System

Abstract

A system is provided.SOLUTION: A system, comprising: means for obtaining historical power usage data; means for obtaining real-time power market data and renewable energy supply data; means for analyzing the historical power usage data and the power market data to identify patterns in power consumption; means for building a model to predict future power demand using a machine learning algorithm; means for executing a power procurement optimization algorithm based on the predictive model to generate an optimal power supply plan; means for presenting the optimal power supply plan to a user; and means for monitoring power usage and modifying the power supply plan in real-time.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] Improving the efficiency and reducing costs of electricity use are important challenges for modern businesses and households. However, the complexity of the electricity market and the need for specialized knowledge about the fluctuations in renewable energy supply make optimal electricity procurement difficult for many users. Furthermore, systems for real-time monitoring and adjustment of electricity use are limited, requiring enormous resources. [Means for solving the problem]

[0005] The present invention includes a means for acquiring historical electricity usage data, real-time electricity market data, and renewable energy supply data. This data is analyzed to identify electricity consumption patterns, and a machine learning algorithm is used to build a model for predicting future electricity demand. Furthermore, a power procurement optimization algorithm is executed based on this predictive model to develop an optimal power supply plan. The plan presented to the user is revised as necessary while monitoring power usage in real time. This allows even users without specialized knowledge to efficiently procure electricity and minimize both costs and environmental impact.

[0006] "Electricity usage data" is information about the amount of electricity consumed by a household or business over a certain period of time.

[0007] "Electricity Market Data" means real-time information about the supply and demand of, and prices for, electricity in a particular region.

[0008] "Renewable energy supply data" refers to data on the amount of electricity supplied from renewable energy sources such as solar, wind, and hydroelectric power, and its forecast.

[0009] "Electricity consumption patterns" refer to consumption trends and usage trends during specific time periods or seasons that can be obtained by analyzing historical electricity usage data.

[0010] A "machine learning algorithm" refers to a computational method that automatically learns patterns and knowledge from data and makes predictions and classifications.

[0011] A "forecasting model" is a mathematical model constructed to estimate future electricity demand using past data.

[0012] An "optimization algorithm" is a mathematical method for formulating the most efficient electricity procurement plan under given constraints.

[0013] An "electricity supply plan" is a plan for procuring electricity efficiently and economically, formulated based on future electricity demand and supply.

[0014] "Real-time monitoring" is the process of instantly monitoring current power usage and making adjustments as needed. [Brief explanation of the drawings]

[0015] [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

[0016] 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.

[0017] First, the terms used in the following description will be explained.

[0018] 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).

[0019] 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.

[0020] 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.

[0021] 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.

[0022] 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."

[0023] [First embodiment]

[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0025] 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.

[0026] 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).

[0027] 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.

[0028] 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.

[0029] 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.

[0030] 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.

[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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."

[0036] The present invention relates to an electricity procurement system that enables businesses and households to optimize the efficiency of their electricity usage and reduce costs and environmental impact. The system of the present invention will be described in detail below.

[0037] First, the user accesses the server and signs in to the system. Once the user uploads their past electricity usage data, the server begins collecting and preprocessing the data. The server obtains the user's electricity usage data from the past year through the smart meter system. At the same time, the server collects electricity market data and renewable energy supply data via the Internet. This data includes real-time electricity price information and renewable energy supply forecasts.

[0038] The server then preprocesses and cleanses the data, imputing or removing missing or outlier values ​​and reordering them by timestamp. The server then analyzes the data to identify patterns in power consumption, for example, using histograms and time series analysis to identify periods of peak and low usage.

[0039] Next, the server uses a machine learning algorithm (e.g., ARIMA model or LSTM model) to build a model that predicts future electricity demand. The server divides the acquired data into a training dataset and a test dataset, and uses the training dataset to train the demand forecasting model. The accuracy of the forecasting model is evaluated to confirm its performance.

[0040] The server runs an optimization algorithm for power procurement based on the demand forecast model to develop an optimal power supply plan. Specifically, it uses multivariable linear programming and dynamic programming to calculate the optimal procurement strategy that minimizes costs and environmental impact. The server then presents this optimal plan to the user. The user can view the power supply plan in graph and chart format through the server's management screen. The server also sends notifications to the user and provides details of the plan.

[0041] The server monitors power usage in real time and adjusts the plan as needed. For example, if power consumption is not as expected, the server detects the anomaly and alerts the user. The device then controls smart home appliances and HVAC systems in real time based on the configured power supply plan.

[0042] As a concrete example, consider the case where a user signs in to the system and uploads electricity usage data from the past year. The server collects real-time electricity market data and renewable energy supply forecasts. Based on this, the server predicts electricity demand for the next month and creates and presents an optimal electricity supply plan. The user receives a plan that allows them to reduce costs by 10% by increasing electricity usage during late-night hours. The device adjusts electricity consumption based on this plan, and the server monitors usage in real time. If a consumption pattern that differs from the forecast occurs, the server sends an alert to the user and provides a new optimized plan if necessary.

[0043] The processing flow will be explained below.

[0044] Step 1:

[0045] The user signs in to the server and uploads their historical electricity usage data. The server is then accessed and retrieves the user's electricity usage data from the smart meter system. The server also collects electricity market data and renewable energy supply data via the internet, allowing the system to obtain all the necessary data.

[0046] Step 2:

[0047] The server cleanses and preprocesses the acquired data. Specifically, the server detects missing values ​​and outliers and completes or removes them. The server also sorts the data by timestamp and generates daily and monthly aggregated data. This preprocessing improves the quality of the data.

[0048] Step 3:

[0049] The server analyzes the pre-processed data to identify patterns in power consumption. It uses histograms and time series analysis to identify peak and low usage periods. It also performs cluster analysis to create groups of different consumption patterns. This analysis identifies the characteristics and trends of power usage.

[0050] Step 4:

[0051] The server uses a machine learning algorithm to build a future electricity demand forecasting model. The server divides the data into a training dataset and a test dataset, and uses the training dataset to train a forecasting model (e.g., an ARIMA model or an LSTM model). The model is then evaluated on the test dataset to confirm its forecast accuracy.

[0052] Step 5:

[0053] The server runs an optimization algorithm for power procurement based on a trained power demand forecasting model. Using multivariable linear programming and dynamic programming, the server formulates an optimal power supply plan that minimizes costs and environmental impacts. This plan is created based on future power demand forecasts.

[0054] Step 6:

[0055] The server presents the optimized power supply plan to the user. The server then visualizes the plan and displays it in the form of graphs and charts on the user's management screen. The server also sends detailed information about the plan to the user's smartphone or email, allowing the user to review and select the proposed plan.

[0056] Step 7:

[0057] The device adjusts power consumption based on the set power supply plan. The device controls smart home appliances and HVAC systems, increasing or decreasing power consumption during designated time periods. The device also transmits real-time power usage data to a server.

[0058] Step 8:

[0059] The server monitors power usage in real time and adjusts plans as needed. The server detects unusual consumption patterns or unexpected situations and alerts users. The server also updates predictive models and runs new optimization algorithms based on real-time data.

[0060] Example 1

[0061] 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."

[0062] In modern society, optimizing the efficiency of power usage in businesses and homes and reducing power costs and environmental impact are major challenges. However, conventional systems do not fully automate the collection and analysis of power usage data, resulting in insufficient data to formulate optimal power supply plans. Furthermore, demand forecasting using machine learning algorithms and real-time monitoring and feedback of power usage status are insufficient, making it difficult to optimize power supply.

[0063] 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.

[0064] In this invention, the server includes means for acquiring historical power usage data, means for acquiring real-time power market data and renewable energy supply data, means for analyzing the historical power usage data and the power market data to identify power consumption patterns, means for constructing a model for predicting future power demand using a machine learning algorithm, means for formulating an optimal power supply plan by executing a power procurement optimization algorithm based on the prediction model, means for presenting the optimal power supply plan to a user, and means for monitoring power usage in real time and modifying the power supply plan. This enables automatic collection and analysis of power usage data, construction and verification of a demand forecast model, and real-time presentation and modification of an optimal power supply plan.

[0065] "Past power usage data" is information relating to the amount of power consumed in the past and the timing of its use.

[0066] "Real-time electricity market data" means up-to-date information about current prices and supply conditions in the electricity market.

[0067] "Renewable energy supply data" is information about the amount and forecast of electricity supplied from renewable energy sources, such as solar, wind, and hydroelectric power.

[0068] "Patterns of electricity consumption" are data that indicate trends and characteristics of electricity usage over a specific period of time.

[0069] A "machine learning algorithm" is a computational method for automatically building predictive and classification models using data.

[0070] A "model for predicting future electricity demand" is a model used to predict future electricity consumption from past data and current conditions.

[0071] An "optimal power supply plan" is a power supply plan designed to maximize the efficiency of power use and minimize costs and environmental impacts.

[0072] A "smart meter system" is a measuring device and communication system that measures electricity consumption in real time and collects and transmits that data remotely.

[0073] A "missing value" refers to a value that is missing in a dataset.

[0074] An "outlier" refers to a value in a data set that falls outside the normal range.

[0075] A "training dataset" is a collection of data used to train a machine learning model.

[0076] A "test dataset" is a collection of data used to evaluate the performance of a trained machine learning model.

[0077] "Real-time monitoring of power usage" refers to the process of monitoring power consumption in real time and taking immediate action if necessary.

[0078] A "terminal" is a device for controlling appliances based on a power supply plan.

[0079] The present invention relates to an electricity procurement system that allows businesses and households to optimize the efficiency of their electricity usage and reduce costs and environmental impacts. The system is comprised of a combination of software and hardware that collects, analyzes, and optimizes historical electricity usage data, real-time electricity market data, and renewable energy supply data.

[0080] First, users access the system's web portal or mobile app and sign in to the system by entering their authentication information on the login screen. They then upload their electricity usage data from the past year in a format such as a CSV file. This data is received by the server and stored. The server then automatically collects electricity usage data through the smart meter system and performs preprocessing to fill in or remove missing or outlier values. The server also obtains the latest real-time electricity market data and renewable energy supply data via the internet.

[0081] The server then uses this preprocessed data to analyze patterns of power consumption, for example by converting it into a data frame using Python's pandas library and generating histograms and time series graphs using the matplotlib library to identify periods of peak and low usage.

[0082] Furthermore, we will build a model to forecast electricity demand using machine learning techniques. Specifically, we will train ARIMA and LSTM models using machine learning libraries such as scikit-learn and TensorFlow, and evaluate the performance of the models using training and test datasets. Based on this trained model, it will be possible to forecast future electricity demand with high accuracy.

[0083] The server then uses multivariable linear programming and dynamic programming to run an optimization algorithm for power procurement based on the demand forecast model. This allows the creation of an optimal power supply plan that minimizes costs and environmental impact. Users can view this optimal supply plan in graph and chart format via a GUI. The server also notifies users of details via email and push notifications.

[0084] When the supply plan is actually executed, the device controls smart home appliances and HVAC (heating, ventilation, and air conditioning) systems in real time based on the plan. The server monitors power usage in real time and alerts the user if consumption does not progress as predicted and provides a new optimization plan.

[0085] As a concrete example, consider a case where a user signs in to the system and uploads their electricity usage data from the past year. This data is collected by a server, which also retrieves real-time electricity market data and renewable energy supply forecasts. Based on this, the server predicts electricity demand for the next month and creates and presents an optimal electricity supply plan. For example, one plan could reduce costs by 10% by increasing electricity usage during late-night hours. The device adjusts its electricity consumption based on this plan, and the server monitors the situation in real time. If a consumption pattern differs from the forecast, the server sends an alert to the user and presents a new plan if necessary.

[0086] Examples of input prompts for generative AI models include:

[0087] Design an electricity procurement system that helps businesses and households optimize their electricity usage efficiency and reduce costs and environmental impact. Use historical electricity usage data, real-time electricity market data, and renewable energy supply forecast data to build a model that forecasts electricity demand using machine learning algorithms such as ARIMA and LSTM models. Then, design a system that calculates and proposes the optimal procurement strategy that minimizes costs and environmental impact.

[0088] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0089] Processing steps of this system's program

[0090] Step 1: Sign in users and upload data

[0091] explanation

[0092] Users access the system's web portal or mobile app and sign in by entering their credentials on the login screen.

[0093] Users upload electricity usage data from the past year in a format such as a CSV file.

[0094] input

[0095] User authentication information (ID and password)

[0096] Power usage data (CSV file)

[0097] output

[0098] Authorization of credentials

[0099] Sending power usage data to a server

[0100] Specific actions

[0101] The user opens a web browser, accesses the system's URL, and enters their ID and password on the login screen.

[0102] After logging in, select "Upload Data" on the dashboard and upload the CSV file.

[0103] Step 2: Data collection and preprocessing by the server

[0104] explanation

[0105] The server receives the power usage data uploaded by the user and stores it in storage.

[0106] The server automatically acquires electricity usage data through the smart meter system and performs pre-processing to complement or remove missing or abnormal values.

[0107] The server collects real-time electricity market data and renewable energy supply data via the Internet.

[0108] input

[0109] User-uploaded electricity usage data

[0110] Electricity usage data from smart meter systems

[0111] Real-time electricity market and renewable energy supply data

[0112] output

[0113] Preprocessed dataset

[0114] Specific actions

[0115] The server parses the uploaded CSV file and stores it in a database.

[0116] The server retrieves data from the smart meter system via API and converts it into a data frame using pandas.

[0117] Missing values ​​are imputed with the mean and outliers are removed. The data is reordered by timestamp.

[0118] Step 3: Analyze data and identify consumption patterns

[0119] explanation

[0120] The server analyzes the pre-processed data and identifies patterns of power consumption through histogram and time series analysis.

[0121] input

[0122] Preprocessed dataset

[0123] output

[0124] Power consumption pattern analysis results

[0125] Specific actions

[0126] The server uses Python's matplotlib library to generate a histogram to see the distribution of consumption.

[0127] Through time series analysis, periods of peak and low usage are graphed.

[0128] Step 4: Building an electricity demand forecast model

[0129] explanation

[0130] The server builds a power demand forecasting model using machine learning algorithms such as the ARIMA model and the LSTM model.

[0131] The server splits the data into training and test datasets and trains and evaluates the model.

[0132] input

[0133] Preprocessed dataset

[0134] output

[0135] A trained electricity demand forecasting model

[0136] Specific actions

[0137] The server uses scikit-learn and TensorFlow to split the data into training and test sets.

[0138] Train the model and evaluate its predictive accuracy on the test set.

[0139] Step 5: Optimize power procurement

[0140] explanation

[0141] Based on the demand forecasting model, the server executes optimization algorithms using multivariable linear programming and dynamic programming to formulate an optimal power supply plan.

[0142] input

[0143] A trained electricity demand forecasting model

[0144] output

[0145] Optimal power supply plan

[0146] Specific actions

[0147] The server uses the SciPy library to perform linear programming and obtain the optimal solution.

[0148] Step 6: Present and notify users of the plan

[0149] explanation

[0150] The server provides and notifies the user of the optimal power supply plan in the form of graphs and charts.

[0151] input

[0152] Optimal power supply plan

[0153] output

[0154] Plan visibility and notification for users

[0155] Specific actions

[0156] The server displays graphs on a web page using D3.js or Chart.js and sends emails and push notifications.

[0157] Step 7: Real-time control and monitoring

[0158] explanation

[0159] The terminal controls smart home appliances and HVAC systems in real time based on the power supply plan received from the server.

[0160] The server monitors power usage in real time and sends an alert to the user if it detects an abnormality.

[0161] input

[0162] Optimal power supply plan

[0163] Real-time electricity usage data

[0164] output

[0165] Control commands for smart appliances and HVAC systems

[0166] Alert notification when an abnormality is detected

[0167] Specific actions

[0168] The terminal automatically controls air conditioning and lighting based on the power supply plan.

[0169] The server runs a real-time monitoring system and sends an email notification to the user when abnormal data is detected.

[0170] (Application example 1)

[0171] 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."

[0172] Reducing energy costs and environmental impacts are important issues for modern factories. However, conventional energy management systems have difficulty optimizing energy consumption in real time and are unable to efficiently manage the energy usage status of each piece of equipment. Therefore, effective means to improve the energy efficiency of the entire factory are needed.

[0173] 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.

[0174] In this invention, the server includes: means for acquiring historical energy usage data; means for acquiring real-time energy market data and renewable energy supply data; means for analyzing the historical energy usage data and the energy market data to identify energy consumption patterns; means for constructing a model for predicting future energy demand using a machine learning algorithm; means for formulating an optimal energy supply plan by executing an energy procurement optimization algorithm based on the prediction model; means for presenting the optimal energy supply plan to a user; means for each device to share energy consumption data; means for a central server to generate an optimal operation schedule using the energy consumption data; and means for monitoring energy usage in real time and modifying the energy supply plan. This enables each device in a factory to use energy efficiently, thereby reducing overall energy costs and environmental impact.

[0175] "Energy usage data" refers to information about energy consumed in the past.

[0176] "Energy Market Data" means real-time information on energy prices and supply conditions.

[0177] "Renewable energy supply data" refers to information on supply and forecasts from renewable energy sources such as solar, wind, and hydroelectric power.

[0178] A "machine learning algorithm" is a computational method for automatically learning patterns from data and making predictions and classifications.

[0179] A "predictive model" is a mathematical or computational model constructed to predict future trends based on past data.

[0180] An "energy procurement optimization algorithm" is a calculation method for optimizing the cost and efficiency of energy procurement.

[0181] An "energy supply plan" is a plan that determines the energy supply schedule and management policy.

[0182] "Means for each device to share energy consumption data" refers to a means for each device in a factory to communicate its own energy consumption status to each other.

[0183] "Means for a central server to generate an optimal operation schedule using energy consumption data" refers to a means for creating an optimal operation schedule for equipment based on energy consumption data collected by a central server within the factory.

[0184] "Means for monitoring energy usage in real time" refers to means for observing current energy usage in real time.

[0185] The "means for modifying an energy supply plan" refers to a means for appropriately modifying an existing energy supply plan in accordance with the actual energy usage situation.

[0186] This invention relates to a management system for improving the efficiency of energy use in factories and reducing costs and environmental impact. The system is composed of a central server and software installed on each piece of factory equipment.

[0187] First, the server acquires past energy usage data. Data is collected from smart meters and sensors connected to each device and sent to the server. Users can also upload their past energy usage data.

[0188] The server then retrieves real-time energy market data and renewable energy supply data via the internet, including energy prices and supply information, which is retrieved through internet services and APIs.

[0189] The server preprocesses the collected data by imputing or removing missing or outlier values ​​and reordering the data by timestamp. The server then analyzes the data and identifies patterns of energy consumption using histograms and time series analysis.

[0190] The server then uses a machine learning algorithm (e.g., an LSTM model) to predict future energy demand. The server divides the acquired data into a training dataset and a test dataset, and uses the training dataset to build and train a demand forecasting model.

[0191] Once the predictive model is complete, the server uses it to run an energy procurement optimization algorithm, specifically using multivariable linear programming and dynamic programming to develop an optimal energy supply plan that minimizes costs and environmental impacts.

[0192] The formulated energy supply plan is presented to the user, who can check the plan in graph and chart format on the management screen. Each factory device shares its energy consumption data, and the central server uses this data to generate an optimal operation schedule.

[0193] The server also monitors energy usage in real time, detecting anomalies and sending alerts to users if consumption patterns differ from those expected, providing them with the means to modify their energy supply plans as needed.

[0194] As a specific example, when a user signs in to the system and uploads energy usage data from the past year, the server collects real-time energy market data and renewable energy supply forecasts to predict energy demand for the next month. Based on this, an optimal energy supply plan is created, and the user receives a plan that allows them to reduce costs by 10% by increasing energy usage during late-night hours. Each piece of equipment in the factory adjusts its energy consumption based on this plan, and the server monitors usage in real time. If a consumption pattern that differs from the forecast occurs, the server sends an alert to the user and provides a new optimized plan.

[0195] Prompt Sentence Examples

[0196] Energy consumption data:

[0197] timestamp: 2023-10-01 00:00:00, consumption: 50

[0198] timestamp: 2023-10-01 01:00:00, consumption: 55

[0199] ...

[0200] timestamp: 2023-10-01 23:00:00, consumption: 58

[0201] Forecast your energy needs for the next month and provide an optimal energy supply plan. Include specific actions to reduce energy usage during peak hours.

[0202] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0203] Step 1:

[0204] The server collects past energy usage data. Data is sent to the server from smart meters and sensors connected to each device. The input here is the measurement data from each smart meter and sensor, and the output is an energy usage dataset that compiles this data. Specifically, the server periodically receives data from each device and stores it in a database.

[0205] Step 2:

[0206] The server obtains real-time energy market data and renewable energy supply data. This data includes energy prices and supply information and is obtained from an external API via the internet. The input is the response data from the API, and the output is the obtained market data and supply data. Specifically, the server calls the API at regular intervals to obtain the latest data.

[0207] Step 3:

[0208] The server preprocesses the collected data. Specifically, it complements or removes missing values ​​and outliers, and rearranges the data in timestamp order. The input is the data collected in steps 1 and 2, and the output is clean data that has been preprocessed. Specifically, the server automatically performs data cleansing and performs any necessary data transformations.

[0209] Step 4:

[0210] The server analyzes the pre-processed data to identify patterns in energy consumption. Histograms and time series analysis are used to identify periods of peak and low usage. The input is clean data, and the output is analysis results that show consumption patterns. Specifically, the server runs time series data analysis algorithms to extract specific patterns and trends.

