply and demand forecast

[0717] The server uses the cleansed and complemented data to input it into an LSTM model built using TensorFlow. The data is then processed by converting the electricity consumption data and weather data into a time series format and applying it to the LSTM model. The LSTM model then predicts future electricity consumption and outputs the supply and demand pattern for the next fiscal year as a result of the prediction.

[0718] Step 4: Price prediction

[0719] The server inputs the results of the supply and demand forecast and uses the ARIMA model to predict electricity prices. It analyzes and processes the supply and demand forecast results and past price data, and predicts future electricity market prices based on this. The output is a predicted future electricity price dataset.

[0720] Step 5: Generate a power procurement plan

[0721] The server generates an electricity procurement plan based on the results of supply and demand forecasts and price forecasts. For example, it generates a 100% renewable energy plan, a 70% renewable energy plan, and a price-focused plan. Supply and demand forecast data and price forecast data are used as input, and optimization is performed when generating each plan, taking into account factors such as the renewable energy ratio, cost, and supply stability. Detailed data for each generated plan is obtained as output.

[0722] Step 6: View and select a plan

[0723] The terminal interactively presents the electricity procurement plans sent from the server to the user. As input, it uses multiple procurement plan data received from the server. Specifically, it displays the details of each plan in an easy-to-understand graph and text format, allowing the user to easily compare and consider them. As output, it generates an information display screen to be presented to the user.

[0724] Step 7: Collect user preferences

[0725] The user inputs desired conditions (for example, maximum cost or minimum percentage of renewable energy) through a terminal. As input, the user's desired condition data is entered into the terminal. The terminal sends this to the server and requests the server to design a plan based on the conditions. As output, the user's desired condition data sent to the server is obtained.

[0726] Step 8: Develop supply and demand plans

[0727] The server formulates detailed supply and demand plans and control plans based on the user's selections and desired conditions. The inputs include the user's desired condition data and detailed data of the selected power procurement plan. To formulate the supply and demand plan, it creates a supply schedule and load balancing plan and issues instructions to various power supply devices and systems. The output generates detailed supply and demand plan data and control plan data.

[0728] Step 9: Automate operations

[0729] The server automatically executes power procurement and supply-demand adjustments based on the formulated plan. Detailed supply-demand and control plan data is used as input. Specifically, it monitors the supply-demand balance in real time and adjusts the procurement plan as needed. The output is an optimized power procurement operation.

[0730] (Application example 1)

[0731] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0732] In modern society, households and businesses lack the knowledge and time to procure optimal, green energy, making efficient and environmentally friendly energy management difficult. In particular, there is a need for a system that can automatically generate plans that take into account the proportion and cost of renewable energy and propose them in a format that is easy for users to understand. Furthermore, it is necessary to adjust supply and demand in real time based on daily electricity supply and demand forecasts and price forecasts, but this is difficult to do without specialized knowledge.

[0733] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0734] In this invention, the server includes: means for collecting past electricity consumption data, weather data, and electricity price data; means for cleansing the collected data and filling in missing data; means for forecasting electricity supply and demand using a machine learning model; means for forecasting electricity prices based on the supply and demand forecast results; means for generating multiple electricity procurement plans based on the supply and demand forecast and price forecast; means for interactively presenting the generated plans to a user; means for receiving the user's selection and desired conditions and formulating a supply and demand plan and a control plan based on the selected plan; means for adjusting supply and demand in real time to ensure compliance with planned values; means for automatically executing various operations required for electricity procurement; means for operating on a smartphone; and data processing means for automatically generating multiple plans with renewable energy usage rates and cost-focused features based on the forecast data. This enables households and businesses to achieve optimal and green electricity procurement and efficient energy management without specialized knowledge.

[0735] "Historical power consumption data" is data that records the amount of power used during a specific period of time.

[0736] "Weather data" refers to data that records meteorological conditions such as temperature, humidity, precipitation, and wind speed.

[0737] "Electricity price data" is data indicating the price of electricity in the market within a certain period of time.

[0738] "Data cleansing methods" are processing methods for removing inaccurate information from collected data and appropriately filling in missing values.

[0739] A "machine learning model" is an algorithm that learns patterns in data and predicts future data.

[0740] "Means for forecasting supply and demand" refers to a method for predicting the balance between electricity demand and supply in advance.

[0741] A "means for forecasting electricity prices" is a method for forecasting future electricity prices based on the results of supply and demand forecasts.

[0742] An "electricity procurement plan" is a specific plan regarding the method and conditions for purchasing electricity.

[0743] "Means of presenting information to the user in an interactive format" refers to a method of displaying information in a way that is easy for the user to understand and providing appropriate options through dialogue.

[0744] "Means for formulating supply and demand plans and control plans" refers to methods for determining specific electricity usage plans and adjustment methods based on the predicted supply and demand balance and price.

[0745] "Means for adjusting supply and demand in real time" refers to a method for monitoring the balance of supply and demand for electricity in real time and making adjustments as necessary.

[0746] "Means for automatic execution" refers to a method for automatically performing various operations related to power procurement without user intervention.

[0747] "Means operating on a smartphone" refers to applications and systems that operate on a smartphone and are available for use by the user.

[0748] "Renewable energy usage rate" refers to the percentage of renewable energy in the total electricity used.

[0749] "Cost-sensitive features" are plan characteristics that emphasize minimizing costs.

[0750] "Data processing means" refers to the method used to process collected data and convert it into the form required for forecasting and plan generation.

[0751] This invention is a system that enables homes and businesses to procure optimal and green electricity without requiring specialized knowledge. This system is primarily composed of a server, terminals, and users, and the interaction between these enables efficient power management.

[0752] Server Operations

[0753] The server collects historical energy consumption, weather, and electricity price data, cleansing it, and filling in missing data using APIs and database connections, such as from smart meters and weather information services.

[0754] Based on the collected and preprocessed data, the server applies machine learning models to forecast electricity supply and demand and prices. This uses the SARIMAX model, a time-series forecasting algorithm. The SARIMAX model analyzes past electricity consumption and weather data to predict future electricity supply and demand.

[0755] Based on the forecast results, multiple power procurement plans are generated. These plans include plans that prioritize renewable energy usage rates and costs, and are optimized according to the user's needs. The generated plans are presented to the user via a smartphone application.

[0756] Device operation

[0757] The device, specifically a smartphone, interactively presents the user with multiple electricity procurement plans sent from the server. The user can then review the details of each plan through the smartphone app's intuitive interface. For example, graphs and text descriptions clearly show each plan's cost, environmental impact, and supply stability.

[0758] Users input their desired conditions (e.g., maximum cost or minimum renewable energy usage rate) and select the optimal plan. The device sends the user's selection and desired conditions to the server, which then uses that information to create detailed supply and demand plans and control plans.

[0759] User operations

[0760] Users can review multiple electricity procurement plans offered through a smartphone application, select the plan that best suits their needs, and enter specific conditions (e.g., maximum cost, minimum required percentage of renewable energy) based on the plan they select.

[0761] After selecting a plan, the server creates a detailed supply and demand plan based on the information entered by the user. This plan includes a supply schedule and load balancing plan. Supply and demand are also adjusted in real time to ensure compliance with the planned balance. This allows users to achieve optimal power procurement without the need for specialized knowledge or effort.

[0762] Specific examples

[0763] For example, a server collects electricity consumption data, weather data, and electricity price data from a household for the past six months. An LSTM model is used to predict electricity supply and demand for up to one month in advance, and an ARIMA model is used to predict future electricity prices based on the supply and demand forecast. The generated plan is checked on a smartphone app, and when the user selects a plan with a 70% renewable energy usage ratio, the server uses that information to create a detailed supply and demand plan.

[0764] Prompt Sentence Examples

[0765] For example, give the generative AI model the following prompt:

[0766] "I want to develop an application that uses a home's electricity consumption, weather data, and electricity price data to propose the optimal electricity procurement plan. Can you predict future electricity consumption and prices based on past data, generate plans that include features such as the proportion of renewable energy used and focus on cost, and provide prompts to present them to the user?"

[0767] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0768] Step 1:

[0769] The server collects past electricity consumption data, weather data, and electricity price data. It obtains data from various APIs and databases as input and obtains raw data as output. Specifically, it obtains electricity consumption data from smart meters, weather data from weather information services, and electricity price data from market databases.

[0770] Step 2:

[0771] The server cleanses the collected data and fills in missing data. It uses the raw data obtained in step 1 as input and obtains cleansed data as output. Specifically, it converts the data into a data frame, removes inaccurate information, and interpolates missing values.

[0772] Step 3:

[0773] The server uses a machine learning model to predict power consumption. It uses cleansed data as input and obtains future power demand and supply forecast data as output. Specifically, it analyzes past data using the SARIMAX model and predicts future consumption.

[0774] Step 4:

[0775] The server predicts electricity prices based on the results of the supply and demand forecast. Supply and demand forecast data is used as input, and future electricity price forecast data is obtained as output. Specifically, the server uses an ARIMA model based on the balance between electricity supply and demand to predict fluctuations in electricity prices.

[0776] Step 5:

[0777] The server generates multiple electricity procurement plans based on supply and demand forecasts and price forecasts. It uses supply and demand forecast data and price forecast data as input, and obtains multiple electricity procurement plans as output. Specifically, it processes data to generate plans that prioritize renewable energy usage rates and costs.

[0778] Step 6:

[0779] The server sends the generated plan to a device (smartphone), which then presents it to the user in an interactive format. The generated electricity procurement plan is used as input, and the details of the plan are displayed as output through an interface that is easy for the user to understand. Specifically, information such as the plan's cost, environmental impact, and stability is provided through graphs and text explanations.

[0780] Step 7:

[0781] Users can check the details of each plan through a smartphone app, input their desired conditions, and then select the most suitable plan. The system uses the user's desired conditions (e.g., maximum cost or minimum required renewable energy share) as input, and the user's selection is sent to the server as output. Specific actions include clicking options and inputting desired conditions.

[0782] Step 8:

[0783] The server formulates supply and demand plans and control plans based on the user's selections and desired conditions. It uses the user's selections and desired conditions as input and obtains a detailed supply and demand plan as output. Specific operations include formulating a supply schedule and load balancing plan that is optimal for the selected plan.

[0784] Step 9:

[0785] The server adjusts supply and demand in real time to ensure compliance with the planned balance and automatically executes various necessary operations. It uses the formulated supply and demand plan as input and performs adjustments to maintain an optimal supply and demand balance as output. Specific operations include monitoring and adjusting power usage in real time.

[0786] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0787] This invention is a system that enables businesses and households to procure optimal, green electricity without requiring specialized knowledge. This system consists of a server, terminals, and users. It collects, analyzes, and predicts past electricity consumption data, weather data, and electricity price data, and then generates, presents, and executes an appropriate electricity procurement plan. Furthermore, it combines an emotion engine that recognizes the user's emotions to improve the user experience.

[0788] Main components and their operation

[0789] 1. Server Operation

[0790] The server collects historical electricity consumption data, weather data, and electricity price data, cleans the data, and fills in missing data. It retrieves the necessary data through APIs and database connections, removes outliers and duplicates, and fills in missing data using linear imputation and statistical methods.

[0791] Using the collected and pre-processed data, the server applies machine learning models (e.g., LSTM and ARIMA models) to make supply and demand forecasts and price predictions.

[0792] Based on the forecast results, multiple power procurement plans are generated, which are optimized based on criteria such as renewable energy share, cost, environmental impact, and stability.

[0793] 2. Device Operation

[0794] The server generates multiple electricity procurement plans and sends them to the terminal, which then presents them to the user in an interactive format. The details of each plan are displayed in graphs and text, providing the user with an interface that is easy to understand and operate.

[0795] The device is equipped with an emotion engine that recognizes the user's emotions and estimates their emotional state from their facial expressions, voice, input actions, etc. This allows the device to adjust the way the plan is presented to them, taking into account the user's emotions.

[0796] The device receives the user's selection and desired conditions and sends that information to the server, where the user can enter specific conditions (e.g., maximum cost or minimum required percentage of renewable energy).

[0797] 3. User Operation

[0798] Users can review multiple electricity procurement plans offered through their devices and select the plan that best suits their needs. The emotion engine monitors users' emotions in real time and assists in the plan selection process.

[0799] Based on the plan selected by the user, the user inputs the necessary adjustments and specific plan details. The information entered by the user is sent via the terminal to the server, which then uses that information to formulate supply and demand plans and control plans.

[0800] Explanation of program processing

[0801] Data collection and preprocessing

[0802] The server collects historical electricity consumption data, weather data, and electricity price data from smart meters, the Japan Meteorological Agency's API, market databases, etc.

[0803] The collected data is cleansed and missing data is filled using linear imputation and statistical methods.

[0804] Supply and demand forecasts and price forecasts

[0805] The server uses machine learning models to forecast supply and demand, including LSTM and ARIMA, to predict future electricity consumption, taking seasonal fluctuations and trends into account.

[0806] Based on the results of the supply and demand forecast, the server also predicts electricity prices, calculating predicted price fluctuations according to the supply and demand balance using ARIMA models and other methods.

[0807] Generate electricity procurement plans

[0808] The server generates multiple procurement plans based on supply and demand forecasts and price forecasts, and the plans are optimized taking into account factors such as the proportion of renewable energy, price, and stability of supply.

[0809] Presenting a plan and utilizing the emotion engine

[0810] The terminal presents the plan sent from the server to the user and provides information in an interactive format. The interface is designed to meet the user's desired conditions and needs.

[0811] The device's emotion engine recognizes the user's facial and vocal expressions and adjusts the way the plan is presented based on this. For example, if the user looks confused, it will provide a simpler explanation.

[0812] Plan selection and supply and demand planning

[0813] The user selects the most suitable plan and enters their desired conditions. The device then sends this information to the server.

[0814] The server creates detailed supply and demand plans and control plans based on the user's selections. The supply and demand plans include supply schedules and load balancing plans, and are designed to ensure compliance with planned values.

[0815] Automatic execution of operations

[0816] The server ensures compliance with planned values ​​and adjusts supply and demand in real time. All necessary procedures and operations are automatically performed, optimizing power procurement without user intervention.

[0817] Specific examples

[0818] For businesses

[0819] 1. Data Collection:

[0820] The server collects electricity consumption data for the past year for the company, local weather data, and electricity market price data.

[0821] 2. Supply and demand forecast:

[0822] The server uses the LSTM model to predict supply and demand for the next year, including increased consumption in the winter.

[0823] 3. Price Prediction:

[0824] The server predicts future electricity prices using an ARIMA model based on supply and demand forecasts.

