system

Abstract

We provide the system. [Solution] A means for collecting and preprocessing energy consumption data, A method for predicting energy consumption spikes using recurrent neural networks and long- and short-term memory, A means of generating multiple consumption scenarios using generative AI, evaluating their effectiveness, and selecting the optimal scenario, A means of automatically generating smart contracts based on selected consumption scenarios, A means of notifying users of consumption shift plans via a terminal and obtaining their approval, A system that includes means for adjusting and executing the operating schedule of equipment based on user approval.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of the chatbot's 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] Due to the sudden increase in energy consumption in a specific time period, there is a problem that the peak load in the power supply system increases, leading to an increase in energy supply costs and a decrease in supply efficiency. It is necessary to address this problem, suppress peak energy consumption, reduce energy costs, and improve supply efficiency. Also, promoting sustainable energy use by leveling energy consumption is an important issue.

Means for Solving the Problems

[0005] This invention provides a system that collects and preprocesses energy consumption data in real time and predicts consumption spikes using recurrent neural networks and long- and short-term memory. Furthermore, it utilizes generative AI to generate multiple consumption scenarios, evaluates their effectiveness, selects the optimal scenario, and automatically generates it as a smart contract. After the user approves the consumption shift plan notified to the user, the system adjusts the operating schedule of equipment via a terminal, effectively shifting peak consumption. This results in reduced energy costs and efficient energy use.

[0006] "Energy consumption data" refers to information about the amount of electricity used by a particular piece of equipment or system within a specific time period.

[0007] "Preprocessing" is the process of improving data quality by imputing missing values ​​and removing outliers before performing analysis or modeling.

[0008] A "recurrent neural network" is a type of artificial neural network developed to handle time-series data, and it is a model that can predict future states by considering past information.

[0009] "Long-short-term memory" is an extension of recurrent neural networks that has the ability to capture long-term dependencies, improving the prediction accuracy of time-series data.

[0010] A "consumption spike" is a phenomenon that shows a sudden increase in energy consumption during a specific period of time.

[0011] "Generative AI" is an artificial intelligence technology that automatically creates future scenarios for energy consumption and selects the optimal one from among them.

[0012] A "consumption scenario" is a simulation result that assumes different patterns of energy consumption and evaluates consumption behavior and impacts in each pattern.

[0013] A "smart contract" is a contract that functions as a self-executing program on the blockchain, automatically performing predefined operations when certain conditions are met.

[0014] A "terminal" refers to a device such as a computer or smartphone used by a user, which functions as an interface to the system.

[0015] A "consumption shift plan" is a specific plan to shift energy consumption from peak hours to off-peak hours. [Brief explanation of the drawing]

[0016] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11]It is a sequence diagram showing the processing flow of the data processing system in Embodiment 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Embodiment 2 when the emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when the emotion engine is combined.

Mode for Carrying Out the Invention

[0017] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0018] First, the terms used in the following description will be explained.

[0019] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0020] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0021] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0022] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0024] [First Embodiment]

[0025] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0026] As shown in Figure 1, the 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.

[0027] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0029] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and 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.

[0030] 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 perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0032] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0034] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0036] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0037] The present invention's system supports energy consumption efficiency by using generative AI to predict future energy consumption spikes and automatically executes plans to level out consumption. This system includes a server that collects and analyzes data in real time, a terminal that notifies and approves of consumption plans, and a user that controls equipment based on the plan.

[0038] The server first collects energy consumption data from smart meters and IoT sensors. This data includes time-based consumption, weather conditions, and facility operating status. After collecting the data, the server preprocesses it to improve data quality by imputing missing values ​​and removing outliers.

[0039] Next, the server uses recurrent neural networks (RNNs) and long-short-term memory (LSTMs) to predict future energy consumption patterns. This allows it to detect consumption spikes at specific times. Based on the prediction results, the server uses generative AI to generate multiple consumption scenarios. The server then evaluates each scenario to select the most efficient consumption shift plan from among them.