[0211] Step 5:

[0212] The server uses a machine learning algorithm (e.g., an LSTM model) to build a model that predicts future energy demand. First, it divides the collected data into a training dataset and a test dataset, and then trains the model based on the training dataset. The input is the preprocessed data and the machine learning algorithm, and the output is a trained predictive model. Specifically, the server uses a machine learning framework (e.g., TensorFlow or Keras) to train the model.

[0213] Step 6:

[0214] The server uses the forecasting model to predict future energy demand. The input is a trained forecasting model and the latest dataset, and the output is the future demand forecast result. Specifically, the server inputs the latest data into the demand forecasting model and generates the forecast result.

[0215] Step 7:

[0216] The server executes an optimization algorithm for energy procurement and formulates an optimal energy supply plan. It minimizes costs and environmental impacts using multivariable linear programming and dynamic programming. The inputs are the demand forecast results and the optimization algorithm, and the output is an optimal energy supply plan. Specifically, the server executes optimization calculations and generates the optimal plan within the specified constraints.

[0217] Step 8:

[0218] The server presents the formulated energy supply plan to the user. The user can check the plan in graph and chart format through the management screen. The input is the optimal energy supply plan, and the output is the supply plan displayed to the user. Specifically, the server visualizes the plan and displays it on a dashboard.

[0219] Step 9:

[0220] Each piece of factory equipment shares its energy consumption data, and a central server uses this data to generate an optimal operation schedule. The input is the energy consumption data from each piece of equipment, and the output is the optimal operation schedule. Specifically, the server periodically collects data from each piece of equipment and calculates the optimal schedule.

[0221] Step 10:

[0222] The server monitors energy usage in real time, alerts users if anomalies occur, and modifies energy supply plans as needed. The input is real-time energy usage data, and the output is modified energy supply plans and alerts. Specifically, the server tracks consumption data through the monitoring system and takes action when anomalies are detected.

[0223] 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.

[0224] The present invention is an electricity procurement system for businesses and households to optimize the efficiency of electricity usage and reduce costs and environmental burdens. The system further incorporates an emotion engine to recognize user emotions and improve the effectiveness of supply plans. The system of the present invention will be described in detail below.

[0225] First, the user accesses the server and signs in to the system. Once the user uploads their past electricity usage data, the server begins collecting and preprocessing the data. The server obtains the user's electricity usage data from the past year through the smart meter system. At the same time, the server collects electricity market data and renewable energy supply data via the Internet. This data includes real-time electricity price information and renewable energy supply forecasts.

[0226] The server then preprocesses and cleanses the data, imputing or removing missing or outlier values ​​and reordering them by timestamp. The server then analyzes the data to identify patterns in power consumption, for example, using histograms and time series analysis to identify periods of peak and low usage.

[0227] Next, the server uses a machine learning algorithm (e.g., ARIMA model or LSTM model) to build a model that predicts future electricity demand. The server divides the acquired data into a training dataset and a test dataset, and uses the training dataset to train the demand forecasting model. The accuracy of the forecasting model is evaluated to confirm its performance.

[0228] The server runs an optimization algorithm for power procurement based on the demand forecast model to develop an optimal power supply plan. Specifically, it uses multivariable linear programming and dynamic programming to calculate the optimal procurement strategy that minimizes costs and environmental impacts. The server then presents this optimal plan to the user. The user can view the power supply plan in graph and chart format through the server's management screen. The server also sends notifications to the user and provides details of the plan.

[0229] This is where the emotion engine comes in. The emotion engine acquires the user's emotion data, analyzes it, and evaluates the effectiveness of the power supply plan. For example, if the user is feeling stressed, it can use that information to suggest an energy consumption plan that will help reduce stress. The emotion engine collects emotion data using voice and image analysis.

[0230] The server monitors power usage in real time and adjusts the plan as needed. For example, if power consumption is not as expected, the server detects the anomaly and alerts the user. The server also updates the predictive model based on real-time data and runs new optimization algorithms.

[0231] As a concrete example, consider the case where a user signs in to the system and uploads electricity usage data from the past year. The server collects real-time electricity market data and renewable energy supply forecasts. Based on this, the server predicts electricity demand for the next month and creates and presents an optimal electricity supply plan. The user reviews the plan, and the emotion engine recognizes the user's stress level, and the server suggests a plan that will reduce stress. The device adjusts electricity consumption based on this plan, and the server monitors usage in real time. If a consumption pattern that differs from the forecast occurs, the server sends an alert to the user and provides a new optimized plan if necessary.

[0232] The processing flow will be explained below.

[0233] Step 1:

[0234] A user signs in to the server and uploads their historical electricity usage data. The server retrieves the past year's electricity usage data from the smart meter system, and also collects electricity market data and renewable energy supply data via the Internet.

[0235] Step 2:

[0236] The server cleanses and preprocesses the acquired data. Specifically, the server detects, imputes, or removes missing or outlier values. It also sorts the data by timestamp and generates daily and monthly aggregated data.

[0237] Step 3:

[0238] The server analyzes the pre-processed data to identify patterns of power consumption, using histograms and time series analysis to identify periods of peak and low usage, and then performs cluster analysis to create distinct consumption pattern groups.

[0239] Step 4:

[0240] The server uses a machine learning algorithm to build a future electricity demand forecasting model. The server divides the data into a training dataset and a test dataset, and trains an ARIMA model or LSTM model using the training dataset. The model's predictive accuracy is then evaluated using the test dataset.

[0241] Step 5:

[0242] The server runs an optimization algorithm for power procurement based on the trained power demand forecasting model. Specifically, the server uses multivariable linear programming and dynamic programming to formulate an optimal power supply plan that minimizes costs and environmental impacts.

[0243] Step 6:

[0244] The server presents the optimized power supply plan to the user, who can then view the plan on the server's management screen and receive detailed information through notifications. The plan is visually displayed in graphs and charts.

[0245] Step 7:

[0246] The emotion engine works to obtain the user's emotion data. The emotion engine uses voice and image analysis to read the user's emotion in real time. For example, it analyzes the user's facial expressions and tone of voice through a camera or microphone.

[0247] Step 8:

[0248] The server analyzes the acquired emotional data and evaluates the user's current emotional state. For example, if the user is feeling stressed, the server adjusts the power supply plan based on that information.

[0249] Step 9:

[0250] Based on the data provided by the emotion engine, the server re-optimizes the power supply plan. A new power supply plan with a stress-reducing effect is formulated and presented to the user. The server then visually displays the plan again and sends a notification to the user.

[0251] Step 10:

[0252] The device adjusts power consumption based on the new power supply plan set by the device, controls smart home appliances and HVAC systems, and increases or decreases power consumption during specified times. The device transmits power usage data to the server in real time.

[0253] Step 11:

[0254] The server monitors power usage in real time and adjusts the plan as needed. If it detects abnormal consumption patterns or unexpected conditions, it will alert the user and provide a new optimized plan accordingly.

[0255] Example 2

[0256] 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."

[0257] The problem that this invention aims to solve is to optimize the efficiency of power usage in businesses and homes and reduce costs and environmental burdens. Conventional power supply systems do not propose power supply plans that take user emotions into consideration, making it difficult to improve user satisfaction. Another problem is the lack of real-time monitoring of power usage and rapid response after anomaly detection.

[0258] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for acquiring historical power usage data; means for acquiring real-time power market data and renewable energy supply data; means for analyzing the historical power usage data and the power market data to identify power consumption patterns; means for constructing a model for predicting future power demand using a machine learning algorithm; means for formulating an optimal power supply plan by executing a power procurement optimization algorithm based on the prediction model; means for presenting the optimal power supply plan to a user; means for acquiring and analyzing user emotion data to evaluate the effectiveness of the power supply plan; and means for monitoring power usage in real time and modifying the power supply plan. This makes it possible to reduce costs and environmental burdens by optimizing power usage efficiency, improve user satisfaction, and achieve quick responses in real time.

[0259] "Past power usage data" refers to data relating to the amount of power consumed by the user in the past and the time of day.

[0260] "Real-time electricity market data" means information about the current electricity market price and supply.

[0261] "Renewable energy supply data" is information about electricity supplied from renewable energy sources such as wind, solar, and hydroelectric power.

[0262] The "means for identifying patterns of power consumption" is a means for analyzing power usage data and identifying trends in power consumption and peak times during a specific period.

[0263] A "machine learning algorithm" is a method or technology for predicting future data by analyzing and learning from data.

[0264] A "model for predicting future electricity demand" is a mathematical or statistical model for estimating future electricity consumption based on collected data.

[0265] The "power procurement optimization algorithm" is a calculation method for efficiently procuring electricity while taking into account factors such as cost and environmental impact.

[0266] An "optimal power supply plan" is a power supply strategy that meets users' power demand while minimizing costs and environmental burdens.

[0267] "User emotional data" is information about the user's psychological state and emotions, and is data collected through voice and image analysis.

[0268] "Means for monitoring power usage in real time" refers to means for constantly monitoring users' power usage and responding immediately if any abnormalities are detected.

[0269] The present invention is an electricity procurement system for optimizing the efficiency of electricity usage in businesses and homes, reducing costs and environmental burdens, and further incorporates an emotion engine for recognizing user emotions to improve the effectiveness of supply plans.

[0270] Overall system overview

[0271] First, the user accesses the server and signs in to the system. Once the user uploads their past electricity usage data, the server begins collecting and preprocessing the data. The server obtains the user's electricity usage data from the past year through the smart meter system. At the same time, the server collects electricity market data and renewable energy supply data via the Internet. This data includes real-time electricity price information and renewable energy supply forecasts.

[0272] Hardware and Software Use

[0273] The server preprocesses the uploaded data using Python's Pandas library and NumPy. Specifically, it imputes missing values ​​and removes outliers. It then uses data visualization tools such as Matplotlib and Seaborn to identify patterns in power consumption through histograms and time series analysis.

[0274] Next, the server builds a machine learning model using Scikit-learn and TensorFlow. It uses ARIMA and LSTM models to predict future electricity demand. In this process, the collected data is divided into a training dataset and a test dataset, and the model is trained.

[0275] To develop an optimal power supply plan, optimization libraries such as Gurobi and PuLP are used. The server uses multivariable linear programming and dynamic programming to calculate the optimal procurement strategy that minimizes costs and environmental impacts. The plan is presented to the user in the form of graphs and charts using the Plotly library.

[0276] Furthermore, the emotion engine collects and analyzes the user's emotional data through voice and image analysis. For example, it uses emotion analysis tools such as AWS Rekognition and Watson. If the user is feeling stressed, it can use that information to suggest a stress-relieving plan.

[0277] Adding specific examples

[0278] As a concrete example, consider the case where a user signs in to the system and uploads electricity usage data from the past year. The server collects real-time electricity market data and renewable energy supply forecasts. Based on this, the server predicts electricity demand for the next month and creates and presents an optimal electricity supply plan. The user reviews the plan, and the emotion engine recognizes the user's stress level, and the server suggests a plan that will reduce stress. The device adjusts electricity consumption based on this plan, and the server monitors usage in real time. If a consumption pattern that differs from the forecast occurs, the server sends an alert to the user and provides a new optimized plan if necessary.

[0279] Prompt Sentence Examples

[0280] An example of an input prompt for a generative AI model is as follows:

[0281] "Please tell me the detailed steps to implement a system where a user can upload electricity usage data from the past year and the system will suggest the optimal electricity supply plan. Please also provide a concrete coding example."

[0282] By implementing this system, it is possible to reduce costs and environmental burdens by optimizing power usage efficiency, improve user satisfaction, and realize quick responses in real time.

[0283] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0284] Step 1:

[0285] A user accesses the system's web application and signs in by entering their username and password. The input is the username and password, and the output is an authentication token. The server checks the user authentication information in the database and generates an authentication token if authentication is successful. Specifically, the user opens a browser, enters a URL to access the system's login page, enters their username and password in the form, and clicks the "Login" button.

[0286] Step 2:

[0287] Users upload the electricity usage data from the past year to the system from their own devices. The input is a file of past electricity usage data, and the output is the electricity usage data stored in the server's database. The server receives this data through the smart meter system and stores it in the database. Specifically, the user clicks the "Upload data" button, selects the data file, and uploads it.

[0288] Step 3:

[0289] The server collects electricity market data and renewable energy supply data via the Internet. The input is electricity market data and renewable energy data from APIs and databases, and the output is the server's database where this data is stored. Specifically, the server periodically calls external APIs to collect electricity market data and renewable energy data.

[0290] Step 4:

[0291] The server performs preprocessing on the collected data. The input is the uploaded electricity usage data, electricity market data, and renewable energy data, and the output is the cleansed data. Specifically, it uses Python's Pandas library and NumPy to fill in missing values ​​and remove outliers. Specifically, the server fills in missing values ​​using Pandas' DataFrame.fillna() method and removes outliers using the DataFrame.drop() method.

[0292] Step 5:

[0293] The server analyzes the cleansed data to identify patterns in power consumption. The input is the cleansed power usage data, and the output is information about power consumption patterns. Histograms and time series analysis are used to identify periods of peak and low usage. Specifically, Matplotlib and Seaborn are used to create histograms and visualize consumption patterns.

[0294] Step 6:

[0295] The server uses a machine learning algorithm to build an electricity demand forecasting model. The input is the cleansed dataset, and the output is the demand forecasting model. Scikit-learn and TensorFlow are used to train ARIMA and LSTM models. Specifically, the dataset is split into training data and test data, and the data is separated using Scikit-learn's train_test_split() method.

[0296] Step 7:

[0297] The server runs an optimization algorithm for power procurement based on the forecast model. The input is a demand forecast model and electricity market data, and the output is an optimal power supply plan. It uses optimization libraries such as Gurobi and PuLP to perform multivariable linear programming and dynamic programming. Specifically, it sets up an optimization problem in Gurobi and calculates the optimal power supply plan.

[0298] Step 8:

[0299] The server presents the optimal power supply plan to the user. The input is the optimal power supply plan, and the output is a plan in the form of graphs and charts that the user can view. Using the Plotly library, the optimal plan is visualized and displayed on a web page. Specifically, the server creates a graph using Plotly and displays it on an admin screen that the user can access.

[0300] Step 9:

[0301] The emotion engine acquires and analyzes the user's emotional data. The input is voice and image data, and the output is analyzed emotional data. Emotions are analyzed using AWS Rekognition or Watson, and the server evaluates the effectiveness of the plan. Specifically, the camera acquires the user's facial expression data, which is then sent to the emotion analysis tool to obtain the analysis results.

[0302] Step 10:

[0303] The server monitors power usage in real time and sends an alert to the user if it detects an abnormality. The input is real-time power usage data and the output is an alert message. Monitoring and alerts are implemented using Prometheus and Grafana. Specifically, the server collects real-time data and sends a notification to the user if the data exceeds a specified threshold.

[0304] (Application example 2)

[0305] 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."

[0306] Conventional power procurement systems formulate power supply plans based on past power usage data and market data, but they have limitations in terms of real-time adjustments to supply plans, particularly optimization that takes user sentiment data into account. Furthermore, optimizing power usage is difficult, particularly in places with large-scale power consumption such as factories, and efficient energy management is required. This presents a challenge in that it is not possible to fully reduce power costs and environmental impact.

[0307] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0308] In this invention, the server includes: means for acquiring historical power usage data; means for acquiring real-time power market data and renewable energy supply data; means for analyzing the historical power usage data and the power market data to identify power consumption patterns; means for constructing a model for predicting future power demand using a machine learning algorithm; means for formulating an optimal power supply plan by executing a power procurement optimization algorithm based on the prediction model; means for presenting the optimal power supply plan to a user; means for monitoring power usage in real time and revising the power supply plan; means for acquiring and analyzing user emotion data; means for revising the power supply plan based on the emotion data; and means for optimizing power usage of factory robots using a smartphone application. This enables real-time optimization of power usage and efficient energy management in a factory while taking user emotion data into consideration.

[0309] "Past power usage data" is history data relating to the amount of power consumed by the user in the past.

[0310] "Real-time electricity market data" means instantaneous data showing the current market electricity prices and supply situation.

[0311] "Renewable energy supply data" refers to data on electricity supply from renewable energy sources such as solar, wind, and hydroelectric power.

[0312] The "power consumption pattern" is an analysis result that indicates the user's power consumption tendency by time period, peak time periods, and low consumption time periods.

[0313] A "machine learning algorithm" is a computer science technique for learning patterns and trends from large amounts of data and making predictions.

[0314] A "model for predicting electricity demand" is a mathematical model constructed to predict future electricity consumption from past data.

[0315] An "optimization algorithm for power procurement" is a mathematical method for calculating the optimal power supply method to minimize power costs and environmental impact.

[0316] An "electricity supply plan" is a plan that determines how electricity will be procured and supplied for future electricity usage.

[0317] The "means for presenting to the user" refers to a method for providing information such as a power supply plan to the user visually or as a notification.

[0318] "Means for monitoring power usage in real time" refers to a system for instantly monitoring current power consumption.

[0319] "User emotion data" is data that indicates the user's emotional state, and is obtained by voice or image analysis, for example.

[0320] A "smartphone application" is software that runs on a smartphone and provides specific functions or services.

[0321] "Means for optimizing power usage of factory robots" is a system for efficiently managing and minimizing power usage of robots operating in factories.

[0322] The present invention is a system for optimizing the efficiency of power usage in businesses and households to reduce costs and environmental burdens, and in particular, uses a smartphone application to improve the efficiency of power usage in factory robots. This system optimizes power supply plans taking into account user emotional data. Specific embodiments of the present invention are described in detail below.

[0323] The system includes a server, a smartphone application, and an emotion data acquisition unit. The server acquires historical electricity usage data and collects real-time electricity market data and renewable energy supply data. It also preprocesses and cleanses this data to identify electricity consumption patterns. This data is then used to build a model for predicting future electricity demand using a machine learning algorithm, and an optimal electricity supply plan is developed.

[0324] Specific software and hardware used in this embodiment include the requests library for data collection, the pandas library for data preprocessing, and the scikit-learn library for machine learning algorithms. Also, EmotionRecognizer is used as software for acquiring user emotion data.

[0325] For example, a user (factory manager) signs in to a smartphone application and uploads electricity usage data from the past year. The server collects real-time electricity market data and renewable energy supply data and analyzes electricity consumption patterns. The server then uses machine learning algorithms to predict future electricity demand and create an optimal electricity supply plan, which it then presents to the user. The server also obtains the user's emotional data and modifies the plan according to their emotional state. The smartphone application also optimizes the power usage of factory robots.

[0326] For example, after a factory manager signs in to the system and uploads data, the server collects electricity market data in real time and executes a process to forecast the next month's electricity demand. During this process, emotion recognition software detects the manager's stress level and presents a power supply plan that will reduce stress. An example of a specific prompt could be to instruct the server, "Please predict electricity usage based on the following data."

[0327] In this way, the system optimizes power usage in factories in real time, enabling efficient energy management.

[0328] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0329] Step 1:

[0330] The server acquires past electricity usage data. When a user uploads electricity usage data from the past year via a smartphone application, the data is sent to the server. The server collects the data using an API endpoint and stores it in a database. The input is the past electricity usage data, and the output is the stored electricity usage data.

[0331] Step 2:

[0332] The server obtains real-time electricity market data and renewable energy supply data. It accesses APIs via the internet to collect current market data and renewable energy supply forecasts. The collected data is stored in the server's database. The input is the real-time electricity market data and renewable energy supply data, and the output is the stored market data and supply data.

[0333] Step 3:

[0334] The server preprocesses the acquired historical electricity usage data and market data to identify electricity consumption patterns. Specifically, it cleanses the data and fills or removes missing values ​​and outliers. Using the cleansed data, it identifies peak and low electricity usage periods through histogram and time series analysis. The input is historical electricity usage data and market data, and the output is electricity consumption patterns.

[0335] Step 4:

[0336] The server uses a machine learning algorithm to build a model to predict future electricity demand. Specifically, it divides the data into a training dataset and a test dataset, and trains a linear regression model using, for example, the scikit-learn library. The input is preprocessed electricity usage data and market data, and the output is the trained demand forecasting model.

[0337] Step 5:

[0338] The server executes an optimization algorithm for power procurement based on the trained forecasting model to formulate an optimal power supply plan. For example, it applies multivariable linear programming based on the predicted power demand to minimize costs and environmental impact. The input is the trained demand forecasting model, and the output is the optimal power supply plan.

[0339] Step 6:

[0340] The server presents the optimal power supply plan to the user, who can then check the plan through a smartphone application. The plan is visually displayed in graphs and charts, and detailed information is also provided. The input is the optimal power supply plan, and the output is the plan presented to the user.

[0341] Step 7:

[0342] The server monitors power usage in real time and modifies the power supply plan. It acquires data sequentially, compares the predictive model with the actual data, detects anomalies, and modifies the plan as necessary. The input is real-time power usage data, and the output is the modified power supply plan.

[0343] Step 8:

[0344] The server acquires and analyzes the user's emotional data. It collects audio and images from smartphones and other devices and analyzes the user's emotional state using EmotionRecognizer software. The input is the user's emotional data, and the output is the analysis results.

[0345] Step 9:

[0346] The server modifies the power supply plan based on the emotion data. For example, if the user is feeling stressed, it generates a plan to reduce power consumption. The input is the analyzed emotion data, and the output is the modified power supply plan.

[0347] Step 10:

[0348] The server uses a smartphone application to optimize the power usage of factory robots. Users can check the robot's operating status and power consumption through the app and operate it according to the optimized plan. The input is the factory robot's power usage data, and the output is the optimized power usage.