[0825] 4. Plan generation and presentation:

[0826] The server generates three plans: a 100% renewable energy plan, a 70% renewable energy plan, and a price-conscious plan, and the terminal presents them to the company's energy management officer.

[0827] 5. Plan Selection and Emotion Engine:

[0828] When the energy manager (user) selects Plan B (70% renewable energy plan), the emotion engine analyzes the manager's facial expressions and voice and adjusts the explanation of the plan as necessary.

[0829] 6. Formulating a supply plan:

[0830] The server will create a detailed supply and demand plan and control plan based on Plan B.

[0831] For ordinary households

[0832] 1. Data Collection:

[0833] The server collects the household's electricity consumption data and weather forecast data for the past six months.

[0834] 2. Supply and demand forecast:

[0835] The server uses a time series prediction model to predict power supply and demand up to one month in advance.

[0836] 3. Price Prediction:

[0837] The server uses the ARIMA model to predict electricity prices.

[0838] 4. Plan generation and presentation:

[0839] The server generates plans that use only renewable energy and have low night-time rates, and the terminal presents them to the household.

[0840] 5. Plan Selection and Emotion Engine:

[0841] When a user selects Plan C (a plan with lower nighttime rates), the emotion engine analyzes the user's facial expressions and voice and provides appropriate information.

[0842] 6. Formulating a supply plan:

[0843] The server develops a power consumption control plan based on Plan C and submits it to the power company.

[0844] In this way, the entire system works together to achieve optimal power procurement, and by combining it with an emotion engine, it improves the user experience.

[0845] The processing flow will be explained below.

[0846] Step 1:

[0847] The server collects historical electricity consumption data, weather data, and electricity price data. Specifically, it obtains consumption data from smart meters, weather data from the Japan Meteorological Agency's API, and electricity price data from a market database.

[0848] Step 2:

[0849] The server cleanses the collected data, removing outliers and duplicate data to make it more reliable, and then fills in missing data using linear interpolation and statistical methods.

[0850] Step 3:

[0851] The server uses machine learning models to forecast supply and demand. Specifically, it uses LSTM and ARIMA models to input past electricity consumption data and weather data to predict future electricity consumption.

[0852] Step 4:

[0853] The server predicts electricity prices based on the results of the supply and demand forecast. It uses the ARIMA model to calculate future electricity prices while taking into account the supply and demand balance.

[0854] Step 5:

[0855] The server generates multiple electricity procurement plans based on supply and demand forecasts and price forecasts, and optimizes the plans by taking into account factors such as the proportion of renewable energy, prices, and supply stability.

[0856] Step 6:

[0857] The server sends the generated electricity procurement plans to the terminal, which provides the user with an interactive interface that displays the details of each plan in graphs and text.

[0858] Step 7:

[0859] The device's emotion engine analyzes the user's facial expressions and voice input in real time while the plan is being presented, recognizing their emotional state. If the user shows confusion or anxiety, the interface will provide a simpler explanation or additional information.

[0860] Step 8:

[0861] The user compares multiple electricity procurement plans offered through the terminal and selects the plan that best suits their needs. The user then inputs the selected plan and desired conditions into the terminal.

[0862] Step 9:

[0863] The terminal transmits the user's selections and desired conditions to the server, which uses the selections and conditions to develop supply and demand plans and control plans.

[0864] Step 10:

[0865] The server creates detailed supply and demand plans and control plans based on user selections, including factors such as supply schedules, load balancing plans, and renewable energy utilization rates.

[0866] Step 11:

[0867] The server adjusts supply and demand in real time to ensure compliance with the planned value. If the supply and demand balance is disrupted, necessary adjustments are automatically made.

[0868] Step 12:

[0869] The server automatically executes the various operations required for power procurement, including submitting plans to the power company and making the necessary reports after procurement is confirmed.

[0870] Step 13:

[0871] The electricity procurement is officially executed based on the plan selected by the user. The server monitors the actual electricity consumption data and collects and analyzes feedback to improve the accuracy of the next forecast.

[0872] Example: In the case of a company

[0873] Data collection and preprocessing

[0874] Step 1:

[0875] The server collects electricity consumption data from the past year, local weather data, and electricity market price data.

[0876] Step 2:

[0877] The server cleanses the collected data and fills in any missing parts. A reliable data set is created through a series of data cleansing processes.

[0878] Supply and demand forecasts and price forecasts

[0879] Step 3:

[0880] The server uses the LSTM model to predict supply and demand for the next year, including increased consumption in the winter.

[0881] Step 4:

[0882] The server uses an ARIMA model to predict future electricity prices based on supply and demand forecasts.

[0883] Plan generation and presentation

[0884] Step 5:

[0885] The server generates three plans: a 100% renewable energy plan, a 70% renewable energy plan, and a price-conscious plan.

[0886] Step 6:

[0887] The server sends the generated plan to the terminal, which then presents it to the company's energy management officer.

[0888] Plan selection and emotional engine utilization

[0889] Step 7:

[0890] The device's emotion engine analyzes the facial expressions and voice of the energy manager to recognize their emotional state in real time.

[0891] Step 8:

[0892] The energy manager (user) selects Plan B (70% renewable energy plan) and enters specific desired conditions.

[0893] Formulating and implementing a supply plan

[0894] Step 9:

[0895] The terminal sends the selection of the person in charge and the desired conditions to the server.

[0896] Step 10:

[0897] The server will create a detailed supply and demand plan and control plan based on Plan B.

[0898] Step 11:

[0899] The server ensures compliance with planned values ​​and adjusts supply and demand in real time.

[0900] Step 12:

[0901] The server will automatically handle the necessary plan submissions and procedures.

[0902] Step 13:

[0903] Power procurement is officially carried out according to the plan selected by the energy manager (user).

[0904] Example: For an average household

[0905] Data collection and preprocessing

[0906] Step 1:

[0907] The server collects the household's electricity consumption data and weather forecast data for the past six months.

[0908] Step 2:

[0909] The server cleanses the collected data and fills in any missing parts.

[0910] Supply and demand forecasts and price forecasts

[0911] Step 3:

[0912] The server uses a time series prediction model to predict power supply and demand up to one month in advance.

[0913] Step 4:

[0914] The server predicts the electricity price based on the supply and demand forecast.

[0915] Plan generation and presentation

[0916] Step 5:

[0917] The server generates renewable energy only plans and plans with cheaper nighttime rates.

[0918] Step 6:

[0919] The server sends the generated plan to the terminal, which then presents it to the household.

[0920] Plan selection and emotional engine utilization

[0921] Step 7:

[0922] The device's emotion engine analyzes the user's facial expressions and voice to recognize their emotional state in real time.

[0923] Step 8:

[0924] The user selects Plan C (a plan with a cheaper nighttime rate) and transmits the selection to the server via the terminal.

[0925] Formulating and implementing a supply plan

[0926] Step 9:

[0927] The terminal transmits the user's selection and desired conditions to the server.

[0928] Step 10:

[0929] The server will create a detailed supply and demand plan and control plan based on Plan C.

[0930] Step 11:

[0931] The server ensures compliance with planned values ​​and adjusts supply and demand in real time.

[0932] Step 12:

[0933] The server automatically handles the necessary plan submissions and procedures.

[0934] Step 13:

[0935] Power procurement is officially carried out according to the plan selected by the user.

[0936] In this way, the entire system works together to achieve optimal power procurement, and by combining it with an emotion engine, it improves the user experience.

[0937] Example 2

[0938] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0939] Conventional energy management systems are inefficient in predicting power consumption and optimizing procurement plans, making it difficult to flexibly respond to user emotions and needs. This requires advanced expertise for businesses and households to procure power in an optimal and environmentally friendly manner. Furthermore, supply-demand adjustments and price forecasts are often inaccurate, leading to insufficient adherence to planned simultaneous balancing. It is necessary to resolve these issues and improve the user experience.

[0940] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0941] In this invention, the server includes means for collecting past electricity consumption data, weather data, and electricity price data, means for cleansing the collected data and filling in missing data, means for forecasting electricity consumption supply and demand using a machine learning model, means for forecasting electricity prices based on the supply and demand forecast results, means for generating multiple electricity procurement plans based on the supply and demand forecast and the price forecast, means for interactively presenting the generated plans to a user, means for recognizing the user's emotions and adjusting the method of presenting the plans depending on the user's emotional state, means for receiving the user's selection and desired conditions and formulating a supply and demand plan and a control plan based on the selected plan, means for adjusting supply and demand in real time to ensure compliance with planned values, and means for automatically executing various operations required for electricity procurement. This enables users to procure electricity in an optimal and environmentally friendly manner without specialized knowledge, and the emotion engine can be used to improve the user experience.

[0942] "Historical electricity consumption data" refers to historical information about electricity usage over a specific period of time, and is digital data obtained from smart meters and passive measuring devices.

[0943] "Weather data" refers to information about past and current weather conditions, including temperature, humidity, precipitation, wind speed, and other factors.

[0944] "Electricity Price Data" means data that includes historical and current information about electricity unit prices and trading prices in a particular market or region.

[0945] "Cleansing" is a data processing method for removing outliers and duplicate data from collected data to improve its quality.

[0946] "Missing data imputation" is the process of filling in gaps in a dataset using statistical or linear interpolation techniques to restore the data to a complete state.

[0947] A "machine learning model" is an algorithm that learns patterns and regularities from data and performs predictions and classifications. Examples include LSTM and ARIMA.

[0948] "Supply and demand forecasting" refers to predicting future electricity demand and supply, a process carried out using machine learning models.

[0949] "Electricity price forecasting" is the process of predicting future fluctuations in electricity prices based on supply and demand forecasts.

[0950] An "electricity procurement plan" is a specific plan for how to procure electricity, and includes factors such as the proportion of renewable energy used, costs, and stability of supply.

[0951] The "means for presenting to the user in an interactive format" refers to a method for interactively displaying the generated electricity procurement plan to the user using graphs and text.

[0952] "Means for recognizing emotions and adjusting the way plans are presented according to the emotional state" refers to technology that has the function of analyzing emotions from the user's facial expressions, voice, etc., and changing the way information is presented according to those emotions.

[0953] "Adherence to planned values ​​and quantities" refers to making real-time adjustments to ensure that planned power supply and consumption match.

[0954] "Means for adjusting supply and demand in real time" refers to a system for instantly adjusting the balance between supply and demand.

[0955] "Means for automatically executing various operations required for power procurement" refers to technology that allows the system to automatically perform operations and procedures for power procurement and management without user intervention.

[0956] This invention is a system that enables businesses and households to procure electricity in an optimal and environmentally friendly manner without requiring specialized knowledge. This system consists of a server, terminals, and users, and performs data collection, data analysis, forecasting, plan generation, plan presentation, emotion recognition, supply and demand planning, and operation automation.

[0957] System configuration and processing overview

[0958] Server-based data collection and cleansing

[0959] The server collects historical electricity consumption data, weather data, and electricity price data from smart meters, the Japan Meteorological Agency's API, market databases, etc.

[0960] Cleanse the collected data and fill in missing data, for example by removing outliers and using linear interpolation and statistical methods to fill in missing data.

[0961] Supply and demand forecasts and price forecasts

[0962] The server uses machine learning models (e.g., LSTM and ARIMA) to predict supply and demand based on past data, taking into account seasonal fluctuations and trends to predict future electricity consumption.

[0963] Based on the results of the supply and demand forecast, the server also predicts electricity prices, predicting price fluctuations according to the supply and demand balance.

[0964] Generate electricity procurement plans

[0965] The server generates multiple electricity procurement plans based on supply-demand and price forecasts, and these plans are optimized taking into account factors such as the proportion of renewable energy, cost, and stability of supply.

[0966] Presenting plans and utilizing emotion recognition

[0967] The terminal presents the electricity procurement plans sent from the server to the user, and details of each plan are displayed in graphs and text, providing an interface that is easy to understand and operate.

[0968] The device is equipped with an emotion engine that recognizes the user's emotions and estimates their emotional state from their facial expressions, voice, input actions, etc. This allows the device to adjust the way the plan is presented to them, taking into account the user's emotions.

[0969] User-Server Interaction

[0970] Users can review the various electricity procurement plans offered through their devices and select the plan that best suits their needs, for example, the "70% renewable energy plan," and specify the maximum monthly cost.

[0971] The terminal receives the user's selection and desired conditions and sends this information to the server, which then uses this information to formulate supply and demand plans and control plans.

[0972] The server formulates a supply and demand plan and a control plan based on the plan selected by the user, and adjusts supply and demand in real time based on this.

[0973] Example: In the case of a company

[0974] 1. Data Collection:

[0975] The server collects a company's electricity consumption data from the past year, local weather data, and electricity market price data.

[0976] 2. Supply and demand forecast:

[0977] The server uses the LSTM model to predict supply and demand for the next year, including increased consumption in the winter.

[0978] 3. Price Prediction:

[0979] The server predicts future electricity prices using an ARIMA model based on supply and demand forecasts.

[0980] 4. Plan generation and presentation:

[0981] The server generates three plans: a 100% renewable energy plan, a 70% renewable energy plan, and a price-conscious plan, and the terminal presents them to the company's energy management officer.

[0982] 5. Plan Selection and Emotion Engine:

[0983] When the energy manager (user) selects Plan B (70% renewable energy plan), the emotion engine analyzes the manager's facial expressions and voice and adjusts the explanation of the plan as necessary.

[0984] 6. Developing supply and demand planning and control plans:

[0985] The server will create a detailed supply and demand plan and control plan based on Plan B.

[0986] Example: For an average household

[0987] 1. Data Collection:

[0988] The server collects the household's electricity consumption data and weather forecast data for the past six months.

[0989] 2. Supply and demand forecast:

[0990] The server uses a time series prediction model to predict power supply and demand up to one month in advance.

[0991] 3. Price Prediction:

[0992] The server uses the ARIMA model to predict electricity prices.

[0993] 4. Plan generation and presentation:

[0994] The server generates plans that use only renewable energy and have low night-time rates, and the terminal presents them to the household.

[0995] 5. Plan Selection and Emotion Engine:

[0996] When a user selects Plan C (a plan with lower nighttime rates), the emotion engine analyzes the user's facial expressions and voice and provides appropriate information.

[0997] 6. Supply and demand planning:

[0998] The server develops a power consumption control plan based on Plan C and submits it to the power company.

[0999] Prompt Sentence Examples

[1000] Below are some example prompts to explain each element or process of the system to the generative AI model:

[1001] User: How do we forecast our company's electricity consumption for next year?