[0040] The selected scenario is automatically generated as a smart contract. This smart contract includes details of the consumption shift plan, trigger conditions for execution, and specific execution steps. The generated contract is notified to the user via the terminal, and the user reviews and approves the plan.

[0041] Once a device receives user approval, it feeds that information back to the server. The server then automatically adjusts the device's operating schedule based on the approved plan.

[0042] As a concrete example, let's consider implementation in a local commercial facility. The facility's server predicts a consumption spike every Monday morning. Generating AI creates a scenario to shift the operation of air conditioning and elevators to a different time and incorporates it into a smart contract. A terminal notifies the facility manager of this plan, and once the manager approves, the terminal automatically executes the schedule change. As a result, peak energy consumption is effectively reduced, leading to lower operating costs.

[0043] The following describes the processing flow.

[0044] Step 1:

[0045] The server collects energy consumption data from smart meters and IoT sensors. This data includes hourly consumption, environmental conditions (temperature, humidity, etc.), and the operating status of the equipment.

[0046] Step 2:

[0047] The server preprocesses the collected data. Data quality is ensured by imputing missing values ​​using the nearest neighbor mean method and detecting and removing outliers using statistical methods.

[0048] Step 3:

[0049] The server uses a recurrent neural network (RNN) or long-short-term memory (LSTM) to predict future energy consumption spikes from preprocessed data. It trains the model and outputs time-series prediction results.

[0050] Step 4:

[0051] Based on the prediction results, the server uses generative AI to create multiple consumption scenarios. These generated scenarios include strategies for shifting consumption during peak hours and reducing load.

[0052] Step 5:

[0053] The server evaluates the effectiveness of each consumption scenario and selects the most effective one. Evaluation criteria include consumption reduction, cost-effectiveness, and feasibility.

[0054] Step 6:

[0055] The server automatically generates smart contracts based on the selected scenario. These contracts include details of the consumption shift plan, execution conditions, and specific triggers.

[0056] Step 7:

[0057] The device will notify the user of the details of their smart contract and prompt them to review and approve their consumption shift plan. Notifications will be sent via app push notifications or email.

[0058] Step 8:

[0059] When a user approves a shift plan via their terminal, that information is fed back to the server. Users can also modify the plan.

[0060] Step 9:

[0061] The server automatically adjusts the equipment's operating schedule via the terminal based on the approved plan. This ensures that consumption is leveled out over specified times.

[0062] Step 10:

[0063] The server monitors consumption data after the planned execution, evaluates the results, and uses them to optimize future plans.

[0064] (Example 1)

[0065] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0066] Current energy management systems lack the means to effectively predict and smooth out rapid fluctuations and spikes in energy consumption. This leads to increased facility operating costs and hinders efficient energy use. Furthermore, providing optimal energy consumption patterns for each facility and end-user presents a challenge.

[0067] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0068] In this invention, the server includes means for collecting information related to energy consumption and preprocessing it to improve data quality; means for predicting future energy consumption patterns using a time-series forecasting model; and means for generating multiple consumption adjustment scenarios using a generative model, evaluating the effects of each, and selecting the optimal scenario. This enables the automation and optimization of plans to predict and smooth out rapid fluctuations in energy consumption.

[0069] "Information related to energy consumption" refers to all data related to energy use, such as consumption per hour, the impact of the external environment, and the operating status of equipment.

[0070] "Methods for preprocessing data to improve its quality" refers to the process of imputing missing values ​​and removing outliers in order to ensure the accuracy of collected data.

[0071] A "time series forecasting model" is an algorithm used to learn how data changes over time and predict future data patterns.

[0072] A "generative model" is a technique used to create new information from existing data and construct multiple scenarios.

[0073] A "consumption adjustment scenario" refers to a specific consumption plan developed to optimize energy usage patterns.

[0074] A "contract" refers to an official document that details the plan and implementation conditions for energy consumption.

[0075] An "information transmission device" refers to hardware or software that provides information to users and enables interaction.

[0076] "End users" refer to individuals or companies that use this system and ultimately benefit from it.

[0077] "Methods for optimizing equipment operation schedules" refers to the process of adjusting the timing of equipment operation in order to improve energy efficiency.