[0349] 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.

[0350] 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.

[0351] 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.

[0352] [Second embodiment]

[0353] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0354] 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.

[0355] 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).

[0356] 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.

[0357] 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.

[0358] 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).

[0359] 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.

[0360] 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.

[0361] 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.

[0362] 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.

[0363] 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.

[0364] 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."

[0365] The present invention relates to an electricity procurement system that enables businesses and households to optimize the efficiency of their electricity usage and reduce costs and environmental impact. The system of the present invention will be described in detail below.

[0366] First, the user accesses the server and signs in to the system. Once the user uploads their past electricity usage data, the server begins collecting and preprocessing the data. The server obtains the user's electricity usage data from the past year through the smart meter system. At the same time, the server collects electricity market data and renewable energy supply data via the Internet. This data includes real-time electricity price information and renewable energy supply forecasts.

[0367] The server then preprocesses and cleanses the data, imputing or removing missing or outlier values ​​and reordering them by timestamp. The server then analyzes the data to identify patterns in power consumption, for example, using histograms and time series analysis to identify periods of peak and low usage.

[0368] Next, the server uses a machine learning algorithm (e.g., ARIMA model or LSTM model) to build a model that predicts future electricity demand. The server divides the acquired data into a training dataset and a test dataset, and uses the training dataset to train the demand forecasting model. The accuracy of the forecasting model is evaluated to confirm its performance.

[0369] The server runs an optimization algorithm for power procurement based on the demand forecast model to develop an optimal power supply plan. Specifically, it uses multivariable linear programming and dynamic programming to calculate the optimal procurement strategy that minimizes costs and environmental impact. The server then presents this optimal plan to the user. The user can view the power supply plan in graph and chart format through the server's management screen. The server also sends notifications to the user and provides details of the plan.

[0370] The server monitors power usage in real time and adjusts the plan as needed. For example, if power consumption is not as expected, the server detects the anomaly and alerts the user. The device then controls smart home appliances and HVAC systems in real time based on the configured power supply plan.

[0371] As a concrete example, consider the case where a user signs in to the system and uploads electricity usage data from the past year. The server collects real-time electricity market data and renewable energy supply forecasts. Based on this, the server predicts electricity demand for the next month and creates and presents an optimal electricity supply plan. The user receives a plan that allows them to reduce costs by 10% by increasing electricity usage during late-night hours. The device adjusts electricity consumption based on this plan, and the server monitors usage in real time. If a consumption pattern that differs from the forecast occurs, the server sends an alert to the user and provides a new optimized plan if necessary.

[0372] The processing flow will be explained below.

[0373] Step 1:

[0374] The user signs in to the server and uploads their historical electricity usage data. The server is then accessed and retrieves the user's electricity usage data from the smart meter system. The server also collects electricity market data and renewable energy supply data via the internet, allowing the system to obtain all the necessary data.

[0375] Step 2:

[0376] The server cleanses and preprocesses the acquired data. Specifically, the server detects missing values ​​and outliers and completes or removes them. The server also sorts the data by timestamp and generates daily and monthly aggregated data. This preprocessing improves the quality of the data.

[0377] Step 3:

[0378] The server analyzes the pre-processed data to identify patterns in power consumption. It uses histograms and time series analysis to identify peak and low usage periods. It also performs cluster analysis to create groups of different consumption patterns. This analysis identifies the characteristics and trends of power usage.

[0379] Step 4:

[0380] The server uses a machine learning algorithm to build a future electricity demand forecasting model. The server divides the data into a training dataset and a test dataset, and uses the training dataset to train a forecasting model (e.g., an ARIMA model or an LSTM model). The model is then evaluated on the test dataset to confirm its forecast accuracy.

[0381] Step 5:

[0382] The server runs an optimization algorithm for power procurement based on a trained power demand forecasting model. Using multivariable linear programming and dynamic programming, the server formulates an optimal power supply plan that minimizes costs and environmental impacts. This plan is created based on future power demand forecasts.

[0383] Step 6:

[0384] The server presents the optimized power supply plan to the user. The server then visualizes the plan and displays it in the form of graphs and charts on the user's management screen. The server also sends detailed information about the plan to the user's smartphone or email, allowing the user to review and select the proposed plan.

[0385] Step 7:

[0386] The device adjusts power consumption based on the set power supply plan. The device controls smart home appliances and HVAC systems, increasing or decreasing power consumption during designated time periods. The device also transmits real-time power usage data to a server.

[0387] Step 8:

[0388] The server monitors power usage in real time and adjusts plans as needed. The server detects unusual consumption patterns or unexpected situations and alerts users. The server also updates predictive models and runs new optimization algorithms based on real-time data.

[0389] Example 1

[0390] 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."

[0391] In modern society, optimizing the efficiency of power usage in businesses and homes and reducing power costs and environmental impact are major challenges. However, conventional systems do not fully automate the collection and analysis of power usage data, resulting in insufficient data to formulate optimal power supply plans. Furthermore, demand forecasting using machine learning algorithms and real-time monitoring and feedback of power usage status are insufficient, making it difficult to optimize power supply.

[0392] 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.

[0393] In this invention, the server includes means for acquiring historical power usage data, means for acquiring real-time power market data and renewable energy supply data, means for analyzing the historical power usage data and the power market data to identify power consumption patterns, means for constructing a model for predicting future power demand using a machine learning algorithm, means for formulating an optimal power supply plan by executing a power procurement optimization algorithm based on the prediction model, means for presenting the optimal power supply plan to a user, and means for monitoring power usage in real time and modifying the power supply plan. This enables automatic collection and analysis of power usage data, construction and verification of a demand forecast model, and real-time presentation and modification of an optimal power supply plan.

[0394] "Past power usage data" is information relating to the amount of power consumed in the past and the timing of its use.

[0395] "Real-time electricity market data" means up-to-date information about current prices and supply conditions in the electricity market.

[0396] "Renewable energy supply data" is information about the amount and forecast of electricity supplied from renewable energy sources, such as solar, wind, and hydroelectric power.

[0397] "Patterns of electricity consumption" are data that indicate trends and characteristics of electricity usage over a specific period of time.

[0398] A "machine learning algorithm" is a computational method for automatically building predictive and classification models using data.

[0399] A "model for predicting future electricity demand" is a model used to predict future electricity consumption from past data and current conditions.

[0400] An "optimal power supply plan" is a power supply plan designed to maximize the efficiency of power use and minimize costs and environmental impacts.

[0401] A "smart meter system" is a measuring device and communication system that measures electricity consumption in real time and collects and transmits that data remotely.

[0402] A "missing value" refers to a value that is missing in a dataset.

[0403] An "outlier" refers to a value in a data set that falls outside the normal range.

[0404] A "training dataset" is a collection of data used to train a machine learning model.

[0405] A "test dataset" is a collection of data used to evaluate the performance of a trained machine learning model.

[0406] "Real-time monitoring of power usage" refers to the process of monitoring power consumption in real time and taking immediate action if necessary.

[0407] A "terminal" is a device for controlling appliances based on a power supply plan.

[0408] The present invention relates to an electricity procurement system that allows businesses and households to optimize the efficiency of their electricity usage and reduce costs and environmental impacts. The system is comprised of a combination of software and hardware that collects, analyzes, and optimizes historical electricity usage data, real-time electricity market data, and renewable energy supply data.

[0409] First, users access the system's web portal or mobile app and sign in to the system by entering their authentication information on the login screen. They then upload their electricity usage data from the past year in a format such as a CSV file. This data is received by the server and stored. The server then automatically collects electricity usage data through the smart meter system and performs preprocessing to fill in or remove missing or outlier values. The server also obtains the latest real-time electricity market data and renewable energy supply data via the internet.

[0410] The server then uses this preprocessed data to analyze patterns of power consumption, for example by converting it into a data frame using Python's pandas library and generating histograms and time series graphs using the matplotlib library to identify periods of peak and low usage.

[0411] Furthermore, we will build a model to forecast electricity demand using machine learning techniques. Specifically, we will train ARIMA and LSTM models using machine learning libraries such as scikit-learn and TensorFlow, and evaluate the performance of the models using training and test datasets. Based on this trained model, it will be possible to forecast future electricity demand with high accuracy.

[0412] The server then uses multivariable linear programming and dynamic programming to run an optimization algorithm for power procurement based on the demand forecast model. This allows the creation of an optimal power supply plan that minimizes costs and environmental impact. Users can view this optimal supply plan in graph and chart format via a GUI. The server also notifies users of details via email and push notifications.

[0413] When the supply plan is actually executed, the device controls smart home appliances and HVAC (heating, ventilation, and air conditioning) systems in real time based on the plan. The server monitors power usage in real time and alerts the user if consumption does not progress as predicted and provides a new optimization plan.

[0414] As a concrete example, consider a case where a user signs in to the system and uploads their electricity usage data from the past year. This data is collected by a server, which also retrieves real-time electricity market data and renewable energy supply forecasts. Based on this, the server predicts electricity demand for the next month and creates and presents an optimal electricity supply plan. For example, one plan could reduce costs by 10% by increasing electricity usage during late-night hours. The device adjusts its electricity consumption based on this plan, and the server monitors the situation in real time. If a consumption pattern differs from the forecast, the server sends an alert to the user and presents a new plan if necessary.

[0415] Examples of input prompts for generative AI models include:

[0416] Design an electricity procurement system that helps businesses and households optimize their electricity usage efficiency and reduce costs and environmental impact. Use historical electricity usage data, real-time electricity market data, and renewable energy supply forecast data to build a model that forecasts electricity demand using machine learning algorithms such as ARIMA and LSTM models. Then, design a system that calculates and proposes the optimal procurement strategy that minimizes costs and environmental impact.

[0417] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0418] Processing steps of this system's program

[0419] Step 1: Sign in users and upload data

[0420] explanation

[0421] Users access the system's web portal or mobile app and sign in by entering their credentials on the login screen.

[0422] Users upload electricity usage data from the past year in a format such as a CSV file.

[0423] input

[0424] User authentication information (ID and password)

[0425] Power usage data (CSV file)

[0426] output

[0427] Authorization of credentials

[0428] Sending power usage data to a server

[0429] Specific actions

[0430] The user opens a web browser, accesses the system's URL, and enters their ID and password on the login screen.

[0431] After logging in, select "Upload Data" on the dashboard and upload the CSV file.

[0432] Step 2: Data collection and preprocessing by the server

[0433] explanation

[0434] The server receives the power usage data uploaded by the user and stores it in storage.

[0435] The server automatically acquires electricity usage data through the smart meter system and performs pre-processing to complement or remove missing or abnormal values.

[0436] The server collects real-time electricity market data and renewable energy supply data via the Internet.

[0437] input

[0438] User-uploaded electricity usage data

[0439] Electricity usage data from smart meter systems

[0440] Real-time electricity market and renewable energy supply data

[0441] output

[0442] Preprocessed dataset

[0443] Specific actions

[0444] The server parses the uploaded CSV file and stores it in a database.

[0445] The server retrieves data from the smart meter system via API and converts it into a data frame using pandas.

[0446] Missing values ​​are imputed with the mean and outliers are removed. The data is reordered by timestamp.

[0447] Step 3: Analyze data and identify consumption patterns

[0448] explanation

[0449] The server analyzes the pre-processed data and identifies patterns of power consumption through histogram and time series analysis.

[0450] input

[0451] Preprocessed dataset

[0452] output

[0453] Power consumption pattern analysis results

[0454] Specific actions

[0455] The server uses Python's matplotlib library to generate a histogram to see the distribution of consumption.

[0456] Through time series analysis, periods of peak and low usage are graphed.

[0457] Step 4: Building an electricity demand forecast model

[0458] explanation

[0459] The server builds a power demand forecasting model using machine learning algorithms such as the ARIMA model and the LSTM model.

[0460] The server splits the data into training and test datasets and trains and evaluates the model.

[0461] input

[0462] Preprocessed dataset

[0463] output

[0464] A trained electricity demand forecasting model

[0465] Specific actions

[0466] The server uses scikit-learn and TensorFlow to split the data into training and test sets.

[0467] Train the model and evaluate its predictive accuracy on the test set.

[0468] Step 5: Optimize power procurement

[0469] explanation

[0470] Based on the demand forecasting model, the server executes optimization algorithms using multivariable linear programming and dynamic programming to formulate an optimal power supply plan.

[0471] input

[0472] A trained electricity demand forecasting model

[0473] output

[0474] Optimal power supply plan

[0475] Specific actions

[0476] The server uses the SciPy library to perform linear programming and obtain the optimal solution.

[0477] Step 6: Present and notify users of the plan

[0478] explanation

[0479] The server provides and notifies the user of the optimal power supply plan in the form of graphs and charts.

[0480] input

[0481] Optimal power supply plan

[0482] output

[0483] Plan visibility and notification for users

[0484] Specific actions

[0485] The server displays graphs on a web page using D3.js or Chart.js and sends emails and push notifications.

[0486] Step 7: Real-time control and monitoring

[0487] explanation

[0488] The terminal controls smart home appliances and HVAC systems in real time based on the power supply plan received from the server.

[0489] The server monitors power usage in real time and sends an alert to the user if it detects an abnormality.

[0490] input

[0491] Optimal power supply plan

[0492] Real-time electricity usage data

[0493] output

[0494] Control commands for smart appliances and HVAC systems

[0495] Alert notification when an abnormality is detected

[0496] Specific actions

[0497] The terminal automatically controls air conditioning and lighting based on the power supply plan.

[0498] The server runs a real-time monitoring system and sends an email notification to the user when abnormal data is detected.

[0499] (Application example 1)

[0500] 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."

[0501] Reducing energy costs and environmental impacts are important issues for modern factories. However, conventional energy management systems have difficulty optimizing energy consumption in real time and are unable to efficiently manage the energy usage status of each piece of equipment. Therefore, effective means to improve the energy efficiency of the entire factory are needed.

[0502] 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.

[0503] In this invention, the server includes: means for acquiring historical energy usage data; means for acquiring real-time energy market data and renewable energy supply data; means for analyzing the historical energy usage data and the energy market data to identify energy consumption patterns; means for constructing a model for predicting future energy demand using a machine learning algorithm; means for formulating an optimal energy supply plan by executing an energy procurement optimization algorithm based on the prediction model; means for presenting the optimal energy supply plan to a user; means for each device to share energy consumption data; means for a central server to generate an optimal operation schedule using the energy consumption data; and means for monitoring energy usage in real time and modifying the energy supply plan. This enables each device in a factory to use energy efficiently, thereby reducing overall energy costs and environmental impact.

[0504] "Energy usage data" refers to information about energy consumed in the past.

[0505] "Energy Market Data" means real-time information on energy prices and supply conditions.

[0506] "Renewable energy supply data" refers to information on supply and forecasts from renewable energy sources such as solar, wind, and hydroelectric power.

[0507] A "machine learning algorithm" is a computational method for automatically learning patterns from data and making predictions and classifications.

[0508] A "predictive model" is a mathematical or computational model constructed to predict future trends based on past data.

[0509] An "energy procurement optimization algorithm" is a calculation method for optimizing the cost and efficiency of energy procurement.

[0510] An "energy supply plan" is a plan that determines the energy supply schedule and management policy.

[0511] "Means for each device to share energy consumption data" refers to a means for each device in a factory to communicate its own energy consumption status to each other.

[0512] "Means for a central server to generate an optimal operation schedule using energy consumption data" refers to a means for creating an optimal operation schedule for equipment based on energy consumption data collected by a central server within the factory.

[0513] "Means for monitoring energy usage in real time" refers to means for observing current energy usage in real time.

[0514] The "means for modifying an energy supply plan" refers to a means for appropriately modifying an existing energy supply plan in accordance with the actual energy usage situation.

[0515] This invention relates to a management system for improving the efficiency of energy use in factories and reducing costs and environmental impact. The system is composed of a central server and software installed on each piece of factory equipment.

[0516] First, the server acquires past energy usage data. Data is collected from smart meters and sensors connected to each device and sent to the server. Users can also upload their past energy usage data.

[0517] The server then retrieves real-time energy market data and renewable energy supply data via the internet, including energy prices and supply information, which is retrieved through internet services and APIs.

[0518] The server preprocesses the collected data by imputing or removing missing or outlier values ​​and reordering the data by timestamp. The server then analyzes the data and identifies patterns of energy consumption using histograms and time series analysis.

[0519] The server then uses a machine learning algorithm (e.g., an LSTM model) to predict future energy demand. The server divides the acquired data into a training dataset and a test dataset, and uses the training dataset to build and train a demand forecasting model.

[0520] Once the predictive model is complete, the server uses it to run an energy procurement optimization algorithm, specifically using multivariable linear programming and dynamic programming to develop an optimal energy supply plan that minimizes costs and environmental impacts.

[0521] The formulated energy supply plan is presented to the user, who can check the plan in graph and chart format on the management screen. Each factory device shares its energy consumption data, and the central server uses this data to generate an optimal operation schedule.

[0522] The server also monitors energy usage in real time, detecting anomalies and sending alerts to users if consumption patterns differ from those expected, providing them with the means to modify their energy supply plans as needed.

[0523] As a specific example, when a user signs in to the system and uploads energy usage data from the past year, the server collects real-time energy market data and renewable energy supply forecasts to predict energy demand for the next month. Based on this, an optimal energy supply plan is created, and the user receives a plan that allows them to reduce costs by 10% by increasing energy usage during late-night hours. Each piece of equipment in the factory adjusts its energy consumption based on this plan, and the server monitors usage in real time. If a consumption pattern that differs from the forecast occurs, the server sends an alert to the user and provides a new optimized plan.

[0524] Prompt Sentence Examples

[0525] Energy consumption data:

[0526] timestamp: 2023-10-01 00:00:00, consumption: 50

[0527] timestamp: 2023-10-01 01:00:00, consumption: 55

[0528] ...

[0529] timestamp: 2023-10-01 23:00:00, consumption: 58

[0530] Forecast your energy needs for the next month and provide an optimal energy supply plan. Include specific actions to reduce energy usage during peak hours.

[0531] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0532] Step 1:

[0533] The server collects past energy usage data. Data is sent to the server from smart meters and sensors connected to each device. The input here is the measurement data from each smart meter and sensor, and the output is an energy usage dataset that compiles this data. Specifically, the server periodically receives data from each device and stores it in a database.

[0534] Step 2:

[0535] The server obtains real-time energy market data and renewable energy supply data. This data includes energy prices and supply information and is obtained from an external API via the internet. The input is the response data from the API, and the output is the obtained market data and supply data. Specifically, the server calls the API at regular intervals to obtain the latest data.

[0536] Step 3:

[0537] The server preprocesses the collected data. Specifically, it complements or removes missing values ​​and outliers, and rearranges the data in timestamp order. The input is the data collected in steps 1 and 2, and the output is clean data that has been preprocessed. Specifically, the server automatically performs data cleansing and performs any necessary data transformations.

[0538] Step 4:

[0539] The server analyzes the pre-processed data to identify patterns in energy consumption. Histograms and time series analysis are used to identify periods of peak and low usage. The input is clean data, and the output is analysis results that show consumption patterns. Specifically, the server runs time series data analysis algorithms to extract specific patterns and trends.

[0540] Step 5:

[0541] The server uses a machine learning algorithm (e.g., an LSTM model) to build a model that predicts future energy demand. First, it divides the collected data into a training dataset and a test dataset, and then trains the model based on the training dataset. The input is the preprocessed data and the machine learning algorithm, and the output is a trained predictive model. Specifically, the server uses a machine learning framework (e.g., TensorFlow or Keras) to train the model.

[0542] Step 6:

[0543] The server uses the forecasting model to predict future energy demand. The input is a trained forecasting model and the latest dataset, and the output is the future demand forecast result. Specifically, the server inputs the latest data into the demand forecasting model and generates the forecast result.

[0544] Step 7:

[0545] The server executes an optimization algorithm for energy procurement and formulates an optimal energy supply plan. It minimizes costs and environmental impacts using multivariable linear programming and dynamic programming. The inputs are the demand forecast results and the optimization algorithm, and the output is an optimal energy supply plan. Specifically, the server executes optimization calculations and generates the optimal plan within the specified constraints.

[0546] Step 8:

[0547] The server presents the formulated energy supply plan to the user. The user can check the plan in graph and chart format through the management screen. The input is the optimal energy supply plan, and the output is the supply plan displayed to the user. Specifically, the server visualizes the plan and displays it on a dashboard.

[0548] Step 9:

[0549] Each piece of factory equipment shares its energy consumption data, and a central server uses this data to generate an optimal operation schedule. The input is the energy consumption data from each piece of equipment, and the output is the optimal operation schedule. Specifically, the server periodically collects data from each piece of equipment and calculates the optimal schedule.

[0550] Step 10:

[0551] The server monitors energy usage in real time, alerts users if anomalies occur, and modifies energy supply plans as needed. The input is real-time energy usage data, and the output is modified energy supply plans and alerts. Specifically, the server tracks consumption data through the monitoring system and takes action when anomalies are detected.

[0552] 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.

[0553] The present invention is an electricity procurement system for businesses and households to optimize the efficiency of electricity usage and reduce costs and environmental burdens. The system further incorporates an emotion engine to recognize user emotions and improve the effectiveness of supply plans. The system of the present invention will be described in detail below.