[1002] Terminal: The server collects historical electricity consumption data and weather data, and uses an LSTM model to predict supply and demand, then uses an ARIMA model to predict prices.

[1003] In this way, the entire system works together to achieve optimal power procurement, and the emotional engine is combined to improve the user experience.

[1004] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1005] Program processing steps

[1006] Step 1: Data collection

[1007] The server collects historical electricity consumption data, weather data, and electricity price data.

[1008] Input: Data sources (smart meter API, Japan Meteorological Agency API, market database)

[1009] Specific operations: Retrieve historical electricity consumption data from the smart meter API and store it in a database. Retrieve local weather data using the Japan Meteorological Agency API and store it in a database. Query and retrieve historical electricity price data from the market database.

[1010] Output: Raw dataset for cleansing and imputation

[1011] Step 2: Data cleansing and imputation

[1012] The server cleanses the collected data and completes any missing data.

[1013] Input: raw dataset

[1014] What it does: Detect and remove outliers from datasets, remove duplicates, and impute missing data using linear interpolation and statistical methods.

[1015] Output: Clean dataset

[1016] Step 3: Supply and demand forecast

[1017] The server uses a machine learning model (e.g., LSTM) to predict supply and demand.

[1018] Input: Clean dataset

[1019] What it does: Train an LSTM model and use the clean data as input to predict future electricity demand, for example, predicting electricity consumption for the next year in 30-minute increments.

[1020] Output: Power supply and demand forecast data

[1021] Step 4: Forecasting electricity prices

[1022] The server predicts the electricity price based on the results of the supply and demand forecast.

[1023] Input: Power supply and demand forecast data

[1024] Specific operation: Uses an ARIMA model to predict future electricity prices based on supply and demand forecast data. For example, predicts electricity prices for the next year in 30-minute increments.

[1025] Output: Power price forecast data

[1026] Step 5: Generate a power procurement plan

[1027] The server generates a plurality of power procurement plans based on the supply and demand forecast and the price forecast.

[1028] Input: Power supply and demand forecast data, power price forecast data

[1029] Specific actions: Taking into account the proportion of renewable energy, cost, and stability of supply, three procurement plans will be created: a 100% renewable energy plan, a 70% renewable energy plan, and a price-focused plan.

[1030] Output: Power Procurement Plan

[1031] Step 6: Present your plan

[1032] The terminal presents the power procurement plan transmitted from the server to the user.

[1033] Input: Power Procurement Plan

[1034] Specific operation: Through the GUI, details of each plan (such as trends in electricity usage and the percentage of renewable energy used) are displayed to the user in graphs and text.

[1035] Output: The plan presented to the user

[1036] Step 7: Emotion recognition and plan adjustment

[1037] The terminal recognizes the user's emotions and adjusts the way it presents plans.

[1038] Input: User's facial expression data, voice data

[1039] Specific operation: Analyzes the user's facial expressions and voice through a camera and microphone, and if the user is confused, provides a more concise explanation and adds diagrams and charts.

[1040] Output: Optimized presentation of information to the user

[1041] Step 8: Select a plan and enter your desired conditions

[1042] The user selects the desired plan and sets specific conditions.

[1043] Input: Offered plan, user's desired conditions

[1044] Specific operation: The user selects the desired plan through the device and enters the maximum monthly cost and minimum required renewable energy percentage.

[1045] Output: User's selections and preferences

[1046] Step 9: Develop supply and demand plans and control plans

[1047] The server formulates specific supply and demand plans and control plans based on the user's selections.

[1048] Input: User selections and preferences

[1049] Specific operation: Creates a supply schedule and load balancing plan based on the conditions of the selected plan and saves it in the database.

[1050] Output: Supply and demand planning and control planning

[1051] Step 10: Real-time supply and demand adjustment and operational automation

[1052] The server ensures compliance with planned values, adjusts supply and demand in real time, and automatically executes various necessary operations.

[1053] Input: Supply and demand plan and control plan

[1054] Specific operations: Monitor the supply-demand balance and take action such as shifting the operation of specific devices in the event of an oversupply. Operations are automated and executed.

[1055] Output: Optimized power supply and demand balance

[1056] (Application example 2)

[1057] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1058] Conventional power procurement systems have made it difficult to optimize power consumption in factories and homes, particularly in promoting the use of renewable energy. Furthermore, the information required for users to select an energy plan is often diverse and difficult to understand. This makes it difficult to select the optimal power procurement plan, resulting in sluggish progress in improving power consumption efficiency and reducing costs. Furthermore, these systems have not been able to utilize emotion recognition to improve the user experience.

[1059] The specification process by the specification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting past power consumption data, weather data, and power price data, means for cleansing the collected data and supplementing missing data, and means for predicting power consumption supply and demand using a machine learning model. This makes it possible to provide optimal and green power procurement plans for power consumption in factories and homes.

[1060] The server further includes means for predicting electricity prices based on the results of the supply and demand forecast, means for generating multiple electricity procurement plans based on the supply and demand forecast and the price forecast, means for interactively presenting the generated plans to the user, means for receiving the user's selection and desired conditions and formulating a supply and demand plan and a control plan based on the selected plan, means for adjusting supply and demand in real time to ensure compliance with planned values, means for automatically executing various operations required for electricity procurement, and means for recognizing the user's emotions and adjusting the method of presenting the plans based on the emotions. This makes it possible to select the optimal electricity procurement plan according to the user's emotions, thereby achieving efficient electricity consumption and cost reduction.

[1061] "Past electricity consumption data" refers to information about the history of electricity used by consumers such as businesses and households over a certain period of time.

[1062] "Weather data" is information indicating weather conditions in a specific region or period, and includes temperature, humidity, rainfall, wind speed, and the like.

[1063] "Electricity Price Data" means information about the price of electricity in a particular market or period.

[1064] "Means for collection" refers to means for acquiring electricity consumption data, weather data, and electricity price data using devices such as servers and sensors.

[1065] "Cleansing and missing data imputation methods" are methods for removing outliers and duplicate data from collected data and filling in missing data using linear imputation and statistical methods.

[1066] "Method for forecasting supply and demand of electricity consumption using machine learning models" refers to a method for forecasting future electricity consumption using machine learning algorithms such as LSTM (Long Short-Term Memory) and ARIMA (AutoRegressive Integrated Moving Average).

[1067] "Means for forecasting electricity prices based on the results of supply and demand forecasts" refers to means for using algorithms or models to forecast future electricity prices using the results of supply and demand forecasts as input.

[1068] The "means for generating multiple electricity procurement plans" refers to a means for creating multiple optimal electricity procurement plans based on predicted supply and demand and price data, taking into consideration factors such as the proportion of renewable energy, costs, and stability of supply.

[1069] The "means for presenting the generated plan to the user in an interactive format" refers to a means for visually or audibly presenting the power procurement plan to the user through an interface, thereby enabling a dialogue with the user.

[1070] "Means for receiving a user's selection and desired conditions and formulating a supply and demand plan and a control plan based on the selected plan" refers to means for formulating detailed power consumption distribution and control schedules in accordance with the power procurement plan selected by the user.

[1071] "Means for adjusting supply and demand in real time to ensure compliance with planned values" refers to means for monitoring the power supply and demand balance in real time and making adjustments as necessary.

[1072] "Means for automatically executing various operations necessary for power procurement" refers to means for automating and executing operations such as power purchasing, load balancing, and power redistribution.

[1073] "Means for recognizing the user's emotions and adjusting the way the plan is presented based on those emotions" refers to means for analyzing the user's facial expressions and voice to identify their emotions and appropriately changing the way the plan is presented depending on their emotional state.

[1074] The present invention provides a system in which a server, a terminal, and a user work together to optimize power consumption and utilize renewable energy.

[1075] First, the server collects historical electricity consumption data, weather data, and electricity price data. This data collection is performed using various APIs and database connections. For example, electricity consumption data is obtained from smart meters, weather data is obtained from the Japan Meteorological Agency's API, and price data is obtained from the electricity market database. This collected data is then cleansed and missing data is filled by removing outliers and duplicates, and linear interpolation and statistical methods are used.

[1076] Next, the server uses a machine learning model to predict power consumption supply and demand. For this purpose, models such as LSTM (Long Short-Term Memory) and ARIMA (AutoRegressive Integrated Moving Average) are used. The server also predicts power prices based on the results of the supply and demand forecast. This makes it possible to predict price fluctuations according to the supply and demand balance. Based on the forecast data, multiple power procurement plans are generated, taking into account factors such as the proportion of renewable energy, price, and supply stability.

[1077] The generated electricity procurement plan is sent from the server to the device, which then presents it to the user in an interactive format. The device displays the details of each plan using graphs and text, providing an interface that is easy for the user to understand. Furthermore, the device is equipped with an emotion engine that estimates the user's emotional state from facial expressions, voice, and input actions. This allows the device to adjust the way information is presented based on the user's emotions, for example, by providing a simple explanation if the user is confused.

[1078] Users can review the multiple power procurement plans presented and select the one that best suits their needs. When the user enters the plan they have selected and their desired conditions into the terminal, the information is sent to the server, which then uses this information to formulate supply and demand plans and control plans. The supply and demand plans include supply schedules and load balancing plans, which ensure compliance with planned values ​​for simultaneous and equal amounts. The server adjusts supply and demand in real time and automatically performs various operations required for power procurement.

[1079] As a specific example, if a factory were to use this invention, the server would collect the past year's worth of power consumption data, local weather data, and power market price data, and use this information to make supply and demand forecasts and price predictions. The server would then generate three plans: a 100% renewable energy plan, a 70% renewable energy plan, and a price-focused plan, and present these to the factory's energy manager via a terminal. When the manager selects a plan, the emotion engine would analyze the manager's facial expressions and voice to provide an explanation appropriate to the situation.

[1080] An example prompt is:

[1081] "Please create a program that will forecast supply and demand and electricity prices based on factory electricity consumption data, weather data, and electricity price data, generate an optimal electricity procurement plan, and make suggestions based on the user's emotional state."

[1082] In an embodiment of the present invention, a server, a terminal, and a user work together to optimize power procurement and improve the user experience. This system promotes the use of renewable energy, thereby achieving cost reduction and improved consumption efficiency.

[1083] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1084] Step 1:

[1085] The server collects historical electricity consumption data, weather data, and electricity price data. Specifically, it obtains data from smart meters, the Japan Meteorological Agency's API, market databases, etc. It receives historical data as input and obtains the collected data set as output. This collected data is stored in a database after converting the responses from each data source.

[1086] Step 2:

[1087] The server cleanses the collected data and fills in missing data. Specifically, it removes outliers and duplicates, and fills in missing data using linear interpolation and statistical methods. It takes the collected data as input and produces a cleansed dataset as output. The cleansing process includes filtering and replacing data.

[1088] Step 3:

[1089] The server uses the cleansed data to forecast electricity consumption. This uses machine learning models such as LSTM and ARIMA. The cleansed dataset is received as input, and future supply and demand forecast data is obtained as output. Specifically, time series data is input into the model to forecast consumption for a certain period of time.

[1090] Step 4:

[1091] The server predicts the electricity price based on the results of the supply and demand forecast. To do this, it uses the ARIMA model to predict price fluctuations. It receives supply and demand forecast data as input and obtains electricity price forecast data as output. The model calculates the price according to the supply and demand balance.

[1092] Step 5:

[1093] The server generates multiple electricity procurement plans based on supply and demand forecasts and electricity price forecast data. The plans are optimized taking into account the proportion of renewable energy, costs, and supply stability. Supply and demand forecast data and electricity price forecast data are received as input, and an electricity procurement plan is obtained as output. Specifically, conditions for each plan are set and scenarios are generated based on them.

[1094] Step 6:

[1095] The server sends the generated electricity procurement plan to the terminal, which then presents it to the user in an interactive format. The terminal receives the electricity procurement plan as input and obtains the information to be presented to the user as output. The terminal displays each plan in graphs and text, providing an easy-to-understand presentation to the user.

[1096] Step 7:

[1097] The device recognizes the user's emotions and adjusts the way the plan is presented based on those emotions. It receives the user's facial expressions, voice, and input actions as input, and outputs the emotion estimation results and an appropriate presentation method based on those emotions. The emotion engine analyzes the user's emotional state in real time, and provides simplified explanations if the user is confused, for example.

[1098] Step 8:

[1099] The user reviews the multiple electricity procurement plans provided and selects the plan that best suits their needs. The system receives the electricity procurement plans as input and obtains the selected plan as output. The user also inputs the selected plan and desired conditions into the terminal.

[1100] Step 9:

[1101] The terminal sends the user's selections and desired conditions to the server, which then formulates supply and demand plans and control plans based on them. The server receives the user's selection information as input and obtains detailed supply and demand plans and control plans as output. The supply and demand plans include supply schedules and load balancing plans.

[1102] Step 10:

[1103] The server ensures compliance with planned values ​​and adjusts supply and demand in real time. It receives real-time consumption data and planned data as input, and obtains adjusted supply and demand balance data as output. The server automatically issues control signals to execute various operations.

[1104] Step 11:

[1105] The server automatically executes various operations required for power procurement. It receives supply and demand planning data as input and obtains data on the operations that have been executed as output. Specifically, this includes operations such as power purchasing, load balancing, and power redistribution.

[1106] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1107] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1108] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[1109] [Third embodiment]

[1110] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1111] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[1112] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1113] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[1114] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1115] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1116] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1117] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1118] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1119] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1120] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1121] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[1122] This invention is a system that enables businesses and households to procure optimal and green electricity without requiring specialized knowledge. This system consists of a server, terminals, and users, and collects, analyzes, and predicts past electricity consumption data, weather data, and electricity price data, and then generates, presents, and executes an appropriate electricity procurement plan.

[1123] Main components and their operation

[1124] 1. Server Operation

[1125] The server collects historical electricity consumption data, weather data, and electricity price data, cleanses this data, and fills in missing data. To do this, the server obtains the necessary data through APIs and database connections.

[1126] Using the collected and pre-processed data, the server applies machine learning models to generate supply and demand forecasts and price predictions, using time series forecasting algorithms such as LSTM and ARIMA models.

[1127] Based on the forecast results, multiple power procurement plans (e.g., a 100% renewable energy plan, a price-focused plan, etc.) are generated. These plans are optimized based on evaluation criteria such as cost, environmental impact, and stability.

[1128] 2. Device Operation

[1129] The server generates multiple electricity procurement plans and sends them to the terminal, which then presents them to the user in an interactive format. The details of each plan are displayed in graphs and text, providing the user with an interface that is easy to understand and operate.