[0078] This invention relates to a system designed for the purpose of improving energy consumption efficiency. The system consists mainly of a server, terminals, and users, and each component works in cooperation to predict and adjust energy consumption.

[0079] The server collects data in real time through smart meters and IoT sensors and stores it in a database. This data includes hourly energy consumption, weather conditions, and facility operating status. The server cleanses the obtained data to improve its quality. This includes imputing missing values ​​and removing outliers.

[0080] Next, the server uses time-series prediction models such as recurrent neural networks (RNNs) and long-short-term memory (LSTMs) to predict future energy consumption patterns. This makes it possible to identify consumption spikes in advance and take countermeasures.

[0081] Using a generative AI model, the server generates multiple consumption adjustment scenarios. Each scenario attempts to level out energy consumption based on various adjustment proposals. The server evaluates the effectiveness of each scenario and selects the optimal one. The selected scenario is automatically generated in the form of a contract as a concrete, actionable consumption adjustment plan.

[0082] The terminal notifies the user of this generated contract. The notification is made through a visualized interface, allowing the user to review the details of the adjustment plan. Once the user approves, that information is fed back from the terminal to the server.

[0083] The server improves energy efficiency by automatically adjusting and executing the equipment's operating schedule based on an approved consumption adjustment plan.

[0084] As a concrete example, in one commercial facility, a server typically predicts a consumption spike on Mondays. Using a generation AI, prompts such as, "Predict the peak time for energy consumption over the next week and propose an optimal consumption shift plan to level out that peak," are used. This results in a plan to level out the operation of the air conditioning system. A terminal then notifies the facility manager of this plan, and after the manager approves it, the schedule is adjusted and implemented. This achieves both peak energy consumption reduction and cost savings.

[0085] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0086] Step 1:

[0087] The server collects energy consumption data from smart meters and IoT sensors. Inputs include hourly consumption, weather conditions, and facility operating status. This information is stored in a central database. Specifically, it acquires data from each sensor via the network and updates the ever-growing dataset in real time. The output is the stored raw data.

[0088] Step 2:

[0089] The server preprocesses the collected data. In this step, missing values ​​are imputed and outliers are removed. The input is the raw data obtained in step 1. Data cleansing techniques are applied, and normalization and filtering are performed to shape the data into an accurate dataset. The output is the preprocessed, clean data.

[0090] Step 3:

[0091] The server predicts energy consumption patterns using a recurrent neural network (RNN) and long-short-term memory (LSTM). The input for this step is the clean data obtained in step 2. Time series analysis is performed to train the model and detect future consumption spikes. Prediction results are generated and output. The output is future energy consumption prediction data.

[0092] Step 4:

[0093] The server generates consumption adjustment scenarios using a generative AI model. In this step, it uses the predicted data obtained in step 3 as input to generate various adjustment options. The generative AI model evaluates the effectiveness of the scenarios using prompt messages. The output is the evaluated multiple scenarios.

[0094] Step 5:

[0095] The server selects the most efficient scenario from the generated scenarios. This step involves comparing multiple input scenarios and determining the optimal scenario based on criteria such as energy efficiency, cost, and feasibility. The output is the selected best-performing scenario.

[0096] Step 6:

[0097] The server generates a smart contract based on the selected scenario. The input for this step is the best-case scenario determined in step 5. It automatically generates a contract that includes details of the specific execution steps, time, and conditions. The output is the generated contract.

[0098] Step 7:

[0099] The terminal notifies the user of the generated contract. The input for this step is the contract generated in step 6. The terminal displays the contract details through the user interface and requests approval. The user reviews the contract and inputs whether they approve or reject it into the terminal. The output is the user's approval result.

[0100] Step 8:

[0101] The server adjusts and executes the equipment's operating schedule based on the user's approval. This step uses the approval result obtained in step 7 as input. The server sends instructions to the equipment to implement specific changes to its operating schedule. The output is the adjusted equipment operating schedule.