[0554] First, the user accesses the server and signs in to the system. Once the user uploads their past electricity usage data, the server begins collecting and preprocessing the data. The server obtains the user's electricity usage data from the past year through the smart meter system. At the same time, the server collects electricity market data and renewable energy supply data via the Internet. This data includes real-time electricity price information and renewable energy supply forecasts.

[0555] The server then preprocesses and cleanses the data, imputing or removing missing or outlier values ​​and reordering them by timestamp. The server then analyzes the data to identify patterns in power consumption, for example, using histograms and time series analysis to identify periods of peak and low usage.

[0556] Next, the server uses a machine learning algorithm (e.g., ARIMA model or LSTM model) to build a model that predicts future electricity demand. The server divides the acquired data into a training dataset and a test dataset, and uses the training dataset to train the demand forecasting model. The accuracy of the forecasting model is evaluated to confirm its performance.

[0557] The server runs an optimization algorithm for power procurement based on the demand forecast model to develop an optimal power supply plan. Specifically, it uses multivariable linear programming and dynamic programming to calculate the optimal procurement strategy that minimizes costs and environmental impacts. The server then presents this optimal plan to the user. The user can view the power supply plan in graph and chart format through the server's management screen. The server also sends notifications to the user and provides details of the plan.

[0558] This is where the emotion engine comes in. The emotion engine acquires the user's emotion data, analyzes it, and evaluates the effectiveness of the power supply plan. For example, if the user is feeling stressed, it can use that information to suggest an energy consumption plan that will help reduce stress. The emotion engine collects emotion data using voice and image analysis.

[0559] The server monitors power usage in real time and adjusts the plan as needed. For example, if power consumption is not as expected, the server detects the anomaly and alerts the user. The server also updates the predictive model based on real-time data and runs new optimization algorithms.

[0560] As a concrete example, consider the case where a user signs in to the system and uploads electricity usage data from the past year. The server collects real-time electricity market data and renewable energy supply forecasts. Based on this, the server predicts electricity demand for the next month and creates and presents an optimal electricity supply plan. The user reviews the plan, and the emotion engine recognizes the user's stress level, and the server suggests a plan that will reduce stress. The device adjusts electricity consumption based on this plan, and the server monitors usage in real time. If a consumption pattern that differs from the forecast occurs, the server sends an alert to the user and provides a new optimized plan if necessary.

[0561] The processing flow will be explained below.

[0562] Step 1:

[0563] A user signs in to the server and uploads their historical electricity usage data. The server retrieves the past year's electricity usage data from the smart meter system, and also collects electricity market data and renewable energy supply data via the Internet.

[0564] Step 2:

[0565] The server cleanses and preprocesses the acquired data. Specifically, the server detects, imputes, or removes missing or outlier values. It also sorts the data by timestamp and generates daily and monthly aggregated data.

[0566] Step 3:

[0567] The server analyzes the pre-processed data to identify patterns of power consumption, using histograms and time series analysis to identify periods of peak and low usage, and then performs cluster analysis to create distinct consumption pattern groups.

[0568] Step 4:

[0569] The server uses a machine learning algorithm to build a future electricity demand forecasting model. The server divides the data into a training dataset and a test dataset, and trains an ARIMA model or LSTM model using the training dataset. The model's predictive accuracy is then evaluated using the test dataset.

[0570] Step 5:

[0571] The server runs an optimization algorithm for power procurement based on the trained power demand forecasting model. Specifically, the server uses multivariable linear programming and dynamic programming to formulate an optimal power supply plan that minimizes costs and environmental impacts.

[0572] Step 6:

[0573] The server presents the optimized power supply plan to the user, who can then view the plan on the server's management screen and receive detailed information through notifications. The plan is visually displayed in graphs and charts.

[0574] Step 7:

[0575] The emotion engine works to obtain the user's emotion data. The emotion engine uses voice and image analysis to read the user's emotion in real time. For example, it analyzes the user's facial expressions and tone of voice through a camera or microphone.

[0576] Step 8:

[0577] The server analyzes the acquired emotional data and evaluates the user's current emotional state. For example, if the user is feeling stressed, the server adjusts the power supply plan based on that information.

[0578] Step 9:

[0579] Based on the data provided by the emotion engine, the server re-optimizes the power supply plan. A new power supply plan with a stress-reducing effect is formulated and presented to the user. The server then visually displays the plan again and sends a notification to the user.

[0580] Step 10:

[0581] The device adjusts power consumption based on the new power supply plan set by the device, controls smart home appliances and HVAC systems, and increases or decreases power consumption during specified times. The device transmits power usage data to the server in real time.

[0582] Step 11:

[0583] The server monitors power usage in real time and adjusts the plan as needed. If it detects abnormal consumption patterns or unexpected conditions, it will alert the user and provide a new optimized plan accordingly.

[0584] Example 2

[0585] 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."

[0586] The problem that this invention aims to solve is to optimize the efficiency of power usage in businesses and homes and reduce costs and environmental burdens. Conventional power supply systems do not propose power supply plans that take user emotions into consideration, making it difficult to improve user satisfaction. Another problem is the lack of real-time monitoring of power usage and rapid response after anomaly detection.

[0587] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for acquiring historical power usage data; means for acquiring real-time power market data and renewable energy supply data; means for analyzing the historical power usage data and the power market data to identify power consumption patterns; means for constructing a model for predicting future power demand using a machine learning algorithm; means for formulating an optimal power supply plan by executing a power procurement optimization algorithm based on the prediction model; means for presenting the optimal power supply plan to a user; means for acquiring and analyzing user emotion data to evaluate the effectiveness of the power supply plan; and means for monitoring power usage in real time and modifying the power supply plan. This makes it possible to reduce costs and environmental burdens by optimizing power usage efficiency, improve user satisfaction, and achieve quick responses in real time.

[0588] "Past power usage data" refers to data relating to the amount of power consumed by the user in the past and the time of day.

[0589] "Real-time electricity market data" means information about the current electricity market price and supply.

[0590] "Renewable energy supply data" is information about electricity supplied from renewable energy sources such as wind, solar, and hydroelectric power.

[0591] The "means for identifying patterns of power consumption" is a means for analyzing power usage data and identifying trends in power consumption and peak times during a specific period.

[0592] A "machine learning algorithm" is a method or technology for predicting future data by analyzing and learning from data.

[0593] A "model for predicting future electricity demand" is a mathematical or statistical model for estimating future electricity consumption based on collected data.

[0594] The "power procurement optimization algorithm" is a calculation method for efficiently procuring electricity while taking into account factors such as cost and environmental impact.

[0595] An "optimal power supply plan" is a power supply strategy that meets users' power demand while minimizing costs and environmental burdens.

[0596] "User emotional data" is information about the user's psychological state and emotions, and is data collected through voice and image analysis.

[0597] "Means for monitoring power usage in real time" refers to means for constantly monitoring users' power usage and responding immediately if any abnormalities are detected.

[0598] The present invention is an electricity procurement system for optimizing the efficiency of electricity usage in businesses and homes, reducing costs and environmental burdens, and further incorporates an emotion engine for recognizing user emotions to improve the effectiveness of supply plans.

[0599] Overall system overview

[0600] First, the user accesses the server and signs in to the system. Once the user uploads their past electricity usage data, the server begins collecting and preprocessing the data. The server obtains the user's electricity usage data from the past year through the smart meter system. At the same time, the server collects electricity market data and renewable energy supply data via the Internet. This data includes real-time electricity price information and renewable energy supply forecasts.

[0601] Hardware and Software Use

[0602] The server preprocesses the uploaded data using Python's Pandas library and NumPy. Specifically, it imputes missing values ​​and removes outliers. It then uses data visualization tools such as Matplotlib and Seaborn to identify patterns in power consumption through histograms and time series analysis.

[0603] Next, the server builds a machine learning model using Scikit-learn and TensorFlow. It uses ARIMA and LSTM models to predict future electricity demand. In this process, the collected data is divided into a training dataset and a test dataset, and the model is trained.

[0604] To develop an optimal power supply plan, optimization libraries such as Gurobi and PuLP are used. The server uses multivariable linear programming and dynamic programming to calculate the optimal procurement strategy that minimizes costs and environmental impacts. The plan is presented to the user in the form of graphs and charts using the Plotly library.

[0605] Furthermore, the emotion engine collects and analyzes the user's emotional data through voice and image analysis. For example, it uses emotion analysis tools such as AWS Rekognition and Watson. If the user is feeling stressed, it can use that information to suggest a stress-relieving plan.

[0606] Adding specific examples

[0607] As a concrete example, consider the case where a user signs in to the system and uploads electricity usage data from the past year. The server collects real-time electricity market data and renewable energy supply forecasts. Based on this, the server predicts electricity demand for the next month and creates and presents an optimal electricity supply plan. The user reviews the plan, and the emotion engine recognizes the user's stress level, and the server suggests a plan that will reduce stress. The device adjusts electricity consumption based on this plan, and the server monitors usage in real time. If a consumption pattern that differs from the forecast occurs, the server sends an alert to the user and provides a new optimized plan if necessary.

[0608] Prompt Sentence Examples

[0609] An example of an input prompt for a generative AI model is as follows:

[0610] "Please tell me the detailed steps to implement a system where a user can upload electricity usage data from the past year and the system will suggest the optimal electricity supply plan. Please also provide a concrete coding example."

[0611] By implementing this system, it is possible to reduce costs and environmental burdens by optimizing power usage efficiency, improve user satisfaction, and realize quick responses in real time.

[0612] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0613] Step 1:

[0614] A user accesses the system's web application and signs in by entering their username and password. The input is the username and password, and the output is an authentication token. The server checks the user authentication information in the database and generates an authentication token if authentication is successful. Specifically, the user opens a browser, enters a URL to access the system's login page, enters their username and password in the form, and clicks the "Login" button.

[0615] Step 2:

[0616] Users upload the electricity usage data from the past year to the system from their own devices. The input is a file of past electricity usage data, and the output is the electricity usage data stored in the server's database. The server receives this data through the smart meter system and stores it in the database. Specifically, the user clicks the "Upload data" button, selects the data file, and uploads it.

[0617] Step 3:

[0618] The server collects electricity market data and renewable energy supply data via the Internet. The input is electricity market data and renewable energy data from APIs and databases, and the output is the server's database where this data is stored. Specifically, the server periodically calls external APIs to collect electricity market data and renewable energy data.

[0619] Step 4:

[0620] The server performs preprocessing on the collected data. The input is the uploaded electricity usage data, electricity market data, and renewable energy data, and the output is the cleansed data. Specifically, it uses Python's Pandas library and NumPy to fill in missing values ​​and remove outliers. Specifically, the server fills in missing values ​​using Pandas' DataFrame.fillna() method and removes outliers using the DataFrame.drop() method.

[0621] Step 5:

[0622] The server analyzes the cleansed data to identify patterns in power consumption. The input is the cleansed power usage data, and the output is information about power consumption patterns. Histograms and time series analysis are used to identify periods of peak and low usage. Specifically, Matplotlib and Seaborn are used to create histograms and visualize consumption patterns.

[0623] Step 6:

[0624] The server uses a machine learning algorithm to build an electricity demand forecasting model. The input is the cleansed dataset, and the output is the demand forecasting model. Scikit-learn and TensorFlow are used to train ARIMA and LSTM models. Specifically, the dataset is split into training data and test data, and the data is separated using Scikit-learn's train_test_split() method.

[0625] Step 7:

[0626] The server runs an optimization algorithm for power procurement based on the forecast model. The input is a demand forecast model and electricity market data, and the output is an optimal power supply plan. It uses optimization libraries such as Gurobi and PuLP to perform multivariable linear programming and dynamic programming. Specifically, it sets up an optimization problem in Gurobi and calculates the optimal power supply plan.

[0627] Step 8:

[0628] The server presents the optimal power supply plan to the user. The input is the optimal power supply plan, and the output is a plan in the form of graphs and charts that the user can view. Using the Plotly library, the optimal plan is visualized and displayed on a web page. Specifically, the server creates a graph using Plotly and displays it on an admin screen that the user can access.

[0629] Step 9:

[0630] The emotion engine acquires and analyzes the user's emotional data. The input is voice and image data, and the output is analyzed emotional data. Emotions are analyzed using AWS Rekognition or Watson, and the server evaluates the effectiveness of the plan. Specifically, the camera acquires the user's facial expression data, which is then sent to the emotion analysis tool to obtain the analysis results.

[0631] Step 10:

[0632] The server monitors power usage in real time and sends an alert to the user if it detects an abnormality. The input is real-time power usage data and the output is an alert message. Monitoring and alerts are implemented using Prometheus and Grafana. Specifically, the server collects real-time data and sends a notification to the user if the data exceeds a specified threshold.

[0633] (Application example 2)

[0634] 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."

[0635] Conventional power procurement systems formulate power supply plans based on past power usage data and market data, but they have limitations in terms of real-time adjustments to supply plans, particularly optimization that takes user sentiment data into account. Furthermore, optimizing power usage is difficult, particularly in places with large-scale power consumption such as factories, and efficient energy management is required. This presents a challenge in that it is not possible to fully reduce power costs and environmental impact.

[0636] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0637] In this invention, the server includes: means for acquiring historical power usage data; means for acquiring real-time power market data and renewable energy supply data; means for analyzing the historical power usage data and the power market data to identify power consumption patterns; means for constructing a model for predicting future power demand using a machine learning algorithm; means for formulating an optimal power supply plan by executing a power procurement optimization algorithm based on the prediction model; means for presenting the optimal power supply plan to a user; means for monitoring power usage in real time and revising the power supply plan; means for acquiring and analyzing user emotion data; means for revising the power supply plan based on the emotion data; and means for optimizing power usage of factory robots using a smartphone application. This enables real-time optimization of power usage and efficient energy management in a factory while taking user emotion data into consideration.

[0638] "Past power usage data" is history data relating to the amount of power consumed by the user in the past.

[0639] "Real-time electricity market data" means instantaneous data showing the current market electricity prices and supply situation.

[0640] "Renewable energy supply data" refers to data on electricity supply from renewable energy sources such as solar, wind, and hydroelectric power.

[0641] The "power consumption pattern" is an analysis result that indicates the user's power consumption tendency by time period, peak time periods, and low consumption time periods.

[0642] A "machine learning algorithm" is a computer science technique for learning patterns and trends from large amounts of data and making predictions.

[0643] A "model for predicting electricity demand" is a mathematical model constructed to predict future electricity consumption from past data.

[0644] An "optimization algorithm for power procurement" is a mathematical method for calculating the optimal power supply method to minimize power costs and environmental impact.

[0645] An "electricity supply plan" is a plan that determines how electricity will be procured and supplied for future electricity usage.

[0646] The "means for presenting to the user" refers to a method for providing information such as a power supply plan to the user visually or as a notification.

[0647] "Means for monitoring power usage in real time" refers to a system for instantly monitoring current power consumption.

[0648] "User emotion data" is data that indicates the user's emotional state, and is obtained by voice or image analysis, for example.

[0649] A "smartphone application" is software that runs on a smartphone and provides specific functions or services.

[0650] "Means for optimizing power usage of factory robots" is a system for efficiently managing and minimizing power usage of robots operating in factories.

[0651] The present invention is a system for optimizing the efficiency of power usage in businesses and households to reduce costs and environmental burdens, and in particular, uses a smartphone application to improve the efficiency of power usage in factory robots. This system optimizes power supply plans taking into account user emotional data. Specific embodiments of the present invention are described in detail below.

[0652] The system includes a server, a smartphone application, and an emotion data acquisition unit. The server acquires historical electricity usage data and collects real-time electricity market data and renewable energy supply data. It also preprocesses and cleanses this data to identify electricity consumption patterns. This data is then used to build a model for predicting future electricity demand using a machine learning algorithm, and an optimal electricity supply plan is developed.

[0653] Specific software and hardware used in this embodiment include the requests library for data collection, the pandas library for data preprocessing, and the scikit-learn library for machine learning algorithms. Also, EmotionRecognizer is used as software for acquiring user emotion data.

[0654] For example, a user (factory manager) signs in to a smartphone application and uploads electricity usage data from the past year. The server collects real-time electricity market data and renewable energy supply data and analyzes electricity consumption patterns. The server then uses machine learning algorithms to predict future electricity demand and create an optimal electricity supply plan, which it then presents to the user. The server also obtains the user's emotional data and modifies the plan according to their emotional state. The smartphone application also optimizes the power usage of factory robots.

[0655] For example, after a factory manager signs in to the system and uploads data, the server collects electricity market data in real time and executes a process to forecast the next month's electricity demand. During this process, emotion recognition software detects the manager's stress level and presents a power supply plan that will reduce stress. An example of a specific prompt could be to instruct the server, "Please predict electricity usage based on the following data."

[0656] In this way, the system optimizes power usage in factories in real time, enabling efficient energy management.

[0657] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0658] Step 1:

[0659] The server acquires past electricity usage data. When a user uploads electricity usage data from the past year via a smartphone application, the data is sent to the server. The server collects the data using an API endpoint and stores it in a database. The input is the past electricity usage data, and the output is the stored electricity usage data.

[0660] Step 2:

[0661] The server obtains real-time electricity market data and renewable energy supply data. It accesses APIs via the internet to collect current market data and renewable energy supply forecasts. The collected data is stored in the server's database. The input is the real-time electricity market data and renewable energy supply data, and the output is the stored market data and supply data.

[0662] Step 3:

[0663] The server preprocesses the acquired historical electricity usage data and market data to identify electricity consumption patterns. Specifically, it cleanses the data and fills or removes missing values ​​and outliers. Using the cleansed data, it identifies peak and low electricity usage periods through histogram and time series analysis. The input is historical electricity usage data and market data, and the output is electricity consumption patterns.

[0664] Step 4:

[0665] The server uses a machine learning algorithm to build a model to predict future electricity demand. Specifically, it divides the data into a training dataset and a test dataset, and trains a linear regression model using, for example, the scikit-learn library. The input is preprocessed electricity usage data and market data, and the output is the trained demand forecasting model.

[0666] Step 5:

[0667] The server executes an optimization algorithm for power procurement based on the trained forecasting model to formulate an optimal power supply plan. For example, it applies multivariable linear programming based on the predicted power demand to minimize costs and environmental impact. The input is the trained demand forecasting model, and the output is the optimal power supply plan.

[0668] Step 6:

[0669] The server presents the optimal power supply plan to the user, who can then check the plan through a smartphone application. The plan is visually displayed in graphs and charts, and detailed information is also provided. The input is the optimal power supply plan, and the output is the plan presented to the user.

[0670] Step 7:

[0671] The server monitors power usage in real time and modifies the power supply plan. It acquires data sequentially, compares the predictive model with the actual data, detects anomalies, and modifies the plan as necessary. The input is real-time power usage data, and the output is the modified power supply plan.

[0672] Step 8:

[0673] The server acquires and analyzes the user's emotional data. It collects audio and images from smartphones and other devices and analyzes the user's emotional state using EmotionRecognizer software. The input is the user's emotional data, and the output is the analysis results.

[0674] Step 9:

[0675] The server modifies the power supply plan based on the emotion data. For example, if the user is feeling stressed, it generates a plan to reduce power consumption. The input is the analyzed emotion data, and the output is the modified power supply plan.

[0676] Step 10:

[0677] The server uses a smartphone application to optimize the power usage of factory robots. Users can check the robot's operating status and power consumption through the app and operate it according to the optimized plan. The input is the factory robot's power usage data, and the output is the optimized power usage.

[0678] 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.

[0679] 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.

[0680] 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.

[0681] [Third embodiment]

[0682] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0683] 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.

[0684] 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).

[0685] 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.

[0686] 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.

[0687] 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).

[0688] 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.

[0689] 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.

[0690] 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.

[0691] 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.

[0692] 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.

[0693] 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."

[0694] The present invention relates to an electricity procurement system that enables businesses and households to optimize the efficiency of their electricity usage and reduce costs and environmental impact. The system of the present invention will be described in detail below.

[0695] First, the user accesses the server and signs in to the system. Once the user uploads their past electricity usage data, the server begins collecting and preprocessing the data. The server obtains the user's electricity usage data from the past year through the smart meter system. At the same time, the server collects electricity market data and renewable energy supply data via the Internet. This data includes real-time electricity price information and renewable energy supply forecasts.

[0696] The server then preprocesses and cleanses the data, imputing or removing missing or outlier values ​​and reordering them by timestamp. The server then analyzes the data to identify patterns in power consumption, for example, using histograms and time series analysis to identify periods of peak and low usage.

[0697] Next, the server uses a machine learning algorithm (e.g., ARIMA model or LSTM model) to build a model that predicts future electricity demand. The server divides the acquired data into a training dataset and a test dataset, and uses the training dataset to train the demand forecasting model. The accuracy of the forecasting model is evaluated to confirm its performance.

[0698] The server runs an optimization algorithm for power procurement based on the demand forecast model to develop an optimal power supply plan. Specifically, it uses multivariable linear programming and dynamic programming to calculate the optimal procurement strategy that minimizes costs and environmental impact. The server then presents this optimal plan to the user. The user can view the power supply plan in graph and chart format through the server's management screen. The server also sends notifications to the user and provides details of the plan.

[0699] The server monitors power usage in real time and adjusts the plan as needed. For example, if power consumption is not as expected, the server detects the anomaly and alerts the user. The device then controls smart home appliances and HVAC systems in real time based on the configured power supply plan.