[1130] The terminal receives the user's selection and desired conditions and transmits them to the server, where the user can enter specific conditions (such as maximum cost or minimum required percentage of renewable energy).

[1131] 3. User Operation

[1132] Users can review multiple electricity procurement plans offered through their terminal and select the plan that best suits their needs.

[1133] Based on the plan selected by the user, the user inputs the necessary adjustments and specific plan details. The information entered by the user is sent via the terminal to the server, which then uses that information to formulate supply and demand plans and control plans.

[1134] Explanation of program processing

[1135] Data collection and preprocessing

[1136] The server collects historical electricity consumption data, weather data, and electricity price data from smart meters, the Japan Meteorological Agency's API, market databases, etc.

[1137] The collected data is cleansed and missing data is filled using linear imputation and statistical methods.

[1138] Supply and demand forecasts and price forecasts

[1139] The server uses machine learning models to forecast supply and demand, including LSTM and ARIMA, to predict future electricity consumption, taking seasonal fluctuations and trends into account.

[1140] Based on the results of the supply and demand forecast, the server also predicts electricity prices, calculating predicted price fluctuations according to the supply and demand balance using ARIMA models and other methods.

[1141] Generate electricity procurement plans

[1142] The server generates multiple procurement plans based on supply and demand forecasts and price forecasts, and the plans are optimized taking into account factors such as the proportion of renewable energy, cost, and stability of supply.

[1143] Plan presentation and selection

[1144] The terminal presents the plan sent from the server to the user and provides information in an interactive format. The interface is designed to meet the user's desired conditions and needs.

[1145] The user selects the most suitable plan and enters their desired conditions. The device then sends this information to the server.

[1146] Supply and demand planning

[1147] Based on the user's selections and desired conditions, the server creates detailed supply and demand plans and control plans, including supply schedules and load balancing plans.

[1148] Automatic execution of operations

[1149] The server ensures compliance with planned values ​​and adjusts supply and demand in real time. All necessary procedures and operations are automatically performed, optimizing power procurement without user intervention.

[1150] Specific examples

[1151] For businesses

[1152] 1. Data Collection:

[1153] The server collects electricity consumption data for the past year for the company, local weather data, and electricity market price data.

[1154] 2. Supply and demand forecast:

[1155] The server uses the LSTM model to predict supply and demand for the next year, including increased consumption in the winter.

[1156] 3. Price Prediction:

[1157] The server predicts future electricity prices using an ARIMA model based on supply and demand forecasts.

[1158] 4. Plan generation and presentation:

[1159] The server generates three plans: a 100% renewable energy plan, a 70% renewable energy plan, and a price-conscious plan, and the terminal presents them to the company's energy management officer.

[1160] 5. Select a plan:

[1161] The user (energy manager) selects Plan B (70% renewable energy plan) and enters specific desired conditions.

[1162] 6. Supply and demand planning:

[1163] The server will create a detailed supply and demand plan and control plan based on Plan B.

[1164] For ordinary households

[1165] 1. Data Collection:

[1166] The server collects the household's electricity consumption data and weather forecast data for the past six months.

[1167] 2. Supply and demand forecast:

[1168] The server uses a time series prediction model to predict power supply and demand up to one month in advance.

[1169] 3. Price Prediction:

[1170] The server uses the ARIMA model to predict electricity prices.

[1171] 4. Plan generation and presentation:

[1172] The server generates plans that use only renewable energy and have low night-time rates, and the terminal presents them to the household.

[1173] 5. Select a plan:

[1174] The user selects Plan C (a plan with a cheaper nighttime rate) and transmits the selection to the server via the terminal.

[1175] 6. Supply and demand planning:

[1176] The server develops a power consumption control plan based on Plan C and submits it to the power company.

[1177] This will provide a system that enables businesses and households to procure electricity effectively without the need for specialized knowledge.

[1178] The processing flow will be explained below.

[1179] Step 1:

[1180] The server collects historical electricity consumption data, weather data, and electricity price data, including data from smart meters, the Japan Meteorological Agency's API, market databases, etc.

[1181] Step 2:

[1182] The server cleanses the collected data, removing outliers and duplicate data to make it more reliable, and then fills in missing data using statistical methods and linear interpolation.

[1183] Step 3:

[1184] The server uses machine learning models to forecast supply and demand. Specifically, it uses LSTM and ARIMA models to input past electricity consumption data and weather data to predict future electricity consumption.

[1185] Step 4:

[1186] The server predicts electricity prices based on the results of the supply and demand forecast, and calculates future electricity prices using ARIMA models and other methods, taking into account the supply and demand balance.

[1187] Step 5:

[1188] The server generates multiple electricity procurement plans based on supply-demand and price forecast data, including factors such as the proportion of renewable energy, price, and stability.

[1189] Step 6:

[1190] The server sends the generated electricity procurement plans to the terminal, which provides the user with an interactive interface and displays the details of each plan in graphs and text.

[1191] Step 7:

[1192] The user compares multiple electricity procurement plans offered through the terminal and selects the plan that best suits their needs. The user then inputs the selected plan and desired conditions into the terminal.

[1193] Step 8:

[1194] The terminal sends the user's selection and desired conditions to the server, which uses the selection and conditions to formulate a supply and demand plan in the next step.

[1195] Step 9:

[1196] The server creates detailed supply and demand plans and control plans based on the user's selections. The supply and demand plans include supply schedules and load balancing plans, and are designed to ensure compliance with planned values.

[1197] Step 10:

[1198] The server adjusts supply and demand in real time to ensure compliance with the planned simultaneous balance. If an imbalance occurs, adjustment measures are automatically implemented.

[1199] Step 11:

[1200] The server automatically executes the various operations required for power procurement, including submitting plans to the power company and making the necessary reports after procurement is confirmed.

[1201] Step 12:

[1202] The electricity procurement is officially executed based on the plan selected by the user. The server monitors the actual electricity consumption data and collects and analyzes feedback to improve the accuracy of the next forecast.

[1203] In this way, the entire system works together to achieve optimal power procurement.

[1204] Example 1

[1205] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1206] Conventional power procurement systems require specialized knowledge, making it difficult for many businesses and households to achieve optimal power procurement. Furthermore, the accuracy of supply and demand forecasts and power price forecasts is low, and environmentally friendly renewable energy use is often not sufficiently considered. Real-time supply and demand adjustments and the provision of information in a format that is easy for users to understand are also insufficient.

[1207] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1208] In this invention, the server includes means for collecting past electricity consumption data, weather data, and electricity price data, means for cleansing the collected data and filling in missing data, means for forecasting electricity consumption supply and demand using a machine learning model, means for forecasting electricity prices based on the supply and demand forecast results, means for generating multiple electricity procurement plans based on the supply and demand forecast and price forecast, means for interactively presenting the generated plans to a user, means for receiving the user's selection and desired conditions and formulating a supply and demand plan and a control plan based on the selected plan, means for adjusting supply and demand in real time to ensure compliance with planned values, means for automatically executing various operations required for electricity procurement, means for providing graphs and text information in a user-understandable format, and means for performing long-term and short-term data analysis for forecasting and planning. This enables users without specialized knowledge to efficiently procure optimal and green electricity.

[1209] "Past power consumption data" is data showing the history of power consumption, and is a record of power usage and consumption patterns over a certain period of time.

[1210] "Weather Data" means data that includes meteorological information such as temperature, humidity, wind speed, and precipitation, and indicates weather conditions for a particular region and period of time.

[1211] "Electricity price data" refers to data showing fluctuations in electricity prices and past price history in the electricity market.

[1212] "Cleansing" refers to the process of organizing collected data and correcting or removing erroneous or incomplete information.

[1213] "Missing data imputation" refers to filling in missing information in an incomplete data set using estimates or other information.

[1214] A "machine learning model" is an algorithm that can learn patterns and rules from data and make predictions and classifications.

[1215] "Demand and supply forecasting" is the process of predicting the future balance of electricity consumption and supply.

[1216] "Electricity price forecasting" refers to predicting future electricity market prices based on the results of supply and demand forecasts, etc.

[1217] An "electricity procurement plan" is a plan that shows how electricity will be procured based on predicted electricity supply and demand and prices.

[1218] "Interactive" refers to a format in which a user interacts with a system to exchange information.

[1219] "Planning based on a selected plan" means creating a detailed supply and demand plan and control plan based on the power procurement plan selected by the user.

[1220] "Adherence to planned values" means making adjustments to match the balance of electricity supply and demand in real time.

[1221] "Automatic execution of operations" means that the system automatically performs the operations and procedures required for power procurement.

[1222] "Providing graphs and text information" means visualizing and presenting information in a way that is easy for users to understand.

[1223] "Long-term and short-term data analysis" is the process of analyzing long-term and short-term data to aid in forecasting and planning.

[1224] This invention is a system that enables businesses and households to procure electricity in an optimal and environmentally friendly manner without requiring specialized knowledge. This system consists of a server, terminals, and users, and collects, analyzes, and predicts past electricity consumption data, weather data, and electricity price data, and then generates, presents, and executes an appropriate electricity procurement plan.

[1225] Server Operations

[1226] Data collection

[1227] The server collects historical electricity consumption data, weather data, and electricity price data using smart meters, the Japan Meteorological Agency's API, and market databases. For example, when obtaining electricity consumption data from smart meters, the server collects the data using a communication protocol.

[1228] Data Cleansing and Imputation

[1229] The server cleanses the collected data and fills in missing data using linear imputation and statistical methods. For example, the cleansing process removes outliers and ensures consistency.

[1230] Supply and demand forecast

[1231] The server uses software such as TensorFlow to build LSTM and ARIMA models, inputs the collected data, and predicts future electricity supply and demand, for example, taking into account peak electricity consumption in the summer.

[1232] Price Prediction

[1233] The server uses the ARIMA model to predict electricity prices based on the results of the supply and demand forecast, using past price data and predicted supply and demand data.

[1234] Generate electricity procurement plans

[1235] The server generates multiple electricity procurement plans based on the results of supply and demand forecasts and price forecasts. The plans include the proportion of renewable energy, cost, stability of supply, etc. For example, it generates a 100% renewable energy plan, a 70% renewable energy plan, a price-focused plan, etc.

[1236] Device operation

[1237] Plan presentation and selection

[1238] The terminal interactively presents the electricity procurement plans sent from the server to the user. The details of each plan are displayed to the user in graphs and text, providing an interface that is easy to understand and operate. For example, a visual graph showing that Plan A has a high proportion of renewable energy use is displayed.

[1239] Collecting user preferences

[1240] The terminal receives the user's desired conditions and sends them to the server. The user can input specific conditions (e.g., maximum cost or minimum required percentage of renewable energy).

[1241] User operations

[1242] Review and select a plan

[1243] Users can review multiple electricity procurement plans offered through their devices and select the plan that best suits their needs. For example, an energy manager might select a plan that uses 70% renewable energy.

[1244] Enter detailed information

[1245] The user inputs the necessary adjustments and specific plan details based on the selected plan. The input information is sent via the terminal to the server, which then uses that information to create a supply-demand plan and a control plan.

[1246] Example prompts for generative AI models

[1247] "Generate a 100% renewable energy electricity procurement plan based on the past year's electricity consumption data, weather data, and electricity price data."

[1248] "Predict electricity consumption for the next year using an LSTM model and present a price-focused electricity procurement plan."

[1249] This invention provides businesses and households with effective and sustainable power supply options tailored to their specific requirements and preferences, enabling them to procure power efficiently without specialized knowledge.

[1250] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1251] Step 1: Data collection

[1252] The server collects historical electricity consumption data, weather data, and electricity price data from smart meters, the Japan Meteorological Agency's API, and the electricity market database. As a specific example, it sends a "GET / electricity / consumption" request from the smart meter to obtain consumption data for the past year and obtains the latest weather data from the Japan Meteorological Agency's API. As a result, it obtains a set of electricity consumption data, weather data, and electricity price data.

[1253] Step 2: Data cleansing and impregnation

[1254] The server cleanses the collected data and imputes missing data. It uses the collected data set as input, and as part of the data cleansing, removes outliers and produces clean data. Missing data is imputed using linear imputation and statistical methods. For example, statistical methods can be used to impute missing values ​​with the mean value, resulting in a complete and consistent data set.

[1255] Step 3: Supply and demand forecast

[1256] The server uses the cleansed and complemented data to input it into an LSTM model built using TensorFlow. The data is then processed by converting the electricity consumption data and weather data into a time series format and applying it to the LSTM model. The LSTM model then predicts future electricity consumption and outputs the supply and demand pattern for the next fiscal year as a result of the prediction.

[1257] Step 4: Price prediction

[1258] The server inputs the results of the supply and demand forecast and uses the ARIMA model to predict electricity prices. It analyzes and processes the supply and demand forecast results and past price data, and predicts future electricity market prices based on this. The output is a predicted future electricity price dataset.

[1259] Step 5: Generate a power procurement plan

[1260] The server generates an electricity procurement plan based on the results of supply and demand forecasts and price forecasts. For example, it generates a 100% renewable energy plan, a 70% renewable energy plan, and a price-focused plan. Supply and demand forecast data and price forecast data are used as input, and optimization is performed when generating each plan, taking into account factors such as the renewable energy ratio, cost, and supply stability. Detailed data for each generated plan is obtained as output.

[1261] Step 6: View and select a plan

[1262] The terminal interactively presents the electricity procurement plans sent from the server to the user. As input, it uses multiple procurement plan data received from the server. Specifically, it displays the details of each plan in an easy-to-understand graph and text format, allowing the user to easily compare and consider them. As output, it generates an information display screen to be presented to the user.

[1263] Step 7: Collect user preferences

[1264] The user inputs desired conditions (for example, maximum cost or minimum percentage of renewable energy) through a terminal. As input, the user's desired condition data is entered into the terminal. The terminal sends this to the server and requests the server to design a plan based on the conditions. As output, the user's desired condition data sent to the server is obtained.

[1265] Step 8: Develop supply and demand plans

[1266] The server formulates detailed supply and demand plans and control plans based on the user's selections and desired conditions. The inputs include the user's desired condition data and detailed data of the selected power procurement plan. To formulate the supply and demand plan, it creates a supply schedule and load balancing plan and issues instructions to various power supply devices and systems. The output generates detailed supply and demand plan data and control plan data.

[1267] Step 9: Automate operations

[1268] The server automatically executes power procurement and supply-demand adjustments based on the formulated plan. Detailed supply-demand and control plan data is used as input. Specifically, it monitors the supply-demand balance in real time and adjusts the procurement plan as needed. The output is an optimized power procurement operation.

[1269] (Application example 1)

[1270] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1271] In modern society, households and businesses lack the knowledge and time to procure optimal, green energy, making efficient and environmentally friendly energy management difficult. In particular, there is a need for a system that can automatically generate plans that take into account the proportion and cost of renewable energy and propose them in a format that is easy for users to understand. Furthermore, it is necessary to adjust supply and demand in real time based on daily electricity supply and demand forecasts and price forecasts, but this is difficult to do without specialized knowledge.

[1272] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1273] In this invention, the server includes: means for collecting past electricity consumption data, weather data, and electricity price data; means for cleansing the collected data and filling in missing data; means for forecasting electricity supply and demand using a machine learning model; means for forecasting electricity prices based on the supply and demand forecast results; means for generating multiple electricity procurement plans based on the supply and demand forecast and price forecast; means for interactively presenting the generated plans to a user; means for receiving the user's selection and desired conditions and formulating a supply and demand plan and a control plan based on the selected plan; means for adjusting supply and demand in real time to ensure compliance with planned values; means for automatically executing various operations required for electricity procurement; means for operating on a smartphone; and data processing means for automatically generating multiple plans with renewable energy usage rates and cost-focused features based on the forecast data. This enables households and businesses to achieve optimal and green electricity procurement and efficient energy management without specialized knowledge.

[1274] "Historical power consumption data" is data that records the amount of power used during a specific period of time.

[1275] "Weather data" refers to data that records meteorological conditions such as temperature, humidity, precipitation, and wind speed.

[1276] "Electricity price data" is data indicating the price of electricity in the market within a certain period of time.

[1277] "Data cleansing methods" are processing methods for removing inaccurate information from collected data and appropriately filling in missing values.

[1278] A "machine learning model" is an algorithm that learns patterns in data and predicts future data.

[1279] "Means for forecasting supply and demand" refers to a method for predicting the balance between electricity demand and supply in advance.

[1280] A "means for forecasting electricity prices" is a method for forecasting future electricity prices based on the results of supply and demand forecasts.

[1281] An "electricity procurement plan" is a specific plan regarding the method and conditions for purchasing electricity.

[1282] "Means of presenting information to the user in an interactive format" refers to a method of displaying information in a way that is easy for the user to understand and providing appropriate options through dialogue.

[1283] "Means for formulating supply and demand plans and control plans" refers to methods for determining specific electricity usage plans and adjustment methods based on the predicted supply and demand balance and price.

[1284] "Means for adjusting supply and demand in real time" refers to a method for monitoring the balance of supply and demand for electricity in real time and making adjustments as necessary.

[1285] "Means for automatic execution" refers to a method for automatically performing various operations related to power procurement without user intervention.

[1286] "Means operating on a smartphone" refers to applications and systems that operate on a smartphone and are available for use by the user.

[1287] "Renewable energy usage rate" refers to the percentage of renewable energy in the total electricity used.

[1288] "Cost-sensitive features" are plan characteristics that emphasize minimizing costs.

[1289] "Data processing means" refers to the method used to process collected data and convert it into the form required for forecasting and plan generation.

[1290] This invention is a system that enables homes and businesses to procure optimal and green electricity without requiring specialized knowledge. This system is primarily composed of a server, terminals, and users, and the interaction between these enables efficient power management.

[1291] Server Operations

[1292] The server collects historical energy consumption, weather, and electricity price data, cleansing it, and filling in missing data using APIs and database connections, such as from smart meters and weather information services.

[1293] Based on the collected and preprocessed data, the server applies machine learning models to forecast electricity supply and demand and prices. This uses the SARIMAX model, a time-series forecasting algorithm. The SARIMAX model analyzes past electricity consumption and weather data to predict future electricity supply and demand.

[1294] Based on the forecast results, multiple power procurement plans are generated. These plans include plans that prioritize renewable energy usage rates and costs, and are optimized according to the user's needs. The generated plans are presented to the user via a smartphone application.

[1295] Device operation

[1296] The device, specifically a smartphone, interactively presents the user with multiple electricity procurement plans sent from the server. The user can then review the details of each plan through the smartphone app's intuitive interface. For example, graphs and text descriptions clearly show each plan's cost, environmental impact, and supply stability.

[1297] Users input their desired conditions (e.g., maximum cost or minimum renewable energy usage rate) and select the optimal plan. The device sends the user's selection and desired conditions to the server, which then uses that information to create detailed supply and demand plans and control plans.

[1298] User operations

[1299] Users can review multiple electricity procurement plans offered through a smartphone application, select the plan that best suits their needs, and enter specific conditions (e.g., maximum cost, minimum required percentage of renewable energy) based on the plan they select.

[1300] After selecting a plan, the server creates a detailed supply and demand plan based on the information entered by the user. This plan includes a supply schedule and load balancing plan. Supply and demand are also adjusted in real time to ensure compliance with the planned balance. This allows users to achieve optimal power procurement without the need for specialized knowledge or effort.

[1301] Specific examples

[1302] For example, a server collects electricity consumption data, weather data, and electricity price data from a household for the past six months. An LSTM model is used to predict electricity supply and demand for up to one month in advance, and an ARIMA model is used to predict future electricity prices based on the supply and demand forecast. The generated plan is checked on a smartphone app, and when the user selects a plan with a 70% renewable energy usage ratio, the server uses that information to create a detailed supply and demand plan.

[1303] Prompt Sentence Examples

[1304] For example, give the generative AI model the following prompt:

[1305] "I want to develop an application that uses a home's electricity consumption, weather data, and electricity price data to propose the optimal electricity procurement plan. Can you predict future electricity consumption and prices based on past data, generate plans that include features such as the proportion of renewable energy used and focus on cost, and provide prompts to present them to the user?"

[1306] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1307] Step 1:

[1308] The server collects past electricity consumption data, weather data, and electricity price data. It obtains data from various APIs and databases as input and obtains raw data as output. Specifically, it obtains electricity consumption data from smart meters, weather data from weather information services, and electricity price data from market databases.

[1309] Step 2:

[1310] The server cleanses the collected data and fills in missing data. It uses the raw data obtained in step 1 as input and obtains cleansed data as output. Specifically, it converts the data into a data frame, removes inaccurate information, and interpolates missing values.

[1311] Step 3:

[1312] The server uses a machine learning model to predict power consumption. It uses cleansed data as input and obtains future power demand and supply forecast data as output. Specifically, it analyzes past data using the SARIMAX model and predicts future consumption.

[1313] Step 4:

[1314] The server predicts electricity prices based on the results of the supply and demand forecast. Supply and demand forecast data is used as input, and future electricity price forecast data is obtained as output. Specifically, the server uses an ARIMA model based on the balance between electricity supply and demand to predict fluctuations in electricity prices.

[1315] Step 5:

[1316] The server generates multiple electricity procurement plans based on supply and demand forecasts and price forecasts. It uses supply and demand forecast data and price forecast data as input, and obtains multiple electricity procurement plans as output. Specifically, it processes data to generate plans that prioritize renewable energy usage rates and costs.

[1317] Step 6:

[1318] The server sends the generated plan to a device (smartphone), which then presents it to the user in an interactive format. The generated electricity procurement plan is used as input, and the details of the plan are displayed as output through an interface that is easy for the user to understand. Specifically, information such as the plan's cost, environmental impact, and stability is provided through graphs and text explanations.

[1319] Step 7:

[1320] Users can check the details of each plan through a smartphone app, input their desired conditions, and then select the most suitable plan. The system uses the user's desired conditions (e.g., maximum cost or minimum required renewable energy share) as input, and the user's selection is sent to the server as output. Specific actions include clicking options and inputting desired conditions.

[1321] Step 8:

[1322] The server formulates supply and demand plans and control plans based on the user's selections and desired conditions. It uses the user's selections and desired conditions as input and obtains a detailed supply and demand plan as output. Specific operations include formulating a supply schedule and load balancing plan that is optimal for the selected plan.

[1323] Step 9:

[1324] The server adjusts supply and demand in real time to ensure compliance with the planned balance and automatically executes various necessary operations. It uses the formulated supply and demand plan as input and performs adjustments to maintain an optimal supply and demand balance as output. Specific operations include monitoring and adjusting power usage in real time.

[1325] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1326] This invention is a system that enables businesses and households to procure optimal, green electricity without requiring specialized knowledge. This system consists of a server, terminals, and users. It collects, analyzes, and predicts past electricity consumption data, weather data, and electricity price data, and then generates, presents, and executes an appropriate electricity procurement plan. Furthermore, it combines an emotion engine that recognizes the user's emotions to improve the user experience.

[1327] Main components and their operation

[1328] 1. Server Operation

[1329] The server collects historical electricity consumption data, weather data, and electricity price data, cleans the data, and fills in missing data. It retrieves the necessary data through APIs and database connections, removes outliers and duplicates, and fills in missing data using linear imputation and statistical methods.

[1330] Using the collected and pre-processed data, the server applies machine learning models (e.g., LSTM and ARIMA models) to make supply and demand forecasts and price predictions.

[1331] Based on the forecast results, multiple power procurement plans are generated, which are optimized based on criteria such as renewable energy share, cost, environmental impact, and stability.

[1332] 2. Device Operation

[1333] The server generates multiple electricity procurement plans and sends them to the terminal, which then presents them to the user in an interactive format. The details of each plan are displayed in graphs and text, providing the user with an interface that is easy to understand and operate.

[1334] The device is equipped with an emotion engine that recognizes the user's emotions and estimates their emotional state from their facial expressions, voice, input actions, etc. This allows the device to adjust the way the plan is presented to them, taking into account the user's emotions.

[1335] The device receives the user's selection and desired conditions and sends that information to the server, where the user can enter specific conditions (e.g., maximum cost or minimum required percentage of renewable energy).

[1336] 3. User Operation

[1337] Users can review multiple electricity procurement plans offered through their devices and select the plan that best suits their needs. The emotion engine monitors users' emotions in real time and assists in the plan selection process.

[1338] Based on the plan selected by the user, the user inputs the necessary adjustments and specific plan details. The information entered by the user is sent via the terminal to the server, which then uses that information to formulate supply and demand plans and control plans.

[1339] Explanation of program processing

[1340] Data collection and preprocessing

[1341] The server collects historical electricity consumption data, weather data, and electricity price data from smart meters, the Japan Meteorological Agency's API, market databases, etc.

[1342] The collected data is cleansed and missing data is filled using linear imputation and statistical methods.

[1343] Supply and demand forecasts and price forecasts

[1344] The server uses machine learning models to forecast supply and demand, including LSTM and ARIMA, to predict future electricity consumption, taking seasonal fluctuations and trends into account.

[1345] Based on the results of the supply and demand forecast, the server also predicts electricity prices, calculating predicted price fluctuations according to the supply and demand balance using ARIMA models and other methods.

[1346] Generate electricity procurement plans

[1347] The server generates multiple procurement plans based on supply and demand forecasts and price forecasts, and the plans are optimized taking into account factors such as the proportion of renewable energy, price, and stability of supply.

[1348] Presenting a plan and utilizing the emotion engine

[1349] The terminal presents the plan sent from the server to the user and provides information in an interactive format. The interface is designed to meet the user's desired conditions and needs.

[1350] The device's emotion engine recognizes the user's facial and vocal expressions and adjusts the way the plan is presented based on this. For example, if the user looks confused, it will provide a simpler explanation.

[1351] Plan selection and supply and demand planning

[1352] The user selects the most suitable plan and enters their desired conditions. The device then sends this information to the server.

[1353] The server creates detailed supply and demand plans and control plans based on the user's selections. The supply and demand plans include supply schedules and load balancing plans, and are designed to ensure compliance with planned values.

[1354] Automatic execution of operations

[1355] The server ensures compliance with planned values ​​and adjusts supply and demand in real time. All necessary procedures and operations are automatically performed, optimizing power procurement without user intervention.

[1356] Specific examples

[1357] For businesses

[1358] 1. Data Collection:

[1359] The server collects electricity consumption data for the past year for the company, local weather data, and electricity market price data.

[1360] 2. Supply and demand forecast:

[1361] The server uses the LSTM model to predict supply and demand for the next year, including increased consumption in the winter.

[1362] 3. Price Prediction:

[1363] The server predicts future electricity prices using an ARIMA model based on supply and demand forecasts.

[1364] 4. Plan generation and presentation:

[1365] The server generates three plans: a 100% renewable energy plan, a 70% renewable energy plan, and a price-conscious plan, and the terminal presents them to the company's energy management officer.

[1366] 5. Plan Selection and Emotion Engine:

[1367] When the energy manager (user) selects Plan B (70% renewable energy plan), the emotion engine analyzes the manager's facial expressions and voice and adjusts the explanation of the plan as necessary.

[1368] 6. Formulating a supply plan:

[1369] The server will create a detailed supply and demand plan and control plan based on Plan B.

[1370] For ordinary households

[1371] 1. Data Collection:

[1372] The server collects the household's electricity consumption data and weather forecast data for the past six months.

[1373] 2. Supply and demand forecast:

[1374] The server uses a time series prediction model to predict power supply and demand up to one month in advance.

[1375] 3. Price Prediction:

[1376] The server uses the ARIMA model to predict electricity prices.

[1377] 4. Plan generation and presentation:

[1378] The server generates plans that use only renewable energy and have low night-time rates, and the terminal presents them to the household.

[1379] 5. Plan Selection and Emotion Engine:

[1380] When a user selects Plan C (a plan with lower nighttime rates), the emotion engine analyzes the user's facial expressions and voice and provides appropriate information.

[1381] 6. Formulating a supply plan:

[1382] The server develops a power consumption control plan based on Plan C and submits it to the power company.

[1383] In this way, the entire system works together to achieve optimal power procurement, and by combining it with an emotion engine, it improves the user experience.

[1384] The processing flow will be explained below.

[1385] Step 1:

[1386] The server collects historical electricity consumption data, weather data, and electricity price data. Specifically, it obtains consumption data from smart meters, weather data from the Japan Meteorological Agency's API, and electricity price data from a market database.

[1387] Step 2:

[1388] The server cleanses the collected data, removing outliers and duplicate data to make it more reliable, and then fills in missing data using linear interpolation and statistical methods.

[1389] Step 3:

[1390] The server uses machine learning models to forecast supply and demand. Specifically, it uses LSTM and ARIMA models to input past electricity consumption data and weather data to predict future electricity consumption.