[0700] As a concrete example, consider the case where a user signs in to the system and uploads electricity usage data from the past year. The server collects real-time electricity market data and renewable energy supply forecasts. Based on this, the server predicts electricity demand for the next month and creates and presents an optimal electricity supply plan. The user receives a plan that allows them to reduce costs by 10% by increasing electricity usage during late-night hours. The device adjusts electricity consumption based on this plan, and the server monitors usage in real time. If a consumption pattern that differs from the forecast occurs, the server sends an alert to the user and provides a new optimized plan if necessary.

[0701] The processing flow will be explained below.

[0702] Step 1:

[0703] The user signs in to the server and uploads their historical electricity usage data. The server is then accessed and retrieves the user's electricity usage data from the smart meter system. The server also collects electricity market data and renewable energy supply data via the internet, allowing the system to obtain all the necessary data.

[0704] Step 2:

[0705] The server cleanses and preprocesses the acquired data. Specifically, the server detects missing values ​​and outliers and completes or removes them. The server also sorts the data by timestamp and generates daily and monthly aggregated data. This preprocessing improves the quality of the data.

[0706] Step 3:

[0707] The server analyzes the pre-processed data to identify patterns in power consumption. It uses histograms and time series analysis to identify peak and low usage periods. It also performs cluster analysis to create groups of different consumption patterns. This analysis identifies the characteristics and trends of power usage.

[0708] Step 4:

[0709] The server uses a machine learning algorithm to build a future electricity demand forecasting model. The server divides the data into a training dataset and a test dataset, and uses the training dataset to train a forecasting model (e.g., an ARIMA model or an LSTM model). The model is then evaluated on the test dataset to confirm its forecast accuracy.

[0710] Step 5:

[0711] The server runs an optimization algorithm for power procurement based on a trained power demand forecasting model. Using multivariable linear programming and dynamic programming, the server formulates an optimal power supply plan that minimizes costs and environmental impacts. This plan is created based on future power demand forecasts.

[0712] Step 6:

[0713] The server presents the optimized power supply plan to the user. The server then visualizes the plan and displays it in the form of graphs and charts on the user's management screen. The server also sends detailed information about the plan to the user's smartphone or email, allowing the user to review and select the proposed plan.

[0714] Step 7:

[0715] The device adjusts power consumption based on the set power supply plan. The device controls smart home appliances and HVAC systems, increasing or decreasing power consumption during designated time periods. The device also transmits real-time power usage data to a server.

[0716] Step 8:

[0717] The server monitors power usage in real time and adjusts plans as needed. The server detects unusual consumption patterns or unexpected situations and alerts users. The server also updates predictive models and runs new optimization algorithms based on real-time data.

[0718] Example 1

[0719] 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."

[0720] In modern society, optimizing the efficiency of power usage in businesses and homes and reducing power costs and environmental impact are major challenges. However, conventional systems do not fully automate the collection and analysis of power usage data, resulting in insufficient data to formulate optimal power supply plans. Furthermore, demand forecasting using machine learning algorithms and real-time monitoring and feedback of power usage status are insufficient, making it difficult to optimize power supply.

[0721] 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.

[0722] In this invention, the server includes means for acquiring historical power usage data, means for acquiring real-time power market data and renewable energy supply data, means for analyzing the historical power usage data and the power market data to identify power consumption patterns, means for constructing a model for predicting future power demand using a machine learning algorithm, means for formulating an optimal power supply plan by executing a power procurement optimization algorithm based on the prediction model, means for presenting the optimal power supply plan to a user, and means for monitoring power usage in real time and modifying the power supply plan. This enables automatic collection and analysis of power usage data, construction and verification of a demand forecast model, and real-time presentation and modification of an optimal power supply plan.

[0723] "Past power usage data" is information relating to the amount of power consumed in the past and the timing of its use.

[0724] "Real-time electricity market data" means up-to-date information about current prices and supply conditions in the electricity market.

[0725] "Renewable energy supply data" is information about the amount and forecast of electricity supplied from renewable energy sources, such as solar, wind, and hydroelectric power.

[0726] "Patterns of electricity consumption" are data that indicate trends and characteristics of electricity usage over a specific period of time.

[0727] A "machine learning algorithm" is a computational method for automatically building predictive and classification models using data.

[0728] A "model for predicting future electricity demand" is a model used to predict future electricity consumption from past data and current conditions.

[0729] An "optimal power supply plan" is a power supply plan designed to maximize the efficiency of power use and minimize costs and environmental impacts.

[0730] A "smart meter system" is a measuring device and communication system that measures electricity consumption in real time and collects and transmits that data remotely.

[0731] A "missing value" refers to a value that is missing in a dataset.

[0732] An "outlier" refers to a value in a data set that falls outside the normal range.

[0733] A "training dataset" is a collection of data used to train a machine learning model.

[0734] A "test dataset" is a collection of data used to evaluate the performance of a trained machine learning model.

[0735] "Real-time monitoring of power usage" refers to the process of monitoring power consumption in real time and taking immediate action if necessary.

[0736] A "terminal" is a device for controlling appliances based on a power supply plan.

[0737] The present invention relates to an electricity procurement system that allows businesses and households to optimize the efficiency of their electricity usage and reduce costs and environmental impacts. The system is comprised of a combination of software and hardware that collects, analyzes, and optimizes historical electricity usage data, real-time electricity market data, and renewable energy supply data.

[0738] First, users access the system's web portal or mobile app and sign in to the system by entering their authentication information on the login screen. They then upload their electricity usage data from the past year in a format such as a CSV file. This data is received by the server and stored. The server then automatically collects electricity usage data through the smart meter system and performs preprocessing to fill in or remove missing or outlier values. The server also obtains the latest real-time electricity market data and renewable energy supply data via the internet.

[0739] The server then uses this preprocessed data to analyze patterns of power consumption, for example by converting it into a data frame using Python's pandas library and generating histograms and time series graphs using the matplotlib library to identify periods of peak and low usage.

[0740] Furthermore, we will build a model to forecast electricity demand using machine learning techniques. Specifically, we will train ARIMA and LSTM models using machine learning libraries such as scikit-learn and TensorFlow, and evaluate the performance of the models using training and test datasets. Based on this trained model, it will be possible to forecast future electricity demand with high accuracy.

[0741] The server then uses multivariable linear programming and dynamic programming to run an optimization algorithm for power procurement based on the demand forecast model. This allows the creation of an optimal power supply plan that minimizes costs and environmental impact. Users can view this optimal supply plan in graph and chart format via a GUI. The server also notifies users of details via email and push notifications.

[0742] When the supply plan is actually executed, the device controls smart home appliances and HVAC (heating, ventilation, and air conditioning) systems in real time based on the plan. The server monitors power usage in real time and alerts the user if consumption does not progress as predicted and provides a new optimization plan.

[0743] As a concrete example, consider a case where a user signs in to the system and uploads their electricity usage data from the past year. This data is collected by a server, which also retrieves real-time electricity market data and renewable energy supply forecasts. Based on this, the server predicts electricity demand for the next month and creates and presents an optimal electricity supply plan. For example, one plan could reduce costs by 10% by increasing electricity usage during late-night hours. The device adjusts its electricity consumption based on this plan, and the server monitors the situation in real time. If a consumption pattern differs from the forecast, the server sends an alert to the user and presents a new plan if necessary.

[0744] Examples of input prompts for generative AI models include:

[0745] Design an electricity procurement system that helps businesses and households optimize their electricity usage efficiency and reduce costs and environmental impact. Use historical electricity usage data, real-time electricity market data, and renewable energy supply forecast data to build a model that forecasts electricity demand using machine learning algorithms such as ARIMA and LSTM models. Then, design a system that calculates and proposes the optimal procurement strategy that minimizes costs and environmental impact.

[0746] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0747] Processing steps of this system's program

[0748] Step 1: Sign in users and upload data

[0749] explanation

[0750] Users access the system's web portal or mobile app and sign in by entering their credentials on the login screen.

[0751] Users upload electricity usage data from the past year in a format such as a CSV file.

[0752] input

[0753] User authentication information (ID and password)

[0754] Power usage data (CSV file)

[0755] output

[0756] Authorization of credentials

[0757] Sending power usage data to a server

[0758] Specific actions

[0759] The user opens a web browser, accesses the system's URL, and enters their ID and password on the login screen.

[0760] After logging in, select "Upload Data" on the dashboard and upload the CSV file.

[0761] Step 2: Data collection and preprocessing by the server

[0762] explanation

[0763] The server receives the power usage data uploaded by the user and stores it in storage.

[0764] The server automatically acquires electricity usage data through the smart meter system and performs pre-processing to complement or remove missing or abnormal values.

[0765] The server collects real-time electricity market data and renewable energy supply data via the Internet.

[0766] input

[0767] User-uploaded electricity usage data

[0768] Electricity usage data from smart meter systems

[0769] Real-time electricity market and renewable energy supply data

[0770] output

[0771] Preprocessed dataset

[0772] Specific actions

[0773] The server parses the uploaded CSV file and stores it in a database.

[0774] The server retrieves data from the smart meter system via API and converts it into a data frame using pandas.

[0775] Missing values ​​are imputed with the mean and outliers are removed. The data is reordered by timestamp.

[0776] Step 3: Analyze data and identify consumption patterns

[0777] explanation

[0778] The server analyzes the pre-processed data and identifies patterns of power consumption through histogram and time series analysis.

[0779] input

[0780] Preprocessed dataset

[0781] output

[0782] Power consumption pattern analysis results

[0783] Specific actions

[0784] The server uses Python's matplotlib library to generate a histogram to see the distribution of consumption.

[0785] Through time series analysis, periods of peak and low usage are graphed.

[0786] Step 4: Building an electricity demand forecast model

[0787] explanation

[0788] The server builds a power demand forecasting model using machine learning algorithms such as the ARIMA model and the LSTM model.

[0789] The server splits the data into training and test datasets and trains and evaluates the model.

[0790] input

[0791] Preprocessed dataset

[0792] output

[0793] A trained electricity demand forecasting model

[0794] Specific actions

[0795] The server uses scikit-learn and TensorFlow to split the data into training and test sets.

[0796] Train the model and evaluate its predictive accuracy on the test set.

[0797] Step 5: Optimize power procurement

[0798] explanation

[0799] Based on the demand forecasting model, the server executes optimization algorithms using multivariable linear programming and dynamic programming to formulate an optimal power supply plan.

[0800] input

[0801] A trained electricity demand forecasting model

[0802] output

[0803] Optimal power supply plan

[0804] Specific actions

[0805] The server uses the SciPy library to perform linear programming and obtain the optimal solution.

[0806] Step 6: Present and notify users of the plan

[0807] explanation

[0808] The server provides and notifies the user of the optimal power supply plan in the form of graphs and charts.

[0809] input

[0810] Optimal power supply plan

[0811] output

[0812] Plan visibility and notification for users

[0813] Specific actions

[0814] The server displays graphs on a web page using D3.js or Chart.js and sends emails and push notifications.

[0815] Step 7: Real-time control and monitoring

[0816] explanation

[0817] The terminal controls smart home appliances and HVAC systems in real time based on the power supply plan received from the server.

[0818] The server monitors power usage in real time and sends an alert to the user if it detects an abnormality.

[0819] input

[0820] Optimal power supply plan

[0821] Real-time electricity usage data

[0822] output

[0823] Control commands for smart appliances and HVAC systems

[0824] Alert notification when an abnormality is detected

[0825] Specific actions

[0826] The terminal automatically controls air conditioning and lighting based on the power supply plan.

[0827] The server runs a real-time monitoring system and sends an email notification to the user when abnormal data is detected.

[0828] (Application example 1)

[0829] 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."

[0830] Reducing energy costs and environmental impacts are important issues for modern factories. However, conventional energy management systems have difficulty optimizing energy consumption in real time and are unable to efficiently manage the energy usage status of each piece of equipment. Therefore, effective means to improve the energy efficiency of the entire factory are needed.

[0831] 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.

[0832] In this invention, the server includes: means for acquiring historical energy usage data; means for acquiring real-time energy market data and renewable energy supply data; means for analyzing the historical energy usage data and the energy market data to identify energy consumption patterns; means for constructing a model for predicting future energy demand using a machine learning algorithm; means for formulating an optimal energy supply plan by executing an energy procurement optimization algorithm based on the prediction model; means for presenting the optimal energy supply plan to a user; means for each device to share energy consumption data; means for a central server to generate an optimal operation schedule using the energy consumption data; and means for monitoring energy usage in real time and modifying the energy supply plan. This enables each device in a factory to use energy efficiently, thereby reducing overall energy costs and environmental impact.

[0833] "Energy usage data" refers to information about energy consumed in the past.

[0834] "Energy Market Data" means real-time information on energy prices and supply conditions.

[0835] "Renewable energy supply data" refers to information on supply and forecasts from renewable energy sources such as solar, wind, and hydroelectric power.

[0836] A "machine learning algorithm" is a computational method for automatically learning patterns from data and making predictions and classifications.

[0837] A "predictive model" is a mathematical or computational model constructed to predict future trends based on past data.

[0838] An "energy procurement optimization algorithm" is a calculation method for optimizing the cost and efficiency of energy procurement.

[0839] An "energy supply plan" is a plan that determines the energy supply schedule and management policy.

[0840] "Means for each device to share energy consumption data" refers to a means for each device in a factory to communicate its own energy consumption status to each other.

[0841] "Means for a central server to generate an optimal operation schedule using energy consumption data" refers to a means for creating an optimal operation schedule for equipment based on energy consumption data collected by a central server within the factory.

[0842] "Means for monitoring energy usage in real time" refers to means for observing current energy usage in real time.

[0843] The "means for modifying an energy supply plan" refers to a means for appropriately modifying an existing energy supply plan in accordance with the actual energy usage situation.

[0844] This invention relates to a management system for improving the efficiency of energy use in factories and reducing costs and environmental impact. The system is composed of a central server and software installed on each piece of factory equipment.

[0845] First, the server acquires past energy usage data. Data is collected from smart meters and sensors connected to each device and sent to the server. Users can also upload their past energy usage data.

[0846] The server then retrieves real-time energy market data and renewable energy supply data via the internet, including energy prices and supply information, which is retrieved through internet services and APIs.

[0847] The server preprocesses the collected data by imputing or removing missing or outlier values ​​and reordering the data by timestamp. The server then analyzes the data and identifies patterns of energy consumption using histograms and time series analysis.

[0848] The server then uses a machine learning algorithm (e.g., an LSTM model) to predict future energy demand. The server divides the acquired data into a training dataset and a test dataset, and uses the training dataset to build and train a demand forecasting model.

[0849] Once the predictive model is complete, the server uses it to run an energy procurement optimization algorithm, specifically using multivariable linear programming and dynamic programming to develop an optimal energy supply plan that minimizes costs and environmental impacts.

[0850] The formulated energy supply plan is presented to the user, who can check the plan in graph and chart format on the management screen. Each factory device shares its energy consumption data, and the central server uses this data to generate an optimal operation schedule.

[0851] The server also monitors energy usage in real time, detecting anomalies and sending alerts to users if consumption patterns differ from those expected, providing them with the means to modify their energy supply plans as needed.

[0852] As a specific example, when a user signs in to the system and uploads energy usage data from the past year, the server collects real-time energy market data and renewable energy supply forecasts to predict energy demand for the next month. Based on this, an optimal energy supply plan is created, and the user receives a plan that allows them to reduce costs by 10% by increasing energy usage during late-night hours. Each piece of equipment in the factory adjusts its energy consumption based on this plan, and the server monitors usage in real time. If a consumption pattern that differs from the forecast occurs, the server sends an alert to the user and provides a new optimized plan.

[0853] Prompt Sentence Examples

[0854] Energy consumption data:

[0855] timestamp: 2023-10-01 00:00:00, consumption: 50

[0856] timestamp: 2023-10-01 01:00:00, consumption: 55

[0857] ...

[0858] timestamp: 2023-10-01 23:00:00, consumption: 58

[0859] Forecast your energy needs for the next month and provide an optimal energy supply plan. Include specific actions to reduce energy usage during peak hours.

[0860] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0861] Step 1:

[0862] The server collects past energy usage data. Data is sent to the server from smart meters and sensors connected to each device. The input here is the measurement data from each smart meter and sensor, and the output is an energy usage dataset that compiles this data. Specifically, the server periodically receives data from each device and stores it in a database.

[0863] Step 2:

[0864] The server obtains real-time energy market data and renewable energy supply data. This data includes energy prices and supply information and is obtained from an external API via the internet. The input is the response data from the API, and the output is the obtained market data and supply data. Specifically, the server calls the API at regular intervals to obtain the latest data.

[0865] Step 3:

[0866] The server preprocesses the collected data. Specifically, it complements or removes missing values ​​and outliers, and rearranges the data in timestamp order. The input is the data collected in steps 1 and 2, and the output is clean data that has been preprocessed. Specifically, the server automatically performs data cleansing and performs any necessary data transformations.

[0867] Step 4:

[0868] The server analyzes the pre-processed data to identify patterns in energy consumption. Histograms and time series analysis are used to identify periods of peak and low usage. The input is clean data, and the output is analysis results that show consumption patterns. Specifically, the server runs time series data analysis algorithms to extract specific patterns and trends.

[0869] Step 5:

[0870] The server uses a machine learning algorithm (e.g., an LSTM model) to build a model that predicts future energy demand. First, it divides the collected data into a training dataset and a test dataset, and then trains the model based on the training dataset. The input is the preprocessed data and the machine learning algorithm, and the output is a trained predictive model. Specifically, the server uses a machine learning framework (e.g., TensorFlow or Keras) to train the model.

[0871] Step 6:

[0872] The server uses the forecasting model to predict future energy demand. The input is a trained forecasting model and the latest dataset, and the output is the future demand forecast result. Specifically, the server inputs the latest data into the demand forecasting model and generates the forecast result.

[0873] Step 7:

[0874] The server executes an optimization algorithm for energy procurement and formulates an optimal energy supply plan. It minimizes costs and environmental impacts using multivariable linear programming and dynamic programming. The inputs are the demand forecast results and the optimization algorithm, and the output is an optimal energy supply plan. Specifically, the server executes optimization calculations and generates the optimal plan within the specified constraints.

[0875] Step 8:

[0876] The server presents the formulated energy supply plan to the user. The user can check the plan in graph and chart format through the management screen. The input is the optimal energy supply plan, and the output is the supply plan displayed to the user. Specifically, the server visualizes the plan and displays it on a dashboard.

[0877] Step 9:

[0878] Each piece of factory equipment shares its energy consumption data, and a central server uses this data to generate an optimal operation schedule. The input is the energy consumption data from each piece of equipment, and the output is the optimal operation schedule. Specifically, the server periodically collects data from each piece of equipment and calculates the optimal schedule.

[0879] Step 10:

[0880] The server monitors energy usage in real time, alerts users if anomalies occur, and modifies energy supply plans as needed. The input is real-time energy usage data, and the output is modified energy supply plans and alerts. Specifically, the server tracks consumption data through the monitoring system and takes action when anomalies are detected.

[0881] 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.

[0882] The present invention is an electricity procurement system for businesses and households to optimize the efficiency of electricity usage and reduce costs and environmental burdens. The system further incorporates an emotion engine to recognize user emotions and improve the effectiveness of supply plans. The system of the present invention will be described in detail below.

[0883] First, the user accesses the server and signs in to the system. Once the user uploads their past electricity usage data, the server begins collecting and preprocessing the data. The server obtains the user's electricity usage data from the past year through the smart meter system. At the same time, the server collects electricity market data and renewable energy supply data via the Internet. This data includes real-time electricity price information and renewable energy supply forecasts.

[0884] The server then preprocesses and cleanses the data, imputing or removing missing or outlier values ​​and reordering them by timestamp. The server then analyzes the data to identify patterns in power consumption, for example, using histograms and time series analysis to identify periods of peak and low usage.

[0885] Next, the server uses a machine learning algorithm (e.g., ARIMA model or LSTM model) to build a model that predicts future electricity demand. The server divides the acquired data into a training dataset and a test dataset, and uses the training dataset to train the demand forecasting model. The accuracy of the forecasting model is evaluated to confirm its performance.

[0886] The server runs an optimization algorithm for power procurement based on the demand forecast model to develop an optimal power supply plan. Specifically, it uses multivariable linear programming and dynamic programming to calculate the optimal procurement strategy that minimizes costs and environmental impacts. The server then presents this optimal plan to the user. The user can view the power supply plan in graph and chart format through the server's management screen. The server also sends notifications to the user and provides details of the plan.

[0887] This is where the emotion engine comes in. The emotion engine acquires the user's emotion data, analyzes it, and evaluates the effectiveness of the power supply plan. For example, if the user is feeling stressed, it can use that information to suggest an energy consumption plan that will help reduce stress. The emotion engine collects emotion data using voice and image analysis.

[0888] The server monitors power usage in real time and adjusts the plan as needed. For example, if power consumption is not as expected, the server detects the anomaly and alerts the user. The server also updates the predictive model based on real-time data and runs new optimization algorithms.

[0889] As a concrete example, consider the case where a user signs in to the system and uploads electricity usage data from the past year. The server collects real-time electricity market data and renewable energy supply forecasts. Based on this, the server predicts electricity demand for the next month and creates and presents an optimal electricity supply plan. The user reviews the plan, and the emotion engine recognizes the user's stress level, and the server suggests a plan that will reduce stress. The device adjusts electricity consumption based on this plan, and the server monitors usage in real time. If a consumption pattern that differs from the forecast occurs, the server sends an alert to the user and provides a new optimized plan if necessary.

[0890] The processing flow will be explained below.

[0891] Step 1:

[0892] A user signs in to the server and uploads their historical electricity usage data. The server retrieves the past year's electricity usage data from the smart meter system, and also collects electricity market data and renewable energy supply data via the Internet.