[1391] Step 4:

[1392] The server predicts electricity prices based on the results of the supply and demand forecast. It uses the ARIMA model to calculate future electricity prices while taking into account the supply and demand balance.

[1393] Step 5:

[1394] The server generates multiple electricity procurement plans based on supply and demand forecasts and price forecasts, and optimizes the plans by taking into account factors such as the proportion of renewable energy, prices, and supply stability.

[1395] Step 6:

[1396] The server sends the generated electricity procurement plans to the terminal, which provides the user with an interactive interface that displays the details of each plan in graphs and text.

[1397] Step 7:

[1398] The device's emotion engine analyzes the user's facial expressions and voice input in real time while the plan is being presented, recognizing their emotional state. If the user shows confusion or anxiety, the interface will provide a simpler explanation or additional information.

[1399] Step 8:

[1400] The user compares multiple electricity procurement plans offered through the terminal and selects the plan that best suits their needs. The user then inputs the selected plan and desired conditions into the terminal.

[1401] Step 9:

[1402] The terminal transmits the user's selections and desired conditions to the server, which uses the selections and conditions to develop supply and demand plans and control plans.

[1403] Step 10:

[1404] The server creates detailed supply and demand plans and control plans based on user selections, including factors such as supply schedules, load balancing plans, and renewable energy utilization rates.

[1405] Step 11:

[1406] The server adjusts supply and demand in real time to ensure compliance with the planned value. If the supply and demand balance is disrupted, necessary adjustments are automatically made.

[1407] Step 12:

[1408] The server automatically executes the various operations required for power procurement, including submitting plans to the power company and making the necessary reports after procurement is confirmed.

[1409] Step 13:

[1410] The electricity procurement is officially executed based on the plan selected by the user. The server monitors the actual electricity consumption data and collects and analyzes feedback to improve the accuracy of the next forecast.

[1411] Example: In the case of a company

[1412] Data collection and preprocessing

[1413] Step 1:

[1414] The server collects electricity consumption data from the past year, local weather data, and electricity market price data.

[1415] Step 2:

[1416] The server cleanses the collected data and fills in any missing parts. A reliable data set is created through a series of data cleansing processes.

[1417] Supply and demand forecasts and price forecasts

[1418] Step 3:

[1419] The server uses the LSTM model to predict supply and demand for the next year, including increased consumption in the winter.

[1420] Step 4:

[1421] The server uses an ARIMA model to predict future electricity prices based on supply and demand forecasts.

[1422] Plan generation and presentation

[1423] Step 5:

[1424] The server generates three plans: a 100% renewable energy plan, a 70% renewable energy plan, and a price-conscious plan.

[1425] Step 6:

[1426] The server sends the generated plan to the terminal, which then presents it to the company's energy management officer.

[1427] Plan selection and emotional engine utilization

[1428] Step 7:

[1429] The device's emotion engine analyzes the facial expressions and voice of the energy manager to recognize their emotional state in real time.

[1430] Step 8:

[1431] The energy manager (user) selects Plan B (70% renewable energy plan) and enters specific desired conditions.

[1432] Formulating and implementing a supply plan

[1433] Step 9:

[1434] The terminal sends the selection of the person in charge and the desired conditions to the server.

[1435] Step 10:

[1436] The server will create a detailed supply and demand plan and control plan based on Plan B.

[1437] Step 11:

[1438] The server ensures compliance with planned values ​​and adjusts supply and demand in real time.

[1439] Step 12:

[1440] The server will automatically handle the necessary plan submissions and procedures.

[1441] Step 13:

[1442] Power procurement is officially carried out according to the plan selected by the energy manager (user).

[1443] Example: For an average household

[1444] Data collection and preprocessing

[1445] Step 1:

[1446] The server collects the household's electricity consumption data and weather forecast data for the past six months.

[1447] Step 2:

[1448] The server cleanses the collected data and fills in any missing parts.

[1449] Supply and demand forecasts and price forecasts

[1450] Step 3:

[1451] The server uses a time series prediction model to predict power supply and demand up to one month in advance.

[1452] Step 4:

[1453] The server predicts the electricity price based on the supply and demand forecast.

[1454] Plan generation and presentation

[1455] Step 5:

[1456] The server generates renewable energy only plans and plans with cheaper nighttime rates.

[1457] Step 6:

[1458] The server sends the generated plan to the terminal, which then presents it to the household.

[1459] Plan selection and emotional engine utilization

[1460] Step 7:

[1461] The device's emotion engine analyzes the user's facial expressions and voice to recognize their emotional state in real time.

[1462] Step 8:

[1463] The user selects Plan C (a plan with a cheaper nighttime rate) and transmits the selection to the server via the terminal.

[1464] Formulating and implementing a supply plan

[1465] Step 9:

[1466] The terminal transmits the user's selection and desired conditions to the server.

[1467] Step 10:

[1468] The server will create a detailed supply and demand plan and control plan based on Plan C.

[1469] Step 11:

[1470] The server ensures compliance with planned values ​​and adjusts supply and demand in real time.

[1471] Step 12:

[1472] The server automatically handles the necessary plan submissions and procedures.

[1473] Step 13:

[1474] Power procurement is officially carried out according to the plan selected by the user.

[1475] In this way, the entire system works together to achieve optimal power procurement, and by combining it with an emotion engine, it improves the user experience.

[1476] Example 2

[1477] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1478] Conventional energy management systems are inefficient in predicting power consumption and optimizing procurement plans, making it difficult to flexibly respond to user emotions and needs. This requires advanced expertise for businesses and households to procure power in an optimal and environmentally friendly manner. Furthermore, supply-demand adjustments and price forecasts are often inaccurate, leading to insufficient adherence to planned simultaneous balancing. It is necessary to resolve these issues and improve the user experience.

[1479] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1480] In this invention, the server includes means for collecting past electricity consumption data, weather data, and electricity price data, means for cleansing the collected data and filling in missing data, means for forecasting electricity consumption supply and demand using a machine learning model, means for forecasting electricity prices based on the supply and demand forecast results, means for generating multiple electricity procurement plans based on the supply and demand forecast and the price forecast, means for interactively presenting the generated plans to a user, means for recognizing the user's emotions and adjusting the method of presenting the plans depending on the user's emotional state, means for receiving the user's selection and desired conditions and formulating a supply and demand plan and a control plan based on the selected plan, means for adjusting supply and demand in real time to ensure compliance with planned values, and means for automatically executing various operations required for electricity procurement. This enables users to procure electricity in an optimal and environmentally friendly manner without specialized knowledge, and the emotion engine can be used to improve the user experience.

[1481] "Historical electricity consumption data" refers to historical information about electricity usage over a specific period of time, and is digital data obtained from smart meters and passive measuring devices.

[1482] "Weather data" refers to information about past and current weather conditions, including temperature, humidity, precipitation, wind speed, and other factors.

[1483] "Electricity Price Data" means data that includes historical and current information about electricity unit prices and trading prices in a particular market or region.

[1484] "Cleansing" is a data processing method for removing outliers and duplicate data from collected data to improve its quality.

[1485] "Missing data imputation" is the process of filling in gaps in a dataset using statistical or linear interpolation techniques to restore the data to a complete state.

[1486] A "machine learning model" is an algorithm that learns patterns and regularities from data and performs predictions and classifications. Examples include LSTM and ARIMA.

[1487] "Supply and demand forecasting" refers to predicting future electricity demand and supply, a process carried out using machine learning models.

[1488] "Electricity price forecasting" is the process of predicting future fluctuations in electricity prices based on supply and demand forecasts.

[1489] An "electricity procurement plan" is a specific plan for how to procure electricity, and includes factors such as the proportion of renewable energy used, costs, and stability of supply.

[1490] The "means for presenting to the user in an interactive format" refers to a method for interactively displaying the generated electricity procurement plan to the user using graphs and text.

[1491] "Means for recognizing emotions and adjusting the way plans are presented according to the emotional state" refers to technology that has the function of analyzing emotions from the user's facial expressions, voice, etc., and changing the way information is presented according to those emotions.

[1492] "Adherence to planned values ​​and quantities" refers to making real-time adjustments to ensure that planned power supply and consumption match.

[1493] "Means for adjusting supply and demand in real time" refers to a system for instantly adjusting the balance between supply and demand.

[1494] "Means for automatically executing various operations required for power procurement" refers to technology that allows the system to automatically perform operations and procedures for power procurement and management without user intervention.

[1495] This invention is a system that enables businesses and households to procure electricity in an optimal and environmentally friendly manner without requiring specialized knowledge. This system consists of a server, terminals, and users, and performs data collection, data analysis, forecasting, plan generation, plan presentation, emotion recognition, supply and demand planning, and operation automation.

[1496] System configuration and processing overview

[1497] Server-based data collection and cleansing

[1498] The server collects historical electricity consumption data, weather data, and electricity price data from smart meters, the Japan Meteorological Agency's API, market databases, etc.

[1499] Cleanse the collected data and fill in missing data, for example by removing outliers and using linear interpolation and statistical methods to fill in missing data.

[1500] Supply and demand forecasts and price forecasts

[1501] The server uses machine learning models (e.g., LSTM and ARIMA) to predict supply and demand based on past data, taking into account seasonal fluctuations and trends to predict future electricity consumption.

[1502] Based on the results of the supply and demand forecast, the server also predicts electricity prices, predicting price fluctuations according to the supply and demand balance.

[1503] Generate electricity procurement plans

[1504] The server generates multiple electricity procurement plans based on supply-demand and price forecasts, and these plans are optimized taking into account factors such as the proportion of renewable energy, cost, and stability of supply.

[1505] Presenting plans and utilizing emotion recognition

[1506] The terminal presents the electricity procurement plans sent from the server to the user, and details of each plan are displayed in graphs and text, providing an interface that is easy to understand and operate.

[1507] The device is equipped with an emotion engine that recognizes the user's emotions and estimates their emotional state from their facial expressions, voice, input actions, etc. This allows the device to adjust the way the plan is presented to them, taking into account the user's emotions.

[1508] User-Server Interaction

[1509] Users can review the various electricity procurement plans offered through their devices and select the plan that best suits their needs, for example, the "70% renewable energy plan," and specify the maximum monthly cost.

[1510] The terminal receives the user's selection and desired conditions and sends this information to the server, which then uses this information to formulate supply and demand plans and control plans.

[1511] The server formulates a supply and demand plan and a control plan based on the plan selected by the user, and adjusts supply and demand in real time based on this.

[1512] Example: In the case of a company

[1513] 1. Data Collection:

[1514] The server collects a company's electricity consumption data from the past year, local weather data, and electricity market price data.

[1515] 2. Supply and demand forecast:

[1516] The server uses the LSTM model to predict supply and demand for the next year, including increased consumption in the winter.

[1517] 3. Price Prediction:

[1518] The server predicts future electricity prices using an ARIMA model based on supply and demand forecasts.

[1519] 4. Plan generation and presentation:

[1520] The server generates three plans: a 100% renewable energy plan, a 70% renewable energy plan, and a price-conscious plan, and the terminal presents them to the company's energy management officer.

[1521] 5. Plan Selection and Emotion Engine:

[1522] When the energy manager (user) selects Plan B (70% renewable energy plan), the emotion engine analyzes the manager's facial expressions and voice and adjusts the explanation of the plan as necessary.

[1523] 6. Developing supply and demand planning and control plans:

[1524] The server will create a detailed supply and demand plan and control plan based on Plan B.

[1525] Example: For an average household

[1526] 1. Data Collection:

[1527] The server collects the household's electricity consumption data and weather forecast data for the past six months.

[1528] 2. Supply and demand forecast:

[1529] The server uses a time series prediction model to predict power supply and demand up to one month in advance.

[1530] 3. Price Prediction:

[1531] The server uses the ARIMA model to predict electricity prices.

[1532] 4. Plan generation and presentation:

[1533] The server generates plans that use only renewable energy and have low night-time rates, and the terminal presents them to the household.

[1534] 5. Plan Selection and Emotion Engine:

[1535] When a user selects Plan C (a plan with lower nighttime rates), the emotion engine analyzes the user's facial expressions and voice and provides appropriate information.

[1536] 6. Supply and demand planning:

[1537] The server develops a power consumption control plan based on Plan C and submits it to the power company.

[1538] Prompt Sentence Examples

[1539] Below are some example prompts to explain each element or process of the system to the generative AI model:

[1540] User: How do we forecast our company's electricity consumption for next year?

[1541] Terminal: The server collects historical electricity consumption data and weather data, and uses an LSTM model to predict supply and demand, then uses an ARIMA model to predict prices.

[1542] In this way, the entire system works together to achieve optimal power procurement, and the emotional engine is combined to improve the user experience.

[1543] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1544] Program processing steps

[1545] Step 1: Data collection

[1546] The server collects historical electricity consumption data, weather data, and electricity price data.

[1547] Input: Data sources (smart meter API, Japan Meteorological Agency API, market database)

[1548] Specific operations: Retrieve historical electricity consumption data from the smart meter API and store it in a database. Retrieve local weather data using the Japan Meteorological Agency API and store it in a database. Query and retrieve historical electricity price data from the market database.

[1549] Output: Raw dataset for cleansing and imputation

[1550] Step 2: Data cleansing and imputation

[1551] The server cleanses the collected data and completes any missing data.

[1552] Input: raw dataset

[1553] What it does: Detect and remove outliers from datasets, remove duplicates, and impute missing data using linear interpolation and statistical methods.

[1554] Output: Clean dataset

[1555] Step 3: Supply and demand forecast

[1556] The server uses a machine learning model (e.g., LSTM) to predict supply and demand.

[1557] Input: Clean dataset

[1558] What it does: Train an LSTM model and use the clean data as input to predict future electricity demand, for example, predicting electricity consumption for the next year in 30-minute increments.

[1559] Output: Power supply and demand forecast data

[1560] Step 4: Forecasting electricity prices

[1561] The server predicts the electricity price based on the results of the supply and demand forecast.

[1562] Input: Power supply and demand forecast data

[1563] Specific operation: Uses an ARIMA model to predict future electricity prices based on supply and demand forecast data. For example, predicts electricity prices for the next year in 30-minute increments.

[1564] Output: Power price forecast data

[1565] Step 5: Generate a power procurement plan

[1566] The server generates a plurality of power procurement plans based on the supply and demand forecast and the price forecast.

[1567] Input: Power supply and demand forecast data, power price forecast data

[1568] Specific actions: Taking into account the proportion of renewable energy, cost, and stability of supply, three procurement plans will be created: a 100% renewable energy plan, a 70% renewable energy plan, and a price-focused plan.