[0893] Step 2:

[0894] The server cleanses and preprocesses the acquired data. Specifically, the server detects, imputes, or removes missing or outlier values. It also sorts the data by timestamp and generates daily and monthly aggregated data.

[0895] Step 3:

[0896] The server analyzes the pre-processed data to identify patterns of power consumption, using histograms and time series analysis to identify periods of peak and low usage, and then performs cluster analysis to create distinct consumption pattern groups.

[0897] Step 4:

[0898] The server uses a machine learning algorithm to build a future electricity demand forecasting model. The server divides the data into a training dataset and a test dataset, and trains an ARIMA model or LSTM model using the training dataset. The model's predictive accuracy is then evaluated using the test dataset.

[0899] Step 5:

[0900] The server runs an optimization algorithm for power procurement based on the trained power demand forecasting model. Specifically, the server uses multivariable linear programming and dynamic programming to formulate an optimal power supply plan that minimizes costs and environmental impacts.

[0901] Step 6:

[0902] The server presents the optimized power supply plan to the user, who can then view the plan on the server's management screen and receive detailed information through notifications. The plan is visually displayed in graphs and charts.

[0903] Step 7:

[0904] The emotion engine works to obtain the user's emotion data. The emotion engine uses voice and image analysis to read the user's emotion in real time. For example, it analyzes the user's facial expressions and tone of voice through a camera or microphone.

[0905] Step 8:

[0906] The server analyzes the acquired emotional data and evaluates the user's current emotional state. For example, if the user is feeling stressed, the server adjusts the power supply plan based on that information.

[0907] Step 9:

[0908] Based on the data provided by the emotion engine, the server re-optimizes the power supply plan. A new power supply plan with a stress-reducing effect is formulated and presented to the user. The server then visually displays the plan again and sends a notification to the user.

[0909] Step 10:

[0910] The device adjusts power consumption based on the new power supply plan set by the device, controls smart home appliances and HVAC systems, and increases or decreases power consumption during specified times. The device transmits power usage data to the server in real time.

[0911] Step 11:

[0912] The server monitors power usage in real time and adjusts the plan as needed. If it detects abnormal consumption patterns or unexpected conditions, it will alert the user and provide a new optimized plan accordingly.

[0913] Example 2

[0914] 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."

[0915] The problem that this invention aims to solve is to optimize the efficiency of power usage in businesses and homes and reduce costs and environmental burdens. Conventional power supply systems do not propose power supply plans that take user emotions into consideration, making it difficult to improve user satisfaction. Another problem is the lack of real-time monitoring of power usage and rapid response after anomaly detection.

[0916] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for acquiring historical power usage data; means for acquiring real-time power market data and renewable energy supply data; means for analyzing the historical power usage data and the power market data to identify power consumption patterns; means for constructing a model for predicting future power demand using a machine learning algorithm; means for formulating an optimal power supply plan by executing a power procurement optimization algorithm based on the prediction model; means for presenting the optimal power supply plan to a user; means for acquiring and analyzing user emotion data to evaluate the effectiveness of the power supply plan; and means for monitoring power usage in real time and modifying the power supply plan. This makes it possible to reduce costs and environmental burdens by optimizing power usage efficiency, improve user satisfaction, and achieve quick responses in real time.

[0917] "Past power usage data" refers to data relating to the amount of power consumed by the user in the past and the time of day.

[0918] "Real-time electricity market data" means information about the current electricity market price and supply.

[0919] "Renewable energy supply data" is information about electricity supplied from renewable energy sources such as wind, solar, and hydroelectric power.

[0920] The "means for identifying patterns of power consumption" is a means for analyzing power usage data and identifying trends in power consumption and peak times during a specific period.

[0921] A "machine learning algorithm" is a method or technology for predicting future data by analyzing and learning from data.

[0922] A "model for predicting future electricity demand" is a mathematical or statistical model for estimating future electricity consumption based on collected data.

[0923] The "power procurement optimization algorithm" is a calculation method for efficiently procuring electricity while taking into account factors such as cost and environmental impact.

[0924] An "optimal power supply plan" is a power supply strategy that meets users' power demand while minimizing costs and environmental burdens.

[0925] "User emotional data" is information about the user's psychological state and emotions, and is data collected through voice and image analysis.

[0926] "Means for monitoring power usage in real time" refers to means for constantly monitoring users' power usage and responding immediately if any abnormalities are detected.

[0927] The present invention is an electricity procurement system for optimizing the efficiency of electricity usage in businesses and homes, reducing costs and environmental burdens, and further incorporates an emotion engine for recognizing user emotions to improve the effectiveness of supply plans.

[0928] Overall system overview

[0929] First, the user accesses the server and signs in to the system. Once the user uploads their past electricity usage data, the server begins collecting and preprocessing the data. The server obtains the user's electricity usage data from the past year through the smart meter system. At the same time, the server collects electricity market data and renewable energy supply data via the Internet. This data includes real-time electricity price information and renewable energy supply forecasts.

[0930] Hardware and Software Use

[0931] The server preprocesses the uploaded data using Python's Pandas library and NumPy. Specifically, it imputes missing values ​​and removes outliers. It then uses data visualization tools such as Matplotlib and Seaborn to identify patterns in power consumption through histograms and time series analysis.

[0932] Next, the server builds a machine learning model using Scikit-learn and TensorFlow. It uses ARIMA and LSTM models to predict future electricity demand. In this process, the collected data is divided into a training dataset and a test dataset, and the model is trained.

[0933] To develop an optimal power supply plan, optimization libraries such as Gurobi and PuLP are used. The server uses multivariable linear programming and dynamic programming to calculate the optimal procurement strategy that minimizes costs and environmental impacts. The plan is presented to the user in the form of graphs and charts using the Plotly library.

[0934] Furthermore, the emotion engine collects and analyzes the user's emotional data through voice and image analysis. For example, it uses emotion analysis tools such as AWS Rekognition and Watson. If the user is feeling stressed, it can use that information to suggest a stress-relieving plan.

[0935] Adding specific examples

[0936] As a concrete example, consider the case where a user signs in to the system and uploads electricity usage data from the past year. The server collects real-time electricity market data and renewable energy supply forecasts. Based on this, the server predicts electricity demand for the next month and creates and presents an optimal electricity supply plan. The user reviews the plan, and the emotion engine recognizes the user's stress level, and the server suggests a plan that will reduce stress. The device adjusts electricity consumption based on this plan, and the server monitors usage in real time. If a consumption pattern that differs from the forecast occurs, the server sends an alert to the user and provides a new optimized plan if necessary.

[0937] Prompt Sentence Examples

[0938] An example of an input prompt for a generative AI model is as follows:

[0939] "Please tell me the detailed steps to implement a system where a user can upload electricity usage data from the past year and the system will suggest the optimal electricity supply plan. Please also provide a concrete coding example."

[0940] By implementing this system, it is possible to reduce costs and environmental burdens by optimizing power usage efficiency, improve user satisfaction, and realize quick responses in real time.

[0941] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0942] Step 1:

[0943] A user accesses the system's web application and signs in by entering their username and password. The input is the username and password, and the output is an authentication token. The server checks the user authentication information in the database and generates an authentication token if authentication is successful. Specifically, the user opens a browser, enters a URL to access the system's login page, enters their username and password in the form, and clicks the "Login" button.

[0944] Step 2:

[0945] Users upload the electricity usage data from the past year to the system from their own devices. The input is a file of past electricity usage data, and the output is the electricity usage data stored in the server's database. The server receives this data through the smart meter system and stores it in the database. Specifically, the user clicks the "Upload data" button, selects the data file, and uploads it.

[0946] Step 3:

[0947] The server collects electricity market data and renewable energy supply data via the Internet. The input is electricity market data and renewable energy data from APIs and databases, and the output is the server's database where this data is stored. Specifically, the server periodically calls external APIs to collect electricity market data and renewable energy data.

[0948] Step 4:

[0949] The server performs preprocessing on the collected data. The input is the uploaded electricity usage data, electricity market data, and renewable energy data, and the output is the cleansed data. Specifically, it uses Python's Pandas library and NumPy to fill in missing values ​​and remove outliers. Specifically, the server fills in missing values ​​using Pandas' DataFrame.fillna() method and removes outliers using the DataFrame.drop() method.

[0950] Step 5:

[0951] The server analyzes the cleansed data to identify patterns in power consumption. The input is the cleansed power usage data, and the output is information about power consumption patterns. Histograms and time series analysis are used to identify periods of peak and low usage. Specifically, Matplotlib and Seaborn are used to create histograms and visualize consumption patterns.

[0952] Step 6:

[0953] The server uses a machine learning algorithm to build an electricity demand forecasting model. The input is the cleansed dataset, and the output is the demand forecasting model. Scikit-learn and TensorFlow are used to train ARIMA and LSTM models. Specifically, the dataset is split into training data and test data, and the data is separated using Scikit-learn's train_test_split() method.

[0954] Step 7:

[0955] The server runs an optimization algorithm for power procurement based on the forecast model. The input is a demand forecast model and electricity market data, and the output is an optimal power supply plan. It uses optimization libraries such as Gurobi and PuLP to perform multivariable linear programming and dynamic programming. Specifically, it sets up an optimization problem in Gurobi and calculates the optimal power supply plan.

[0956] Step 8:

[0957] The server presents the optimal power supply plan to the user. The input is the optimal power supply plan, and the output is a plan in the form of graphs and charts that the user can view. Using the Plotly library, the optimal plan is visualized and displayed on a web page. Specifically, the server creates a graph using Plotly and displays it on an admin screen that the user can access.

[0958] Step 9:

[0959] The emotion engine acquires and analyzes the user's emotional data. The input is voice and image data, and the output is analyzed emotional data. Emotions are analyzed using AWS Rekognition or Watson, and the server evaluates the effectiveness of the plan. Specifically, the camera acquires the user's facial expression data, which is then sent to the emotion analysis tool to obtain the analysis results.

[0960] Step 10:

[0961] The server monitors power usage in real time and sends an alert to the user if it detects an abnormality. The input is real-time power usage data and the output is an alert message. Monitoring and alerts are implemented using Prometheus and Grafana. Specifically, the server collects real-time data and sends a notification to the user if the data exceeds a specified threshold.

[0962] (Application example 2)

[0963] 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."

[0964] Conventional power procurement systems formulate power supply plans based on past power usage data and market data, but they have limitations in terms of real-time adjustments to supply plans, particularly optimization that takes user sentiment data into account. Furthermore, optimizing power usage is difficult, particularly in places with large-scale power consumption such as factories, and efficient energy management is required. This presents a challenge in that it is not possible to fully reduce power costs and environmental impact.

[0965] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0966] In this invention, the server includes: means for acquiring historical power usage data; means for acquiring real-time power market data and renewable energy supply data; means for analyzing the historical power usage data and the power market data to identify power consumption patterns; means for constructing a model for predicting future power demand using a machine learning algorithm; means for formulating an optimal power supply plan by executing a power procurement optimization algorithm based on the prediction model; means for presenting the optimal power supply plan to a user; means for monitoring power usage in real time and revising the power supply plan; means for acquiring and analyzing user emotion data; means for revising the power supply plan based on the emotion data; and means for optimizing power usage of factory robots using a smartphone application. This enables real-time optimization of power usage and efficient energy management in a factory while taking user emotion data into consideration.

[0967] "Past power usage data" is history data relating to the amount of power consumed by the user in the past.

[0968] "Real-time electricity market data" means instantaneous data showing the current market electricity prices and supply situation.

[0969] "Renewable energy supply data" refers to data on electricity supply from renewable energy sources such as solar, wind, and hydroelectric power.

[0970] The "power consumption pattern" is an analysis result that indicates the user's power consumption tendency by time period, peak time periods, and low consumption time periods.

[0971] A "machine learning algorithm" is a computer science technique for learning patterns and trends from large amounts of data and making predictions.

[0972] A "model for predicting electricity demand" is a mathematical model constructed to predict future electricity consumption from past data.

[0973] An "optimization algorithm for power procurement" is a mathematical method for calculating the optimal power supply method to minimize power costs and environmental impact.

[0974] An "electricity supply plan" is a plan that determines how electricity will be procured and supplied for future electricity usage.

[0975] The "means for presenting to the user" refers to a method for providing information such as a power supply plan to the user visually or as a notification.

[0976] "Means for monitoring power usage in real time" refers to a system for instantly monitoring current power consumption.

[0977] "User emotion data" is data that indicates the user's emotional state, and is obtained by voice or image analysis, for example.

[0978] A "smartphone application" is software that runs on a smartphone and provides specific functions or services.

[0979] "Means for optimizing power usage of factory robots" is a system for efficiently managing and minimizing power usage of robots operating in factories.

[0980] The present invention is a system for optimizing the efficiency of power usage in businesses and households to reduce costs and environmental burdens, and in particular, uses a smartphone application to improve the efficiency of power usage in factory robots. This system optimizes power supply plans taking into account user emotional data. Specific embodiments of the present invention are described in detail below.

[0981] The system includes a server, a smartphone application, and an emotion data acquisition unit. The server acquires historical electricity usage data and collects real-time electricity market data and renewable energy supply data. It also preprocesses and cleanses this data to identify electricity consumption patterns. This data is then used to build a model for predicting future electricity demand using a machine learning algorithm, and an optimal electricity supply plan is developed.

[0982] Specific software and hardware used in this embodiment include the requests library for data collection, the pandas library for data preprocessing, and the scikit-learn library for machine learning algorithms. Also, EmotionRecognizer is used as software for acquiring user emotion data.

[0983] For example, a user (factory manager) signs in to a smartphone application and uploads electricity usage data from the past year. The server collects real-time electricity market data and renewable energy supply data and analyzes electricity consumption patterns. The server then uses machine learning algorithms to predict future electricity demand and create an optimal electricity supply plan, which it then presents to the user. The server also obtains the user's emotional data and modifies the plan according to their emotional state. The smartphone application also optimizes the power usage of factory robots.

[0984] For example, after a factory manager signs in to the system and uploads data, the server collects electricity market data in real time and executes a process to forecast the next month's electricity demand. During this process, emotion recognition software detects the manager's stress level and presents a power supply plan that will reduce stress. An example of a specific prompt could be to instruct the server, "Please predict electricity usage based on the following data."

[0985] In this way, the system optimizes power usage in factories in real time, enabling efficient energy management.

[0986] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0987] Step 1:

[0988] The server acquires past electricity usage data. When a user uploads electricity usage data from the past year via a smartphone application, the data is sent to the server. The server collects the data using an API endpoint and stores it in a database. The input is the past electricity usage data, and the output is the stored electricity usage data.

[0989] Step 2:

[0990] The server obtains real-time electricity market data and renewable energy supply data. It accesses APIs via the internet to collect current market data and renewable energy supply forecasts. The collected data is stored in the server's database. The input is the real-time electricity market data and renewable energy supply data, and the output is the stored market data and supply data.

[0991] Step 3:

[0992] The server preprocesses the acquired historical electricity usage data and market data to identify electricity consumption patterns. Specifically, it cleanses the data and fills or removes missing values ​​and outliers. Using the cleansed data, it identifies peak and low electricity usage periods through histogram and time series analysis. The input is historical electricity usage data and market data, and the output is electricity consumption patterns.

[0993] Step 4:

[0994] The server uses a machine learning algorithm to build a model to predict future electricity demand. Specifically, it divides the data into a training dataset and a test dataset, and trains a linear regression model using, for example, the scikit-learn library. The input is preprocessed electricity usage data and market data, and the output is the trained demand forecasting model.

[0995] Step 5:

[0996] The server executes an optimization algorithm for power procurement based on the trained forecasting model to formulate an optimal power supply plan. For example, it applies multivariable linear programming based on the predicted power demand to minimize costs and environmental impact. The input is the trained demand forecasting model, and the output is the optimal power supply plan.

[0997] Step 6:

[0998] The server presents the optimal power supply plan to the user, who can then check the plan through a smartphone application. The plan is visually displayed in graphs and charts, and detailed information is also provided. The input is the optimal power supply plan, and the output is the plan presented to the user.

[0999] Step 7:

[1000] The server monitors power usage in real time and modifies the power supply plan. It acquires data sequentially, compares the predictive model with the actual data, detects anomalies, and modifies the plan as necessary. The input is real-time power usage data, and the output is the modified power supply plan.

[1001] Step 8:

[1002] The server acquires and analyzes the user's emotional data. It collects audio and images from smartphones and other devices and analyzes the user's emotional state using EmotionRecognizer software. The input is the user's emotional data, and the output is the analysis results.

[1003] Step 9:

[1004] The server modifies the power supply plan based on the emotion data. For example, if the user is feeling stressed, it generates a plan to reduce power consumption. The input is the analyzed emotion data, and the output is the modified power supply plan.

[1005] Step 10:

[1006] The server uses a smartphone application to optimize the power usage of factory robots. Users can check the robot's operating status and power consumption through the app and operate it according to the optimized plan. The input is the factory robot's power usage data, and the output is the optimized power usage.

[1007] 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.

[1008] 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.

[1009] 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.

[1010] [Fourth embodiment]

[1011] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1012] 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.

[1013] 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).

[1014] 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.

[1015] 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.

[1016] 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).

[1017] 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.

[1018] 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.

[1019] 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.

[1020] 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.

[1021] 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.

[1022] 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.

[1023] 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."

[1024] The present invention relates to an electricity procurement system that enables businesses and households to optimize the efficiency of their electricity usage and reduce costs and environmental impact. The system of the present invention will be described in detail below.

[1025] First, the user accesses the server and signs in to the system. Once the user uploads their past electricity usage data, the server begins collecting and preprocessing the data. The server obtains the user's electricity usage data from the past year through the smart meter system. At the same time, the server collects electricity market data and renewable energy supply data via the Internet. This data includes real-time electricity price information and renewable energy supply forecasts.

[1026] The server then preprocesses and cleanses the data, imputing or removing missing or outlier values ​​and reordering them by timestamp. The server then analyzes the data to identify patterns in power consumption, for example, using histograms and time series analysis to identify periods of peak and low usage.

[1027] Next, the server uses a machine learning algorithm (e.g., ARIMA model or LSTM model) to build a model that predicts future electricity demand. The server divides the acquired data into a training dataset and a test dataset, and uses the training dataset to train the demand forecasting model. The accuracy of the forecasting model is evaluated to confirm its performance.

[1028] The server runs an optimization algorithm for power procurement based on the demand forecast model to develop an optimal power supply plan. Specifically, it uses multivariable linear programming and dynamic programming to calculate the optimal procurement strategy that minimizes costs and environmental impact. The server then presents this optimal plan to the user. The user can view the power supply plan in graph and chart format through the server's management screen. The server also sends notifications to the user and provides details of the plan.

[1029] The server monitors power usage in real time and adjusts the plan as needed. For example, if power consumption is not as expected, the server detects the anomaly and alerts the user. The device then controls smart home appliances and HVAC systems in real time based on the configured power supply plan.

[1030] As a concrete example, consider the case where a user signs in to the system and uploads electricity usage data from the past year. The server collects real-time electricity market data and renewable energy supply forecasts. Based on this, the server predicts electricity demand for the next month and creates and presents an optimal electricity supply plan. The user receives a plan that allows them to reduce costs by 10% by increasing electricity usage during late-night hours. The device adjusts electricity consumption based on this plan, and the server monitors usage in real time. If a consumption pattern that differs from the forecast occurs, the server sends an alert to the user and provides a new optimized plan if necessary.

[1031] The processing flow will be explained below.

[1032] Step 1:

[1033] The user signs in to the server and uploads their historical electricity usage data. The server is then accessed and retrieves the user's electricity usage data from the smart meter system. The server also collects electricity market data and renewable energy supply data via the internet, allowing the system to obtain all the necessary data.

[1034] Step 2:

[1035] The server cleanses and preprocesses the acquired data. Specifically, the server detects missing values ​​and outliers and completes or removes them. The server also sorts the data by timestamp and generates daily and monthly aggregated data. This preprocessing improves the quality of the data.

[1036] Step 3:

[1037] The server analyzes the pre-processed data to identify patterns in power consumption. It uses histograms and time series analysis to identify peak and low usage periods. It also performs cluster analysis to create groups of different consumption patterns. This analysis identifies the characteristics and trends of power usage.

[1038] Step 4:

[1039] The server uses a machine learning algorithm to build a future electricity demand forecasting model. The server divides the data into a training dataset and a test dataset, and uses the training dataset to train a forecasting model (e.g., an ARIMA model or an LSTM model). The model is then evaluated on the test dataset to confirm its forecast accuracy.

[1040] Step 5:

[1041] The server runs an optimization algorithm for power procurement based on a trained power demand forecasting model. Using multivariable linear programming and dynamic programming, the server formulates an optimal power supply plan that minimizes costs and environmental impacts. This plan is created based on future power demand forecasts.

[1042] Step 6:

[1043] The server presents the optimized power supply plan to the user. The server then visualizes the plan and displays it in the form of graphs and charts on the user's management screen. The server also sends detailed information about the plan to the user's smartphone or email, allowing the user to review and select the proposed plan.

[1044] Step 7:

[1045] The device adjusts power consumption based on the set power supply plan. The device controls smart home appliances and HVAC systems, increasing or decreasing power consumption during designated time periods. The device also transmits real-time power usage data to a server.

[1046] Step 8:

[1047] The server monitors power usage in real time and adjusts plans as needed. The server detects unusual consumption patterns or unexpected situations and alerts users. The server also updates predictive models and runs new optimization algorithms based on real-time data.

[1048] Example 1

[1049] 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."

[1050] In modern society, optimizing the efficiency of power usage in businesses and homes and reducing power costs and environmental impact are major challenges. However, conventional systems do not fully automate the collection and analysis of power usage data, resulting in insufficient data to formulate optimal power supply plans. Furthermore, demand forecasting using machine learning algorithms and real-time monitoring and feedback of power usage status are insufficient, making it difficult to optimize power supply.

[1051] 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.

[1052] In this invention, the server includes means for acquiring historical power usage data, means for acquiring real-time power market data and renewable energy supply data, means for analyzing the historical power usage data and the power market data to identify power consumption patterns, means for constructing a model for predicting future power demand using a machine learning algorithm, means for formulating an optimal power supply plan by executing a power procurement optimization algorithm based on the prediction model, means for presenting the optimal power supply plan to a user, and means for monitoring power usage in real time and modifying the power supply plan. This enables automatic collection and analysis of power usage data, construction and verification of a demand forecast model, and real-time presentation and modification of an optimal power supply plan.

[1053] "Past power usage data" is information relating to the amount of power consumed in the past and the timing of its use.

[1054] "Real-time electricity market data" means up-to-date information about current prices and supply conditions in the electricity market.

[1055] "Renewable energy supply data" is information about the amount and forecast of electricity supplied from renewable energy sources, such as solar, wind, and hydroelectric power.

[1056] "Patterns of electricity consumption" are data that indicate trends and characteristics of electricity usage over a specific period of time.

[1057] A "machine learning algorithm" is a computational method for automatically building predictive and classification models using data.

[1058] A "model for predicting future electricity demand" is a model used to predict future electricity consumption from past data and current conditions.

[1059] An "optimal power supply plan" is a power supply plan designed to maximize the efficiency of power use and minimize costs and environmental impacts.

[1060] A "smart meter system" is a measuring device and communication system that measures electricity consumption in real time and collects and transmits that data remotely.

[1061] A "missing value" refers to a value that is missing in a dataset.

[1062] An "outlier" refers to a value in a data set that falls outside the normal range.

[1063] A "training dataset" is a collection of data used to train a machine learning model.

[1064] A "test dataset" is a collection of data used to evaluate the performance of a trained machine learning model.

[1065] "Real-time monitoring of power usage" refers to the process of monitoring power consumption in real time and taking immediate action if necessary.

[1066] A "terminal" is a device for controlling appliances based on a power supply plan.

[1067] The present invention relates to an electricity procurement system that allows businesses and households to optimize the efficiency of their electricity usage and reduce costs and environmental impacts. The system is comprised of a combination of software and hardware that collects, analyzes, and optimizes historical electricity usage data, real-time electricity market data, and renewable energy supply data.

[1068] First, users access the system's web portal or mobile app and sign in to the system by entering their authentication information on the login screen. They then upload their electricity usage data from the past year in a format such as a CSV file. This data is received by the server and stored. The server then automatically collects electricity usage data through the smart meter system and performs preprocessing to fill in or remove missing or outlier values. The server also obtains the latest real-time electricity market data and renewable energy supply data via the internet.

[1069] The server then uses this preprocessed data to analyze patterns of power consumption, for example by converting it into a data frame using Python's pandas library and generating histograms and time series graphs using the matplotlib library to identify periods of peak and low usage.

[1070] Furthermore, we will build a model to forecast electricity demand using machine learning techniques. Specifically, we will train ARIMA and LSTM models using machine learning libraries such as scikit-learn and TensorFlow, and evaluate the performance of the models using training and test datasets. Based on this trained model, it will be possible to forecast future electricity demand with high accuracy.

[1071] The server then uses multivariable linear programming and dynamic programming to run an optimization algorithm for power procurement based on the demand forecast model. This allows the creation of an optimal power supply plan that minimizes costs and environmental impact. Users can view this optimal supply plan in graph and chart format via a GUI. The server also notifies users of details via email and push notifications.

[1072] When the supply plan is actually executed, the device controls smart home appliances and HVAC (heating, ventilation, and air conditioning) systems in real time based on the plan. The server monitors power usage in real time and alerts the user if consumption does not progress as predicted and provides a new optimization plan.

[1073] As a concrete example, consider a case where a user signs in to the system and uploads their electricity usage data from the past year. This data is collected by a server, which also retrieves real-time electricity market data and renewable energy supply forecasts. Based on this, the server predicts electricity demand for the next month and creates and presents an optimal electricity supply plan. For example, one plan could reduce costs by 10% by increasing electricity usage during late-night hours. The device adjusts its electricity consumption based on this plan, and the server monitors the situation in real time. If a consumption pattern differs from the forecast, the server sends an alert to the user and presents a new plan if necessary.

[1074] Examples of input prompts for generative AI models include:

[1075] Design an electricity procurement system that helps businesses and households optimize their electricity usage efficiency and reduce costs and environmental impact. Use historical electricity usage data, real-time electricity market data, and renewable energy supply forecast data to build a model that forecasts electricity demand using machine learning algorithms such as ARIMA and LSTM models. Then, design a system that calculates and proposes the optimal procurement strategy that minimizes costs and environmental impact.

[1076] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1077] Processing steps of this system's program

[1078] Step 1: Sign in users and upload data

[1079] explanation

[1080] Users access the system's web portal or mobile app and sign in by entering their credentials on the login screen.

[1081] Users upload electricity usage data from the past year in a format such as a CSV file.

[1082] input

[1083] User authentication information (ID and password)

[1084] Power usage data (CSV file)

[1085] output

[1086] Authorization of credentials

[1087] Sending power usage data to a server

[1088] Specific actions

[1089] The user opens a web browser, accesses the system's URL, and enters their ID and password on the login screen.

[1090] After logging in, select "Upload Data" on the dashboard and upload the CSV file.

[1091] Step 2: Data collection and preprocessing by the server

[1092] explanation

[1093] The server receives the power usage data uploaded by the user and stores it in storage.

[1094] The server automatically acquires electricity usage data through the smart meter system and performs pre-processing to complement or remove missing or abnormal values.

[1095] The server collects real-time electricity market data and renewable energy supply data via the Internet.

[1096] input

[1097] User-uploaded electricity usage data

[1098] Electricity usage data from smart meter systems

[1099] Real-time electricity market and renewable energy supply data

[1100] output

[1101] Preprocessed dataset

[1102] Specific actions

[1103] The server parses the uploaded CSV file and stores it in a database.

[1104] The server retrieves data from the smart meter system via API and converts it into a data frame using pandas.

[1105] Missing values ​​are imputed with the mean and outliers are removed. The data is reordered by timestamp.

[1106] Step 3: Analyze data and identify consumption patterns

[1107] explanation

[1108] The server analyzes the pre-processed data and identifies patterns of power consumption through histogram and time series analysis.

[1109] input

[1110] Preprocessed dataset

[1111] output

[1112] Power consumption pattern analysis results

[1113] Specific actions

[1114] The server uses Python's matplotlib library to generate a histogram to see the distribution of consumption.

[1115] Through time series analysis, periods of peak and low usage are graphed.

[1116] Step 4: Building an electricity demand forecast model

[1117] explanation

[1118] The server builds a power demand forecasting model using machine learning algorithms such as the ARIMA model and the LSTM model.

[1119] The server splits the data into training and test datasets and trains and evaluates the model.

[1120] input

[1121] Preprocessed dataset

[1122] output

[1123] A trained electricity demand forecasting model

[1124] Specific actions

[1125] The server uses scikit-learn and TensorFlow to split the data into training and test sets.

[1126] Train the model and evaluate its predictive accuracy on the test set.

[1127] Step 5: Optimize power procurement

[1128] explanation

[1129] Based on the demand forecasting model, the server executes optimization algorithms using multivariable linear programming and dynamic programming to formulate an optimal power supply plan.

[1130] input

[1131] A trained electricity demand forecasting model

[1132] output

[1133] Optimal power supply plan

[1134] Specific actions

[1135] The server uses the SciPy library to perform linear programming and obtain the optimal solution.

[1136] Step 6: Present and notify users of the plan

[1137] explanation

[1138] The server provides and notifies the user of the optimal power supply plan in the form of graphs and charts.

[1139] input

[1140] Optimal power supply plan

[1141] output

[1142] Plan visibility and notification for users

[1143] Specific actions

[1144] The server displays graphs on a web page using D3.js or Chart.js and sends emails and push notifications.

[1145] Step 7: Real-time control and monitoring

[1146] explanation

[1147] The terminal controls smart home appliances and HVAC systems in real time based on the power supply plan received from the server.

[1148] The server monitors power usage in real time and sends an alert to the user if it detects an abnormality.

[1149] input

[1150] Optimal power supply plan

[1151] Real-time electricity usage data

[1152] output

[1153] Control commands for smart appliances and HVAC systems

[1154] Alert notification when an abnormality is detected

[1155] Specific actions

[1156] The terminal automatically controls air conditioning and lighting based on the power supply plan.

[1157] The server runs a real-time monitoring system and sends an email notification to the user when abnormal data is detected.

[1158] (Application example 1)

[1159] 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."

[1160] Reducing energy costs and environmental impacts are important issues for modern factories. However, conventional energy management systems have difficulty optimizing energy consumption in real time and are unable to efficiently manage the energy usage status of each piece of equipment. Therefore, effective means to improve the energy efficiency of the entire factory are needed.

[1161] 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.

[1162] In this invention, the server includes: means for acquiring historical energy usage data; means for acquiring real-time energy market data and renewable energy supply data; means for analyzing the historical energy usage data and the energy market data to identify energy consumption patterns; means for constructing a model for predicting future energy demand using a machine learning algorithm; means for formulating an optimal energy supply plan by executing an energy procurement optimization algorithm based on the prediction model; means for presenting the optimal energy supply plan to a user; means for each device to share energy consumption data; means for a central server to generate an optimal operation schedule using the energy consumption data; and means for monitoring energy usage in real time and modifying the energy supply plan. This enables each device in a factory to use energy efficiently, thereby reducing overall energy costs and environmental impact.

[1163] "Energy usage data" refers to information about energy consumed in the past.

[1164] "Energy Market Data" means real-time information on energy prices and supply conditions.

[1165] "Renewable energy supply data" refers to information on supply and forecasts from renewable energy sources such as solar, wind, and hydroelectric power.

[1166] A "machine learning algorithm" is a computational method for automatically learning patterns from data and making predictions and classifications.

[1167] A "predictive model" is a mathematical or computational model constructed to predict future trends based on past data.

[1168] An "energy procurement optimization algorithm" is a calculation method for optimizing the cost and efficiency of energy procurement.

[1169] An "energy supply plan" is a plan that determines the energy supply schedule and management policy.

[1170] "Means for each device to share energy consumption data" refers to a means for each device in a factory to communicate its own energy consumption status to each other.

[1171] "Means for a central server to generate an optimal operation schedule using energy consumption data" refers to a means for creating an optimal operation schedule for equipment based on energy consumption data collected by a central server within the factory.

[1172] "Means for monitoring energy usage in real time" refers to means for observing current energy usage in real time.

[1173] The "means for modifying an energy supply plan" refers to a means for appropriately modifying an existing energy supply plan in accordance with the actual energy usage situation.

[1174] This invention relates to a management system for improving the efficiency of energy use in factories and reducing costs and environmental impact. The system is composed of a central server and software installed on each piece of factory equipment.

[1175] First, the server acquires past energy usage data. Data is collected from smart meters and sensors connected to each device and sent to the server. Users can also upload their past energy usage data.

[1176] The server then retrieves real-time energy market data and renewable energy supply data via the internet, including energy prices and supply information, which is retrieved through internet services and APIs.

[1177] The server preprocesses the collected data by imputing or removing missing or outlier values ​​and reordering the data by timestamp. The server then analyzes the data and identifies patterns of energy consumption using histograms and time series analysis.

[1178] The server then uses a machine learning algorithm (e.g., an LSTM model) to predict future energy demand. The server divides the acquired data into a training dataset and a test dataset, and uses the training dataset to build and train a demand forecasting model.

[1179] Once the predictive model is complete, the server uses it to run an energy procurement optimization algorithm, specifically using multivariable linear programming and dynamic programming to develop an optimal energy supply plan that minimizes costs and environmental impacts.

[1180] The formulated energy supply plan is presented to the user, who can check the plan in graph and chart format on the management screen. Each factory device shares its energy consumption data, and the central server uses this data to generate an optimal operation schedule.

[1181] The server also monitors energy usage in real time, detecting anomalies and sending alerts to users if consumption patterns differ from those expected, providing them with the means to modify their energy supply plans as needed.

[1182] As a specific example, when a user signs in to the system and uploads energy usage data from the past year, the server collects real-time energy market data and renewable energy supply forecasts to predict energy demand for the next month. Based on this, an optimal energy supply plan is created, and the user receives a plan that allows them to reduce costs by 10% by increasing energy usage during late-night hours. Each piece of equipment in the factory adjusts its energy consumption based on this plan, and the server monitors usage in real time. If a consumption pattern that differs from the forecast occurs, the server sends an alert to the user and provides a new optimized plan.

[1183] Prompt Sentence Examples

[1184] Energy consumption data:

[1185] timestamp: 2023-10-01 00:00:00, consumption: 50

[1186] timestamp: 2023-10-01 01:00:00, consumption: 55

[1187] ...

[1188] timestamp: 2023-10-01 23:00:00, consumption: 58

[1189] Forecast your energy needs for the next month and provide an optimal energy supply plan. Include specific actions to reduce energy usage during peak hours.

[1190] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1191] Step 1:

[1192] The server collects past energy usage data. Data is sent to the server from smart meters and sensors connected to each device. The input here is the measurement data from each smart meter and sensor, and the output is an energy usage dataset that compiles this data. Specifically, the server periodically receives data from each device and stores it in a database.

[1193] Step 2:

[1194] The server obtains real-time energy market data and renewable energy supply data. This data includes energy prices and supply information and is obtained from an external API via the internet. The input is the response data from the API, and the output is the obtained market data and supply data. Specifically, the server calls the API at regular intervals to obtain the latest data.

[1195] Step 3:

[1196] The server preprocesses the collected data. Specifically, it complements or removes missing values ​​and outliers, and rearranges the data in timestamp order. The input is the data collected in steps 1 and 2, and the output is clean data that has been preprocessed. Specifically, the server automatically performs data cleansing and performs any necessary data transformations.

[1197] Step 4:

[1198] The server analyzes the pre-processed data to identify patterns in energy consumption. Histograms and time series analysis are used to identify periods of peak and low usage. The input is clean data, and the output is analysis results that show consumption patterns. Specifically, the server runs time series data analysis algorithms to extract specific patterns and trends.

[1199] Step 5:

[1200] The server uses a machine learning algorithm (e.g., an LSTM model) to build a model that predicts future energy demand. First, it divides the collected data into a training dataset and a test dataset, and then trains the model based on the training dataset. The input is the preprocessed data and the machine learning algorithm, and the output is a trained predictive model. Specifically, the server uses a machine learning framework (e.g., TensorFlow or Keras) to train the model.

[1201] Step 6:

[1202] The server uses the forecasting model to predict future energy demand. The input is a trained forecasting model and the latest dataset, and the output is the future demand forecast result. Specifically, the server inputs the latest data into the demand forecasting model and generates the forecast result.

[1203] Step 7:

[1204] The server executes an optimization algorithm for energy procurement and formulates an optimal energy supply plan. It minimizes costs and environmental impacts using multivariable linear programming and dynamic programming. The inputs are the demand forecast results and the optimization algorithm, and the output is an optimal energy supply plan. Specifically, the server executes optimization calculations and generates the optimal plan within the specified constraints.

[1205] Step 8:

[1206] The server presents the formulated energy supply plan to the user. The user can check the plan in graph and chart format through the management screen. The input is the optimal energy supply plan, and the output is the supply plan displayed to the user. Specifically, the server visualizes the plan and displays it on a dashboard.

[1207] Step 9:

[1208] Each piece of factory equipment shares its energy consumption data, and a central server uses this data to generate an optimal operation schedule. The input is the energy consumption data from each piece of equipment, and the output is the optimal operation schedule. Specifically, the server periodically collects data from each piece of equipment and calculates the optimal schedule.

[1209] Step 10:

[1210] The server monitors energy usage in real time, alerts users if anomalies occur, and modifies energy supply plans as needed. The input is real-time energy usage data, and the output is modified energy supply plans and alerts. Specifically, the server tracks consumption data through the monitoring system and takes action when anomalies are detected.

[1211] 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.

[1212] The present invention is an electricity procurement system for businesses and households to optimize the efficiency of electricity usage and reduce costs and environmental burdens. The system further incorporates an emotion engine to recognize user emotions and improve the effectiveness of supply plans. The system of the present invention will be described in detail below.

[1213] First, the user accesses the server and signs in to the system. Once the user uploads their past electricity usage data, the server begins collecting and preprocessing the data. The server obtains the user's electricity usage data from the past year through the smart meter system. At the same time, the server collects electricity market data and renewable energy supply data via the Internet. This data includes real-time electricity price information and renewable energy supply forecasts.

[1214] The server then preprocesses and cleanses the data, imputing or removing missing or outlier values ​​and reordering them by timestamp. The server then analyzes the data to identify patterns in power consumption, for example, using histograms and time series analysis to identify periods of peak and low usage.

[1215] Next, the server uses a machine learning algorithm (e.g., ARIMA model or LSTM model) to build a model that predicts future electricity demand. The server divides the acquired data into a training dataset and a test dataset, and uses the training dataset to train the demand forecasting model. The accuracy of the forecasting model is evaluated to confirm its performance.

[1216] The server runs an optimization algorithm for power procurement based on the demand forecast model to develop an optimal power supply plan. Specifically, it uses multivariable linear programming and dynamic programming to calculate the optimal procurement strategy that minimizes costs and environmental impacts. The server then presents this optimal plan to the user. The user can view the power supply plan in graph and chart format through the server's management screen. The server also sends notifications to the user and provides details of the plan.

[1217] This is where the emotion engine comes in. The emotion engine acquires the user's emotion data, analyzes it, and evaluates the effectiveness of the power supply plan. For example, if the user is feeling stressed, it can use that information to suggest an energy consumption plan that will help reduce stress. The emotion engine collects emotion data using voice and image analysis.

[1218] The server monitors power usage in real time and adjusts the plan as needed. For example, if power consumption is not as expected, the server detects the anomaly and alerts the user. The server also updates the predictive model based on real-time data and runs new optimization algorithms.

[1219] As a concrete example, consider the case where a user signs in to the system and uploads electricity usage data from the past year. The server collects real-time electricity market data and renewable energy supply forecasts. Based on this, the server predicts electricity demand for the next month and creates and presents an optimal electricity supply plan. The user reviews the plan, and the emotion engine recognizes the user's stress level, and the server suggests a plan that will reduce stress. The device adjusts electricity consumption based on this plan, and the server monitors usage in real time. If a consumption pattern that differs from the forecast occurs, the server sends an alert to the user and provides a new optimized plan if necessary.

[1220] The processing flow will be explained below.

[1221] Step 1:

[1222] A user signs in to the server and uploads their historical electricity usage data. The server retrieves the past year's electricity usage data from the smart meter system, and also collects electricity market data and renewable energy supply data via the Internet.

[1223] Step 2:

[1224] The server cleanses and preprocesses the acquired data. Specifically, the server detects, imputes, or removes missing or outlier values. It also sorts the data by timestamp and generates daily and monthly aggregated data.

[1225] Step 3:

[1226] The server analyzes the pre-processed data to identify patterns of power consumption, using histograms and time series analysis to identify periods of peak and low usage, and then performs cluster analysis to create distinct consumption pattern groups.

[1227] Step 4:

[1228] The server uses a machine learning algorithm to build a future electricity demand forecasting model. The server divides the data into a training dataset and a test dataset, and trains an ARIMA model or LSTM model using the training dataset. The model's predictive accuracy is then evaluated using the test dataset.

[1229] Step 5:

[1230] The server runs an optimization algorithm for power procurement based on the trained power demand forecasting model. Specifically, the server uses multivariable linear programming and dynamic programming to formulate an optimal power supply plan that minimizes costs and environmental impacts.

[1231] Step 6:

[1232] The server presents the optimized power supply plan to the user, who can then view the plan on the server's management screen and receive detailed information through notifications. The plan is visually displayed in graphs and charts.

[1233] Step 7:

[1234] The emotion engine works to obtain the user's emotion data. The emotion engine uses voice and image analysis to read the user's emotion in real time. For example, it analyzes the user's facial expressions and tone of voice through a camera or microphone.

[1235] Step 8:

[1236] The server analyzes the acquired emotional data and evaluates the user's current emotional state. For example, if the user is feeling stressed, the server adjusts the power supply plan based on that information.

[1237] Step 9:

[1238] Based on the data provided by the emotion engine, the server re-optimizes the power supply plan. A new power supply plan with a stress-reducing effect is formulated and presented to the user. The server then visually displays the plan again and sends a notification to the user.

[1239] Step 10:

[1240] The device adjusts power consumption based on the new power supply plan set by the device, controls smart home appliances and HVAC systems, and increases or decreases power consumption during specified times. The device transmits power usage data to the server in real time.

[1241] Step 11:

[1242] The server monitors power usage in real time and adjusts the plan as needed. If it detects abnormal consumption patterns or unexpected conditions, it will alert the user and provide a new optimized plan accordingly.

[1243] Example 2

[1244] 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 robot 414 will be referred to as a "terminal."