[1569] Output: Power Procurement Plan

[1570] Step 6: Present your plan

[1571] The terminal presents the power procurement plan transmitted from the server to the user.

[1572] Input: Power Procurement Plan

[1573] Specific operation: Through the GUI, details of each plan (such as trends in electricity usage and the percentage of renewable energy used) are displayed to the user in graphs and text.

[1574] Output: The plan presented to the user

[1575] Step 7: Emotion recognition and plan adjustment

[1576] The terminal recognizes the user's emotions and adjusts the way it presents plans.

[1577] Input: User's facial expression data, voice data

[1578] Specific operation: Analyzes the user's facial expressions and voice through a camera and microphone, and if the user is confused, provides a more concise explanation and adds diagrams and charts.

[1579] Output: Optimized presentation of information to the user

[1580] Step 8: Select a plan and enter your desired conditions

[1581] The user selects the desired plan and sets specific conditions.

[1582] Input: Offered plan, user's desired conditions

[1583] Specific operation: The user selects the desired plan through the device and enters the maximum monthly cost and minimum required renewable energy percentage.

[1584] Output: User's selections and preferences

[1585] Step 9: Develop supply and demand plans and control plans

[1586] The server formulates specific supply and demand plans and control plans based on the user's selections.

[1587] Input: User selections and preferences

[1588] Specific operation: Creates a supply schedule and load balancing plan based on the conditions of the selected plan and saves it in the database.

[1589] Output: Supply and demand planning and control planning

[1590] Step 10: Real-time supply and demand adjustment and operational automation

[1591] The server ensures compliance with planned values, adjusts supply and demand in real time, and automatically executes various necessary operations.

[1592] Input: Supply and demand plan and control plan

[1593] Specific operations: Monitor the supply-demand balance and take action such as shifting the operation of specific devices in the event of an oversupply. Operations are automated and executed.

[1594] Output: Optimized power supply and demand balance

[1595] (Application example 2)

[1596] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1597] Conventional power procurement systems have made it difficult to optimize power consumption in factories and homes, particularly in promoting the use of renewable energy. Furthermore, the information required for users to select an energy plan is often diverse and difficult to understand. This makes it difficult to select the optimal power procurement plan, resulting in sluggish progress in improving power consumption efficiency and reducing costs. Furthermore, these systems have not been able to utilize emotion recognition to improve the user experience.

[1598] The specification process by the specification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting past power consumption data, weather data, and power price data, means for cleansing the collected data and supplementing missing data, and means for predicting power consumption supply and demand using a machine learning model. This makes it possible to provide optimal and green power procurement plans for power consumption in factories and homes.

[1599] The server further includes means for predicting electricity prices based on the results of the supply and demand forecast, means for generating multiple electricity procurement plans based on the supply and demand forecast and the price forecast, means for interactively presenting the generated plans to the user, means for receiving the user's selection and desired conditions and formulating a supply and demand plan and a control plan based on the selected plan, means for adjusting supply and demand in real time to ensure compliance with planned values, means for automatically executing various operations required for electricity procurement, and means for recognizing the user's emotions and adjusting the method of presenting the plans based on the emotions. This makes it possible to select the optimal electricity procurement plan according to the user's emotions, thereby achieving efficient electricity consumption and cost reduction.

[1600] "Past electricity consumption data" refers to information about the history of electricity used by consumers such as businesses and households over a certain period of time.

[1601] "Weather data" is information indicating weather conditions in a specific region or period, and includes temperature, humidity, rainfall, wind speed, and the like.

[1602] "Electricity Price Data" means information about the price of electricity in a particular market or period.

[1603] "Means for collection" refers to means for acquiring electricity consumption data, weather data, and electricity price data using devices such as servers and sensors.

[1604] "Cleansing and missing data imputation methods" are methods for removing outliers and duplicate data from collected data and filling in missing data using linear imputation and statistical methods.

[1605] "Method for forecasting supply and demand of electricity consumption using machine learning models" refers to a method for forecasting future electricity consumption using machine learning algorithms such as LSTM (Long Short-Term Memory) and ARIMA (AutoRegressive Integrated Moving Average).

[1606] "Means for forecasting electricity prices based on the results of supply and demand forecasts" refers to means for using algorithms or models to forecast future electricity prices using the results of supply and demand forecasts as input.

[1607] The "means for generating multiple electricity procurement plans" refers to a means for creating multiple optimal electricity procurement plans based on predicted supply and demand and price data, taking into consideration factors such as the proportion of renewable energy, costs, and stability of supply.

[1608] The "means for presenting the generated plan to the user in an interactive format" refers to a means for visually or audibly presenting the power procurement plan to the user through an interface, thereby enabling a dialogue with the user.

[1609] "Means for receiving a user's selection and desired conditions and formulating a supply and demand plan and a control plan based on the selected plan" refers to means for formulating detailed power consumption distribution and control schedules in accordance with the power procurement plan selected by the user.

[1610] "Means for adjusting supply and demand in real time to ensure compliance with planned values" refers to means for monitoring the power supply and demand balance in real time and making adjustments as necessary.

[1611] "Means for automatically executing various operations necessary for power procurement" refers to means for automating and executing operations such as power purchasing, load balancing, and power redistribution.

[1612] "Means for recognizing the user's emotions and adjusting the way the plan is presented based on those emotions" refers to means for analyzing the user's facial expressions and voice to identify their emotions and appropriately changing the way the plan is presented depending on their emotional state.

[1613] The present invention provides a system in which a server, a terminal, and a user work together to optimize power consumption and utilize renewable energy.

[1614] First, the server collects historical electricity consumption data, weather data, and electricity price data. This data collection is performed using various APIs and database connections. For example, electricity consumption data is obtained from smart meters, weather data is obtained from the Japan Meteorological Agency's API, and price data is obtained from the electricity market database. This collected data is then cleansed and missing data is filled by removing outliers and duplicates, and linear interpolation and statistical methods are used.

[1615] Next, the server uses a machine learning model to predict power consumption supply and demand. For this purpose, models such as LSTM (Long Short-Term Memory) and ARIMA (AutoRegressive Integrated Moving Average) are used. The server also predicts power prices based on the results of the supply and demand forecast. This makes it possible to predict price fluctuations according to the supply and demand balance. Based on the forecast data, multiple power procurement plans are generated, taking into account factors such as the proportion of renewable energy, price, and supply stability.

[1616] The generated electricity procurement plan is sent from the server to the device, which then presents it to the user in an interactive format. The device displays the details of each plan using graphs and text, providing an interface that is easy for the user to understand. Furthermore, the device is equipped with an emotion engine that estimates the user's emotional state from facial expressions, voice, and input actions. This allows the device to adjust the way information is presented based on the user's emotions, for example, by providing a simple explanation if the user is confused.

[1617] Users can review the multiple power procurement plans presented and select the one that best suits their needs. When the user enters the plan they have selected and their desired conditions into the terminal, the information is sent to the server, which then uses this information to formulate supply and demand plans and control plans. The supply and demand plans include supply schedules and load balancing plans, which ensure compliance with planned values ​​for simultaneous and equal amounts. The server adjusts supply and demand in real time and automatically performs various operations required for power procurement.

[1618] As a specific example, if a factory were to use this invention, the server would collect the past year's worth of power consumption data, local weather data, and power market price data, and use this information to make supply and demand forecasts and price predictions. The server would then generate three plans: a 100% renewable energy plan, a 70% renewable energy plan, and a price-focused plan, and present these to the factory's energy manager via a terminal. When the manager selects a plan, the emotion engine would analyze the manager's facial expressions and voice to provide an explanation appropriate to the situation.

[1619] An example prompt is:

[1620] "Please create a program that will forecast supply and demand and electricity prices based on factory electricity consumption data, weather data, and electricity price data, generate an optimal electricity procurement plan, and make suggestions based on the user's emotional state."

[1621] In an embodiment of the present invention, a server, a terminal, and a user work together to optimize power procurement and improve the user experience. This system promotes the use of renewable energy, thereby achieving cost reduction and improved consumption efficiency.

[1622] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1623] Step 1:

[1624] The server collects historical electricity consumption data, weather data, and electricity price data. Specifically, it obtains data from smart meters, the Japan Meteorological Agency's API, market databases, etc. It receives historical data as input and obtains the collected data set as output. This collected data is stored in a database after converting the responses from each data source.

[1625] Step 2:

[1626] The server cleanses the collected data and fills in missing data. Specifically, it removes outliers and duplicates, and fills in missing data using linear interpolation and statistical methods. It takes the collected data as input and produces a cleansed dataset as output. The cleansing process includes filtering and replacing data.

[1627] Step 3:

[1628] The server uses the cleansed data to forecast electricity consumption. This uses machine learning models such as LSTM and ARIMA. The cleansed dataset is received as input, and future supply and demand forecast data is obtained as output. Specifically, time series data is input into the model to forecast consumption for a certain period of time.

[1629] Step 4:

[1630] The server predicts the electricity price based on the results of the supply and demand forecast. To do this, it uses the ARIMA model to predict price fluctuations. It receives supply and demand forecast data as input and obtains electricity price forecast data as output. The model calculates the price according to the supply and demand balance.

[1631] Step 5:

[1632] The server generates multiple electricity procurement plans based on supply and demand forecasts and electricity price forecast data. The plans are optimized taking into account the proportion of renewable energy, costs, and supply stability. Supply and demand forecast data and electricity price forecast data are received as input, and an electricity procurement plan is obtained as output. Specifically, conditions for each plan are set and scenarios are generated based on them.

[1633] Step 6:

[1634] The server sends the generated electricity procurement plan to the terminal, which then presents it to the user in an interactive format. The terminal receives the electricity procurement plan as input and obtains the information to be presented to the user as output. The terminal displays each plan in graphs and text, providing an easy-to-understand presentation to the user.

[1635] Step 7:

[1636] The device recognizes the user's emotions and adjusts the way the plan is presented based on those emotions. It receives the user's facial expressions, voice, and input actions as input, and outputs the emotion estimation results and an appropriate presentation method based on those emotions. The emotion engine analyzes the user's emotional state in real time, and provides simplified explanations if the user is confused, for example.

[1637] Step 8:

[1638] The user reviews the multiple electricity procurement plans provided and selects the plan that best suits their needs. The system receives the electricity procurement plans as input and obtains the selected plan as output. The user also inputs the selected plan and desired conditions into the terminal.

[1639] Step 9:

[1640] The terminal sends the user's selections and desired conditions to the server, which then formulates supply and demand plans and control plans based on them. The server receives the user's selection information as input and obtains detailed supply and demand plans and control plans as output. The supply and demand plans include supply schedules and load balancing plans.

[1641] Step 10:

[1642] The server ensures compliance with planned values ​​and adjusts supply and demand in real time. It receives real-time consumption data and planned data as input, and obtains adjusted supply and demand balance data as output. The server automatically issues control signals to execute various operations.

[1643] Step 11:

[1644] The server automatically executes various operations required for power procurement. It receives supply and demand planning data as input and obtains data on the operations that have been executed as output. Specifically, this includes operations such as power purchasing, load balancing, and power redistribution.

[1645] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1646] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1647] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1648] [Fourth embodiment]

[1649] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1650] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1651] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1652] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1653] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1654] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1655] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1656] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1657] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1658] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1659] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1660] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1661] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1662] This invention is a system that enables businesses and households to procure optimal and green electricity without requiring specialized knowledge. This system consists of a server, terminals, and users, and collects, analyzes, and predicts past electricity consumption data, weather data, and electricity price data, and then generates, presents, and executes an appropriate electricity procurement plan.

[1663] Main components and their operation

[1664] 1. Server Operation

[1665] The server collects historical electricity consumption data, weather data, and electricity price data, cleanses this data, and fills in missing data. To do this, the server obtains the necessary data through APIs and database connections.

[1666] Using the collected and pre-processed data, the server applies machine learning models to generate supply and demand forecasts and price predictions, using time series forecasting algorithms such as LSTM and ARIMA models.

[1667] Based on the forecast results, multiple power procurement plans (e.g., a 100% renewable energy plan, a price-focused plan, etc.) are generated. These plans are optimized based on evaluation criteria such as cost, environmental impact, and stability.

[1668] 2. Device Operation

[1669] The server generates multiple electricity procurement plans and sends them to the terminal, which then presents them to the user in an interactive format. The details of each plan are displayed in graphs and text, providing the user with an interface that is easy to understand and operate.

[1670] The terminal receives the user's selection and desired conditions and transmits them to the server, where the user can enter specific conditions (such as maximum cost or minimum required percentage of renewable energy).

[1671] 3. User Operation

[1672] Users can review multiple electricity procurement plans offered through their terminal and select the plan that best suits their needs.

[1673] Based on the plan selected by the user, the user inputs the necessary adjustments and specific plan details. The information entered by the user is sent via the terminal to the server, which then uses that information to formulate supply and demand plans and control plans.

[1674] Explanation of program processing

[1675] Data collection and preprocessing

[1676] The server collects historical electricity consumption data, weather data, and electricity price data from smart meters, the Japan Meteorological Agency's API, market databases, etc.

[1677] The collected data is cleansed and missing data is filled using linear imputation and statistical methods.

[1678] Supply and demand forecasts and price forecasts

[1679] The server uses machine learning models to forecast supply and demand, including LSTM and ARIMA, to predict future electricity consumption, taking seasonal fluctuations and trends into account.

[1680] Based on the results of the supply and demand forecast, the server also predicts electricity prices, calculating predicted price fluctuations according to the supply and demand balance using ARIMA models and other methods.

[1681] Generate electricity procurement plans

[1682] The server generates multiple procurement plans based on supply and demand forecasts and price forecasts, and the plans are optimized taking into account factors such as the proportion of renewable energy, cost, and stability of supply.

[1683] Plan presentation and selection

[1684] The terminal presents the plan sent from the server to the user and provides information in an interactive format. The interface is designed to meet the user's desired conditions and needs.

[1685] The user selects the most suitable plan and enters their desired conditions. The device then sends this information to the server.

[1686] Supply and demand planning

[1687] Based on the user's selections and desired conditions, the server creates detailed supply and demand plans and control plans, including supply schedules and load balancing plans.

[1688] Automatic execution of operations

[1689] The server ensures compliance with planned values ​​and adjusts supply and demand in real time. All necessary procedures and operations are automatically performed, optimizing power procurement without user intervention.

[1690] Specific examples

[1691] For businesses

[1692] 1. Data Collection:

[1693] The server collects electricity consumption data for the past year for the company, local weather data, and electricity market price data.