[1245] The problem that this invention aims to solve is to optimize the efficiency of power usage in businesses and homes and reduce costs and environmental burdens. Conventional power supply systems do not propose power supply plans that take user emotions into consideration, making it difficult to improve user satisfaction. Another problem is the lack of real-time monitoring of power usage and rapid response after anomaly detection.

[1246] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for acquiring historical power usage data; means for acquiring real-time power market data and renewable energy supply data; means for analyzing the historical power usage data and the power market data to identify power consumption patterns; means for constructing a model for predicting future power demand using a machine learning algorithm; means for formulating an optimal power supply plan by executing a power procurement optimization algorithm based on the prediction model; means for presenting the optimal power supply plan to a user; means for acquiring and analyzing user emotion data to evaluate the effectiveness of the power supply plan; and means for monitoring power usage in real time and modifying the power supply plan. This makes it possible to reduce costs and environmental burdens by optimizing power usage efficiency, improve user satisfaction, and achieve quick responses in real time.

[1247] "Past power usage data" refers to data relating to the amount of power consumed by the user in the past and the time of day.

[1248] "Real-time electricity market data" means information about the current electricity market price and supply.

[1249] "Renewable energy supply data" is information about electricity supplied from renewable energy sources such as wind, solar, and hydroelectric power.

[1250] The "means for identifying patterns of power consumption" is a means for analyzing power usage data and identifying trends in power consumption and peak times during a specific period.

[1251] A "machine learning algorithm" is a method or technology for predicting future data by analyzing and learning from data.

[1252] A "model for predicting future electricity demand" is a mathematical or statistical model for estimating future electricity consumption based on collected data.

[1253] The "power procurement optimization algorithm" is a calculation method for efficiently procuring electricity while taking into account factors such as cost and environmental impact.

[1254] An "optimal power supply plan" is a power supply strategy that meets users' power demand while minimizing costs and environmental burdens.

[1255] "User emotional data" is information about the user's psychological state and emotions, and is data collected through voice and image analysis.

[1256] "Means for monitoring power usage in real time" refers to means for constantly monitoring users' power usage and responding immediately if any abnormalities are detected.

[1257] The present invention is an electricity procurement system for optimizing the efficiency of electricity usage in businesses and homes, reducing costs and environmental burdens, and further incorporates an emotion engine for recognizing user emotions to improve the effectiveness of supply plans.

[1258] Overall system overview

[1259] First, the user accesses the server and signs in to the system. Once the user uploads their past electricity usage data, the server begins collecting and preprocessing the data. The server obtains the user's electricity usage data from the past year through the smart meter system. At the same time, the server collects electricity market data and renewable energy supply data via the Internet. This data includes real-time electricity price information and renewable energy supply forecasts.

[1260] Hardware and Software Use

[1261] The server preprocesses the uploaded data using Python's Pandas library and NumPy. Specifically, it imputes missing values ​​and removes outliers. It then uses data visualization tools such as Matplotlib and Seaborn to identify patterns in power consumption through histograms and time series analysis.

[1262] Next, the server builds a machine learning model using Scikit-learn and TensorFlow. It uses ARIMA and LSTM models to predict future electricity demand. In this process, the collected data is divided into a training dataset and a test dataset, and the model is trained.

[1263] To develop an optimal power supply plan, optimization libraries such as Gurobi and PuLP are used. The server uses multivariable linear programming and dynamic programming to calculate the optimal procurement strategy that minimizes costs and environmental impacts. The plan is presented to the user in the form of graphs and charts using the Plotly library.

[1264] Furthermore, the emotion engine collects and analyzes the user's emotional data through voice and image analysis. For example, it uses emotion analysis tools such as AWS Rekognition and Watson. If the user is feeling stressed, it can use that information to suggest a stress-relieving plan.

[1265] Adding specific examples

[1266] As a concrete example, consider the case where a user signs in to the system and uploads electricity usage data from the past year. The server collects real-time electricity market data and renewable energy supply forecasts. Based on this, the server predicts electricity demand for the next month and creates and presents an optimal electricity supply plan. The user reviews the plan, and the emotion engine recognizes the user's stress level, and the server suggests a plan that will reduce stress. The device adjusts electricity consumption based on this plan, and the server monitors usage in real time. If a consumption pattern that differs from the forecast occurs, the server sends an alert to the user and provides a new optimized plan if necessary.

[1267] Prompt Sentence Examples

[1268] An example of an input prompt for a generative AI model is as follows:

[1269] "Please tell me the detailed steps to implement a system where a user can upload electricity usage data from the past year and the system will suggest the optimal electricity supply plan. Please also provide a concrete coding example."

[1270] By implementing this system, it is possible to reduce costs and environmental burdens by optimizing power usage efficiency, improve user satisfaction, and realize quick responses in real time.

[1271] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1272] Step 1:

[1273] A user accesses the system's web application and signs in by entering their username and password. The input is the username and password, and the output is an authentication token. The server checks the user authentication information in the database and generates an authentication token if authentication is successful. Specifically, the user opens a browser, enters a URL to access the system's login page, enters their username and password in the form, and clicks the "Login" button.

[1274] Step 2:

[1275] Users upload the electricity usage data from the past year to the system from their own devices. The input is a file of past electricity usage data, and the output is the electricity usage data stored in the server's database. The server receives this data through the smart meter system and stores it in the database. Specifically, the user clicks the "Upload data" button, selects the data file, and uploads it.

[1276] Step 3:

[1277] The server collects electricity market data and renewable energy supply data via the Internet. The input is electricity market data and renewable energy data from APIs and databases, and the output is the server's database where this data is stored. Specifically, the server periodically calls external APIs to collect electricity market data and renewable energy data.

[1278] Step 4:

[1279] The server performs preprocessing on the collected data. The input is the uploaded electricity usage data, electricity market data, and renewable energy data, and the output is the cleansed data. Specifically, it uses Python's Pandas library and NumPy to fill in missing values ​​and remove outliers. Specifically, the server fills in missing values ​​using Pandas' DataFrame.fillna() method and removes outliers using the DataFrame.drop() method.

[1280] Step 5:

[1281] The server analyzes the cleansed data to identify patterns in power consumption. The input is the cleansed power usage data, and the output is information about power consumption patterns. Histograms and time series analysis are used to identify periods of peak and low usage. Specifically, Matplotlib and Seaborn are used to create histograms and visualize consumption patterns.

[1282] Step 6:

[1283] The server uses a machine learning algorithm to build an electricity demand forecasting model. The input is the cleansed dataset, and the output is the demand forecasting model. Scikit-learn and TensorFlow are used to train ARIMA and LSTM models. Specifically, the dataset is split into training data and test data, and the data is separated using Scikit-learn's train_test_split() method.

[1284] Step 7:

[1285] The server runs an optimization algorithm for power procurement based on the forecast model. The input is a demand forecast model and electricity market data, and the output is an optimal power supply plan. It uses optimization libraries such as Gurobi and PuLP to perform multivariable linear programming and dynamic programming. Specifically, it sets up an optimization problem in Gurobi and calculates the optimal power supply plan.

[1286] Step 8:

[1287] The server presents the optimal power supply plan to the user. The input is the optimal power supply plan, and the output is a plan in the form of graphs and charts that the user can view. Using the Plotly library, the optimal plan is visualized and displayed on a web page. Specifically, the server creates a graph using Plotly and displays it on an admin screen that the user can access.

[1288] Step 9:

[1289] The emotion engine acquires and analyzes the user's emotional data. The input is voice and image data, and the output is analyzed emotional data. Emotions are analyzed using AWS Rekognition or Watson, and the server evaluates the effectiveness of the plan. Specifically, the camera acquires the user's facial expression data, which is then sent to the emotion analysis tool to obtain the analysis results.

[1290] Step 10:

[1291] The server monitors power usage in real time and sends an alert to the user if it detects an abnormality. The input is real-time power usage data and the output is an alert message. Monitoring and alerts are implemented using Prometheus and Grafana. Specifically, the server collects real-time data and sends a notification to the user if the data exceeds a specified threshold.

[1292] (Application example 2)

[1293] 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 robot 414 will be referred to as a "terminal."

[1294] Conventional power procurement systems formulate power supply plans based on past power usage data and market data, but they have limitations in terms of real-time adjustments to supply plans, particularly optimization that takes user sentiment data into account. Furthermore, optimizing power usage is difficult, particularly in places with large-scale power consumption such as factories, and efficient energy management is required. This presents a challenge in that it is not possible to fully reduce power costs and environmental impact.

[1295] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1296] In this invention, the server includes: means for acquiring historical power usage data; means for acquiring real-time power market data and renewable energy supply data; means for analyzing the historical power usage data and the power market data to identify power consumption patterns; means for constructing a model for predicting future power demand using a machine learning algorithm; means for formulating an optimal power supply plan by executing a power procurement optimization algorithm based on the prediction model; means for presenting the optimal power supply plan to a user; means for monitoring power usage in real time and revising the power supply plan; means for acquiring and analyzing user emotion data; means for revising the power supply plan based on the emotion data; and means for optimizing power usage of factory robots using a smartphone application. This enables real-time optimization of power usage and efficient energy management in a factory while taking user emotion data into consideration.

[1297] "Past power usage data" is history data relating to the amount of power consumed by the user in the past.

[1298] "Real-time electricity market data" means instantaneous data showing the current market electricity prices and supply situation.

[1299] "Renewable energy supply data" refers to data on electricity supply from renewable energy sources such as solar, wind, and hydroelectric power.

[1300] The "power consumption pattern" is an analysis result that indicates the user's power consumption tendency by time period, peak time periods, and low consumption time periods.

[1301] A "machine learning algorithm" is a computer science technique for learning patterns and trends from large amounts of data and making predictions.

[1302] A "model for predicting electricity demand" is a mathematical model constructed to predict future electricity consumption from past data.

[1303] An "optimization algorithm for power procurement" is a mathematical method for calculating the optimal power supply method to minimize power costs and environmental impact.

[1304] An "electricity supply plan" is a plan that determines how electricity will be procured and supplied for future electricity usage.

[1305] The "means for presenting to the user" refers to a method for providing information such as a power supply plan to the user visually or as a notification.

[1306] "Means for monitoring power usage in real time" refers to a system for instantly monitoring current power consumption.

[1307] "User emotion data" is data that indicates the user's emotional state, and is obtained by voice or image analysis, for example.

[1308] A "smartphone application" is software that runs on a smartphone and provides specific functions or services.

[1309] "Means for optimizing power usage of factory robots" is a system for efficiently managing and minimizing power usage of robots operating in factories.

[1310] The present invention is a system for optimizing the efficiency of power usage in businesses and households to reduce costs and environmental burdens, and in particular, uses a smartphone application to improve the efficiency of power usage in factory robots. This system optimizes power supply plans taking into account user emotional data. Specific embodiments of the present invention are described in detail below.

[1311] The system includes a server, a smartphone application, and an emotion data acquisition unit. The server acquires historical electricity usage data and collects real-time electricity market data and renewable energy supply data. It also preprocesses and cleanses this data to identify electricity consumption patterns. This data is then used to build a model for predicting future electricity demand using a machine learning algorithm, and an optimal electricity supply plan is developed.

[1312] Specific software and hardware used in this embodiment include the requests library for data collection, the pandas library for data preprocessing, and the scikit-learn library for machine learning algorithms. Also, EmotionRecognizer is used as software for acquiring user emotion data.

[1313] For example, a user (factory manager) signs in to a smartphone application and uploads electricity usage data from the past year. The server collects real-time electricity market data and renewable energy supply data and analyzes electricity consumption patterns. The server then uses machine learning algorithms to predict future electricity demand and create an optimal electricity supply plan, which it then presents to the user. The server also obtains the user's emotional data and modifies the plan according to their emotional state. The smartphone application also optimizes the power usage of factory robots.

[1314] For example, after a factory manager signs in to the system and uploads data, the server collects electricity market data in real time and executes a process to forecast the next month's electricity demand. During this process, emotion recognition software detects the manager's stress level and presents a power supply plan that will reduce stress. An example of a specific prompt could be to instruct the server, "Please predict electricity usage based on the following data."

[1315] In this way, the system optimizes power usage in factories in real time, enabling efficient energy management.

[1316] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1317] Step 1:

[1318] The server acquires past electricity usage data. When a user uploads electricity usage data from the past year via a smartphone application, the data is sent to the server. The server collects the data using an API endpoint and stores it in a database. The input is the past electricity usage data, and the output is the stored electricity usage data.

[1319] Step 2:

[1320] The server obtains real-time electricity market data and renewable energy supply data. It accesses APIs via the internet to collect current market data and renewable energy supply forecasts. The collected data is stored in the server's database. The input is the real-time electricity market data and renewable energy supply data, and the output is the stored market data and supply data.

[1321] Step 3:

[1322] The server preprocesses the acquired historical electricity usage data and market data to identify electricity consumption patterns. Specifically, it cleanses the data and fills or removes missing values ​​and outliers. Using the cleansed data, it identifies peak and low electricity usage periods through histogram and time series analysis. The input is historical electricity usage data and market data, and the output is electricity consumption patterns.

[1323] Step 4:

[1324] The server uses a machine learning algorithm to build a model to predict future electricity demand. Specifically, it divides the data into a training dataset and a test dataset, and trains a linear regression model using, for example, the scikit-learn library. The input is preprocessed electricity usage data and market data, and the output is the trained demand forecasting model.

[1325] Step 5:

[1326] The server executes an optimization algorithm for power procurement based on the trained forecasting model to formulate an optimal power supply plan. For example, it applies multivariable linear programming based on the predicted power demand to minimize costs and environmental impact. The input is the trained demand forecasting model, and the output is the optimal power supply plan.

[1327] Step 6:

[1328] The server presents the optimal power supply plan to the user, who can then check the plan through a smartphone application. The plan is visually displayed in graphs and charts, and detailed information is also provided. The input is the optimal power supply plan, and the output is the plan presented to the user.

[1329] Step 7:

[1330] The server monitors power usage in real time and modifies the power supply plan. It acquires data sequentially, compares the predictive model with the actual data, detects anomalies, and modifies the plan as necessary. The input is real-time power usage data, and the output is the modified power supply plan.

[1331] Step 8:

[1332] The server acquires and analyzes the user's emotional data. It collects audio and images from smartphones and other devices and analyzes the user's emotional state using EmotionRecognizer software. The input is the user's emotional data, and the output is the analysis results.

[1333] Step 9:

[1334] The server modifies the power supply plan based on the emotion data. For example, if the user is feeling stressed, it generates a plan to reduce power consumption. The input is the analyzed emotion data, and the output is the modified power supply plan.

[1335] Step 10:

[1336] The server uses a smartphone application to optimize the power usage of factory robots. Users can check the robot's operating status and power consumption through the app and operate it according to the optimized plan. The input is the factory robot's power usage data, and the output is the optimized power usage.

[1337] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice 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 voice data.

[1338] 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.

[1339] 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 robot 414.

[1340] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1341] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1342] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1343] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1344] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1345] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1346] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1347] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1348] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1349] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1350] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1351] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1352] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1353] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1354] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1355] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1356] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1357] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1358] The following is further disclosed regarding the above embodiment.

[1359] (Claim 1)

[1360] a means for obtaining historical electricity usage data;

[1361] a means for obtaining real-time electricity market data and renewable energy supply data;

[1362] means for analyzing the historical electricity usage data and the electricity market data to identify patterns of electricity consumption;

[1363] a means for constructing a model that uses machine learning algorithms to predict future electricity demand;

[1364] a means for executing an optimization algorithm for power procurement based on the prediction model to formulate an optimal power supply plan;

[1365] means for presenting the optimal power supply plan to a user;

[1366] means for monitoring power usage in real time and modifying the power supply plan;

[1367] A system including:

[1368] (Claim 2)

[1369] The system of claim 1, further comprising: preprocessing the electricity usage data to impute or remove missing values ​​and outliers.

[1370] (Claim 3)

[1371] The system according to claim 1, wherein a demand forecasting model is trained using a training data set and a test data set in a machine learning algorithm, and electricity demand is forecast based on the trained model.

[1372] "Example 1"

[1373] (Claim 1)

[1374] a means for obtaining historical electricity usage data;

[1375] a means for obtaining real-time electricity market data and renewable energy supply data;

[1376] means for analyzing the historical electricity usage data and the electricity market data to identify patterns of electricity consumption;

[1377] a means for constructing a model that uses machine learning algorithms to predict future electricity demand;

[1378] a means for executing an optimization algorithm for power procurement based on the prediction model to formulate an optimal power supply plan;

[1379] means for presenting the optimal power supply plan to a user;

[1380] means for monitoring power usage in real time and modifying the power supply plan;

[1381] a means for users to access the system and upload data;

[1382] a means for acquiring electricity usage data through a smart meter system;

[1383] a means of imputing or removing missing and outlier values;

[1384] a means for preprocessing the data to build a demand forecasting model;

[1385] a means for training a model using a machine learning algorithm and evaluating its performance;

[1386] A means to present the optimal power supply plan through a GUI,

[1387] A means for the terminal to control the appliance based on the power supply plan;

[1388] A system including:

[1389] (Claim 2)

[1390] The system of claim 1, further comprising: preprocessing the electricity usage data to impute or remove missing values ​​and outliers.

[1391] (Claim 3)

[1392] The system according to claim 1, wherein a demand forecasting model is trained using a training data set and a test data set in a machine learning algorithm, and electricity demand is forecast based on the trained model.

[1393] "Application Example 1"

[1394] (Claim 1)

[1395] a means for obtaining historical energy usage data;

[1396] a means for obtaining real-time energy market data and renewable energy supply data;

[1397] means for analyzing the historical energy usage data and the energy market data to identify patterns of energy consumption;

[1398] a means for building a model that uses machine learning algorithms to predict future energy demand;

[1399] a means for formulating an optimal energy supply plan by executing an optimization algorithm for energy procurement based on the prediction model;

[1400] means for presenting the optimal energy supply plan to a user;

[1401] A means for each device to share energy consumption data;

[1402] A means for a central server to generate an optimal operating schedule using energy consumption data;

[1403] means for monitoring energy usage in real time and modifying said energy supply plan;

[1404] A system including:

[1405] (Claim 2)

[1406] 10. The system of claim 1, wherein the energy usage data is preprocessed to impute or remove missing values ​​and outliers.

[1407] (Claim 3)

[1408] The system of claim 1, wherein a demand forecasting model is trained using a training data set and a test data set in a machine learning algorithm, and energy demand is forecast based on the trained model.

[1409] "Example 2: Combining Emotion Engines"

[1410] (Claim 1)

[1411] a means for obtaining historical electricity usage data;

[1412] a means for obtaining real-time electricity market data and renewable energy supply data;

[1413] means for analyzing the historical electricity usage data and the electricity market data to identify patterns of electricity consumption;

[1414] a means for constructing a model that uses machine learning algorithms to predict future electricity demand;

[1415] a means for executing an optimization algorithm for power procurement based on the prediction model to formulate an optimal power supply plan;

[1416] means for presenting the optimal power supply plan to a user;

[1417] a means for acquiring and analyzing user emotion data to evaluate the effectiveness of the power supply plan;

[1418] means for monitoring power usage in real time and modifying the power supply plan;

[1419] A system including:

[1420] (Claim 2)

[1421] The system of claim 1, further comprising: preprocessing the electricity usage data to impute or remove missing values ​​and outliers.

[1422] (Claim 3)

[1423] The system according to claim 1, wherein a demand forecasting model is trained using a training data set and a test data set in a machine learning algorithm, and electricity demand is forecast based on the trained model.

[1424] "Application example 2 when combining emotion engines"

[1425] (Claim 1)

[1426] a means for obtaining historical electricity usage data;

[1427] a means for obtaining real-time electricity market data and renewable energy supply data;

[1428] means for analyzing the historical electricity usage data and the electricity market data to identify patterns of electricity consumption;

[1429] a means for constructing a model that uses machine learning algorithms to predict future electricity demand;

[1430] a means for executing an optimization algorithm for power procurement based on the prediction model to formulate an optimal power supply plan;

[1431] means for presenting the optimal power supply plan to a user;

[1432] means for monitoring power usage in real time and modifying the power supply plan;

[1433] A means for acquiring and analyzing user emotion data;

[1434] means for modifying a power supply plan based on the emotion data;

[1435] A means to optimize the power usage of factory robots using a smartphone application;

[1436] A system including:

[1437] (Claim 2)

[1438] The system of claim 1, further comprising: preprocessing the electricity usage data to impute or remove missing values ​​and outliers.

[1439] (Claim 3)

[1440] The system according to claim 1, wherein a demand forecasting model is trained using a training data set and a test data set in a machine learning algorithm, and electricity demand is forecast based on the trained model. [Explanation of symbols]

[1441] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for obtaining historical electricity usage data; a means for obtaining real-time electricity market data and renewable energy supply data; means for analyzing the historical electricity usage data and the electricity market data to identify patterns of electricity consumption; a means for constructing a model that uses machine learning algorithms to predict future electricity demand; a means for executing an optimization algorithm for power procurement based on the prediction model to formulate an optimal power supply plan; means for presenting the optimal power supply plan to a user; means for monitoring power usage in real time and modifying the power supply plan; A system including:

2. The system of claim 1 , wherein the system preprocesses the power usage data to impute or remove missing values ​​and outliers.

3. The system according to claim 1 , wherein a demand forecasting model is trained using a training data set and a test data set in a machine learning algorithm, and electricity demand is forecast based on the trained model.

Images