[1694] 2. Supply and demand forecast:

[1695] The server uses the LSTM model to predict supply and demand for the next year, including increased consumption in the winter.

[1696] 3. Price Prediction:

[1697] The server predicts future electricity prices using an ARIMA model based on supply and demand forecasts.

[1698] 4. Plan generation and presentation:

[1699] The server generates three plans: a 100% renewable energy plan, a 70% renewable energy plan, and a price-conscious plan, and the terminal presents them to the company's energy management officer.

[1700] 5. Select a plan:

[1701] The user (energy manager) selects Plan B (70% renewable energy plan) and enters specific desired conditions.

[1702] 6. Supply and demand planning:

[1703] The server will create a detailed supply and demand plan and control plan based on Plan B.

[1704] For ordinary households

[1705] 1. Data Collection:

[1706] The server collects the household's electricity consumption data and weather forecast data for the past six months.

[1707] 2. Supply and demand forecast:

[1708] The server uses a time series prediction model to predict power supply and demand up to one month in advance.

[1709] 3. Price Prediction:

[1710] The server uses the ARIMA model to predict electricity prices.

[1711] 4. Plan generation and presentation:

[1712] The server generates plans that use only renewable energy and have low night-time rates, and the terminal presents them to the household.

[1713] 5. Select a plan:

[1714] The user selects Plan C (a plan with a cheaper nighttime rate) and transmits the selection to the server via the terminal.

[1715] 6. Supply and demand planning:

[1716] The server develops a power consumption control plan based on Plan C and submits it to the power company.

[1717] This will provide a system that enables businesses and households to procure electricity effectively without the need for specialized knowledge.

[1718] The processing flow will be explained below.

[1719] Step 1:

[1720] The server collects historical electricity consumption data, weather data, and electricity price data, including data from smart meters, the Japan Meteorological Agency's API, market databases, etc.

[1721] Step 2:

[1722] The server cleanses the collected data, removing outliers and duplicate data to make it more reliable, and then fills in missing data using statistical methods and linear interpolation.

[1723] Step 3:

[1724] The server uses machine learning models to forecast supply and demand. Specifically, it uses LSTM and ARIMA models to input past electricity consumption data and weather data to predict future electricity consumption.

[1725] Step 4:

[1726] The server predicts electricity prices based on the results of the supply and demand forecast, and calculates future electricity prices using ARIMA models and other methods, taking into account the supply and demand balance.

[1727] Step 5:

[1728] The server generates multiple electricity procurement plans based on supply-demand and price forecast data, including factors such as the proportion of renewable energy, price, and stability.

[1729] Step 6:

[1730] The server sends the generated electricity procurement plans to the terminal, which provides the user with an interactive interface that displays the details of each plan in graphs and text.

[1731] Step 7:

[1732] The user compares multiple electricity procurement plans offered through the terminal and selects the plan that best suits their needs. The user then inputs the selected plan and desired conditions into the terminal.

[1733] Step 8:

[1734] The terminal sends the user's selection and desired conditions to the server, which uses the selection and conditions to formulate a supply and demand plan in the next step.

[1735] Step 9:

[1736] The server creates detailed supply and demand plans and control plans based on the user's selections. The supply and demand plans include supply schedules and load balancing plans, and are designed to ensure compliance with planned values.

[1737] Step 10:

[1738] The server adjusts supply and demand in real time to ensure compliance with the planned simultaneous balance. If an imbalance occurs, adjustment measures are automatically implemented.

[1739] Step 11:

[1740] The server automatically executes the various operations required for power procurement, including submitting plans to the power company and making the necessary reports after procurement is confirmed.

[1741] Step 12:

[1742] The electricity procurement is officially executed based on the plan selected by the user. The server monitors the actual electricity consumption data and collects and analyzes feedback to improve the accuracy of the next forecast.

[1743] In this way, the entire system works together to achieve optimal power procurement.

[1744] Example 1

[1745] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1746] Conventional power procurement systems require specialized knowledge, making it difficult for many businesses and households to achieve optimal power procurement. Furthermore, the accuracy of supply and demand forecasts and power price forecasts is low, and environmentally friendly renewable energy use is often not sufficiently considered. Real-time supply and demand adjustments and the provision of information in a format that is easy for users to understand are also insufficient.

[1747] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1748] In this invention, the server includes means for collecting past electricity consumption data, weather data, and electricity price data, means for cleansing the collected data and filling in missing data, means for forecasting electricity consumption supply and demand using a machine learning model, means for forecasting electricity prices based on the supply and demand forecast results, means for generating multiple electricity procurement plans based on the supply and demand forecast and price forecast, means for interactively presenting the generated plans to a user, means for receiving the user's selection and desired conditions and formulating a supply and demand plan and a control plan based on the selected plan, means for adjusting supply and demand in real time to ensure compliance with planned values, means for automatically executing various operations required for electricity procurement, means for providing graphs and text information in a user-understandable format, and means for performing long-term and short-term data analysis for forecasting and planning. This enables users without specialized knowledge to efficiently procure optimal and green electricity.

[1749] "Past power consumption data" is data showing the history of power consumption, and is a record of power usage and consumption patterns over a certain period of time.

[1750] "Weather Data" means data that includes meteorological information such as temperature, humidity, wind speed, and precipitation, and indicates weather conditions for a particular region and period of time.

[1751] "Electricity price data" refers to data showing fluctuations in electricity prices and past price history in the electricity market.

[1752] "Cleansing" refers to the process of organizing collected data and correcting or removing erroneous or incomplete information.

[1753] "Missing data imputation" refers to filling in missing information in an incomplete data set using estimates or other information.

[1754] A "machine learning model" is an algorithm that can learn patterns and rules from data and make predictions and classifications.

[1755] "Demand and supply forecasting" is the process of predicting the future balance of electricity consumption and supply.

[1756] "Electricity price forecasting" refers to predicting future electricity market prices based on the results of supply and demand forecasts, etc.

[1757] An "electricity procurement plan" is a plan that shows how electricity will be procured based on predicted electricity supply and demand and prices.

[1758] "Interactive" refers to a format in which a user interacts with a system to exchange information.

[1759] "Planning based on a selected plan" means creating a detailed supply and demand plan and control plan based on the power procurement plan selected by the user.

[1760] "Adherence to planned values" means making adjustments to match the balance of electricity supply and demand in real time.

[1761] "Automatic execution of operations" means that the system automatically performs the operations and procedures required for power procurement.

[1762] "Providing graphs and text information" means visualizing and presenting information in a way that is easy for users to understand.

[1763] "Long-term and short-term data analysis" is the process of analyzing long-term and short-term data to aid in forecasting and planning.

[1764] This invention is a system that enables businesses and households to procure electricity in an optimal and environmentally friendly manner without requiring specialized knowledge. This system consists of a server, terminals, and users, and collects, analyzes, and predicts past electricity consumption data, weather data, and electricity price data, and then generates, presents, and executes an appropriate electricity procurement plan.

[1765] Server Operations

[1766] Data collection

[1767] The server collects historical electricity consumption data, weather data, and electricity price data using smart meters, the Japan Meteorological Agency's API, and market databases. For example, when obtaining electricity consumption data from smart meters, the server collects the data using a communication protocol.

[1768] Data Cleansing and Imputation

[1769] The server cleanses the collected data and fills in missing data using linear imputation and statistical methods. For example, the cleansing process removes outliers and ensures consistency.

[1770] Supply and demand forecast

[1771] The server uses software such as TensorFlow to build LSTM and ARIMA models, inputs the collected data, and predicts future electricity supply and demand, for example, taking into account peak electricity consumption in the summer.

[1772] Price Prediction

[1773] The server uses the ARIMA model to predict electricity prices based on the results of the supply and demand forecast, using past price data and predicted supply and demand data.

[1774] Generate electricity procurement plans

[1775] The server generates multiple electricity procurement plans based on the results of supply and demand forecasts and price forecasts. The plans include the proportion of renewable energy, cost, stability of supply, etc. For example, it generates a 100% renewable energy plan, a 70% renewable energy plan, a price-focused plan, etc.

[1776] Device operation

[1777] Plan presentation and selection

[1778] The terminal interactively presents the electricity procurement plans sent from the server to the user. The details of each plan are displayed to the user in graphs and text, providing an interface that is easy to understand and operate. For example, a visual graph showing that Plan A has a high proportion of renewable energy use is displayed.

[1779] Collecting user preferences

[1780] The terminal receives the user's desired conditions and sends them to the server. The user can input specific conditions (e.g., maximum cost or minimum required percentage of renewable energy).

[1781] User operations

[1782] Review and select a plan

[1783] Users can review multiple electricity procurement plans offered through their devices and select the plan that best suits their needs. For example, an energy manager might select a plan that uses 70% renewable energy.

[1784] Enter detailed information

[1785] The user inputs the necessary adjustments and specific plan details based on the selected plan. The input information is sent via the terminal to the server, which then uses that information to create a supply-demand plan and a control plan.

[1786] Example prompts for generative AI models

[1787] "Generate a 100% renewable energy electricity procurement plan based on the past year's electricity consumption data, weather data, and electricity price data."

[1788] "Predict electricity consumption for the next year using an LSTM model and present a price-focused electricity procurement plan."

[1789] This invention provides businesses and households with effective and sustainable power supply options tailored to their specific requirements and preferences, enabling them to procure power efficiently without specialized knowledge.

[1790] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1791] Step 1: Data collection

[1792] The server collects historical electricity consumption data, weather data, and electricity price data from smart meters, the Japan Meteorological Agency's API, and the electricity market database. As a specific example, it sends a "GET / electricity / consumption" request from the smart meter to obtain consumption data for the past year and obtains the latest weather data from the Japan Meteorological Agency's API. As a result, it obtains a set of electricity consumption data, weather data, and electricity price data.

[1793] Step 2: Data cleansing and impregnation

[1794] The server cleanses the collected data and imputes missing data. It uses the collected data set as input, and as part of the data cleansing, removes outliers and produces clean data. Missing data is imputed using linear imputation and statistical methods. For example, statistical methods can be used to impute missing values ​​with the mean value, resulting in a complete and consistent data set.

[1795] Step 3: Supply and demand forecast

[1796] The server uses the cleansed and complemented data to input it into an LSTM model built using TensorFlow. The data is then processed by converting the electricity consumption data and weather data into a time series format and applying it to the LSTM model. The LSTM model then predicts future electricity consumption and outputs the supply and demand pattern for the next fiscal year as a result of the prediction.

[1797] Step 4: Price prediction

[1798] The server inputs the results of the supply and demand forecast and uses the ARIMA model to predict electricity prices. It analyzes and processes the supply and demand forecast results and past price data, and predicts future electricity market prices based on this. The output is a predicted future electricity price dataset.

[1799] Step 5: Generate a power procurement plan

[1800] The server generates an electricity procurement plan based on the results of supply and demand forecasts and price forecasts. For example, it generates a 100% renewable energy plan, a 70% renewable energy plan, and a price-focused plan. Supply and demand forecast data and price forecast data are used as input, and optimization is performed when generating each plan, taking into account factors such as the renewable energy ratio, cost, and supply stability. Detailed data for each generated plan is obtained as output.

[1801] Step 6: View and select a plan

[1802] The terminal interactively presents the electricity procurement plans sent from the server to the user. As input, it uses multiple procurement plan data received from the server. Specifically, it displays the details of each plan in an easy-to-understand graph and text format, allowing the user to easily compare and consider them. As output, it generates an information display screen to be presented to the user.

[1803] Step 7: Collect user preferences

[1804] The user inputs desired conditions (for example, maximum cost or minimum percentage of renewable energy) through a terminal. As input, the user's desired condition data is entered into the terminal. The terminal sends this to the server and requests the server to design a plan based on the conditions. As output, the user's desired condition data sent to the server is obtained.

[1805] Step 8: Develop supply and demand plans

[1806] The server formulates detailed supply and demand plans and control plans based on the user's selections and desired conditions. The inputs include the user's desired condition data and detailed data of the selected power procurement plan. To formulate the supply and demand plan, it creates a supply schedule and load balancing plan and issues instructions to various power supply devices and systems. The output generates detailed supply and demand plan data and control plan data.

[1807] Step 9: Automate operations

[1808] The server automatically executes power procurement and supply-demand adjustments based on the formulated plan. Detailed supply-demand and control plan data is used as input. Specifically, it monitors the supply-demand balance in real time and adjusts the procurement plan as needed. The output is an optimized power procurement operation.

[1809] (Application example 1)

[1810] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1811] In modern society, households and businesses lack the knowledge and time to procure optimal, green energy, making efficient and environmentally friendly energy management difficult. In particular, there is a need for a system that can automatically generate plans that take into account the proportion and cost of renewable energy and propose them in a format that is easy for users to understand. Furthermore, it is necessary to adjust supply and demand in real time based on daily electricity supply and demand forecasts and price forecasts, but this is difficult to do without specialized knowledge.

[1812] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1813] In this invention, the server includes: means for collecting past electricity consumption data, weather data, and electricity price data; means for cleansing the collected data and filling in missing data; means for forecasting electricity supply and demand using a machine learning model; means for forecasting electricity prices based on the supply and demand forecast results; means for generating multiple electricity procurement plans based on the supply and demand forecast and price forecast; means for interactively presenting the generated plans to a user; means for receiving the user's selection and desired conditions and formulating a supply and demand plan and a control plan based on the selected plan; means for adjusting supply and demand in real time to ensure compliance with planned values; means for automatically executing various operations required for electricity procurement; means for operating on a smartphone; and data processing means for automatically generating multiple plans with renewable energy usage rates and cost-focused features based on the forecast data. This enables households and businesses to achieve optimal and green electricity procurement and efficient energy management without specialized knowledge.

[1814] "Historical power consumption data" is data...

Claims

1. a means for collecting historical electricity consumption data, weather data, and electricity price data; A means of cleansing the collected data and completing missing data; A means for forecasting supply and demand of electricity consumption using a machine learning model; A means for predicting electricity prices based on the results of the supply and demand forecast; a means for generating a plurality of electricity procurement plans based on supply and demand forecasts and price forecasts; means for interactively presenting the generated plan to a user; means for receiving a user's selection and desired conditions and formulating a supply and demand plan and a control plan based on the selected plan; a means for adjusting supply and demand in real time to ensure compliance with planned values; A means to automatically execute various operations required for power procurement, A system including:

2. The system of claim 1 , wherein the plurality of power procurement plans include a plan using renewable energy and a price-focused plan.

3. The system according to claim 1 , wherein the machine learning model uses a time series prediction model and a regression model.

Images