[0102] (Application Example 1)

[0103] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0104] Urban energy consumption presents numerous problems during peak hours. For example, it can compromise the stability of the power supply and increase energy costs. Furthermore, inefficient energy use can lead to increased environmental impact. To address these challenges, methods are needed to predict and efficiently adjust urban-wide energy consumption in real time.

[0105] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0106] In this invention, the server includes means for collecting and preprocessing energy usage information, means for predicting increases in energy consumption using a recurrent neural network and long-term short-term memory, and means for generating multiple consumption plans using generative AI, evaluating their effectiveness, and selecting the optimal plan. This enables efficient optimization of urban energy use, mitigation of the impact of energy spikes, and reduction of environmental impact and costs.

[0107] "Energy usage information" refers to data showing the energy consumption situation in cities and facilities, and specifically includes indicators related to electricity consumption and usage.

[0108] "Preprocessing" is the process of removing noise and invalid data from collected energy usage information and preparing it into an analyzable format.

[0109] A "recurrent neural network" is a type of machine learning model designed to handle time-series data, possessing the ability to predict future data patterns while considering past information.

[0110] "Long-term short-term memory" is a part of recurrent neural networks, a special type of neural network with a structure that enables predictions that take long-term dependencies into account.

[0111] A "consumption plan" is a plan that outlines a specific implementation schedule for adjusting the timing and amount of energy consumption in order to efficiently manage energy use.

[0112] An "automated contract" is a contract that is automatically executed under specific conditions based on a generated consumption plan, and is a mechanism for programmatically adjusting energy use.

[0113] A "communication terminal" is an electronic device used by users to check and approve energy consumption plans, such as smartphones and tablet devices.

[0114] "User" refers to an individual or organization that has the authority to receive, approve, or modify information regarding energy consumption plans.

[0115] "Equipment operation plan" refers to a detailed schedule for operating equipment to optimize energy consumption in facilities and urban infrastructure.

[0116] "Optimizing energy use across an entire city" refers to the process of balancing energy supply and demand on a city-wide scale, and adjusting energy to use it efficiently and without waste.

[0117] This invention is realized through the cooperation of servers, communication terminals, and users in order to optimize energy consumption across the entire city.

[0118] The server collects energy usage information in real time from smart meters and IoT sensors. The collected data is preprocessed to impart missing information and remove outliers. During this process, programming languages ​​such as Python are used to improve data quality. The server also uses recurrent neural networks and long-term short-term memory (LSTM) to predict future increases in energy consumption. This prediction is performed using machine learning libraries such as TENSORFLOW®.

[0119] Using a generation AI, multiple consumption plans are generated on the server. These plans are evaluated for their effectiveness, and the optimal plan is selected. Based on the selected plan, an automated contract is generated. This contract includes the specific elements of the consumption shift plan, the timing of its execution, and the conditions.

[0120] The communication terminal is responsible for notifying users of the generated consumption shift plan. Users receive the notification via smartphone or tablet and approve or modify the plan. After approval, the server automatically adjusts and executes the equipment operation plan based on the user's approval.

[0121] As a concrete example, the server generates a scenario that suggests adjusting the air conditioning schedule and optimizing lighting brightness in response to a predicted energy consumption spike on Monday morning. This scenario is then automatically executed after user approval.

[0122] An example of a prompt for a generative AI model is, "Predict the expected energy consumption spikes for the next week and generate consumption shift scenarios to optimize energy use." This prompt prepares the AI ​​model to generate the necessary scenarios.

[0123] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0124] Step 1:

[0125] The server collects energy usage information from smart meters and IoT sensors. The input is real-time consumption data, and the output is a dataset containing this data. The server periodically retrieves data from these devices via an API.

[0126] Step 2:

[0127] The server preprocesses the collected energy usage information. It performs data imputation and removes outliers to generate a high-quality dataset. The input is the raw data obtained in step 1, and the output is the preprocessed data. The server uses Python libraries to perform noise reduction and data cleaning.

[0128] Step 3:

[0129] The server uses pre-processed data to predict increases in energy consumption based on recurrent neural networks and long short-term memory (LSTM) models. The input is pre-processed data, and the output is the predicted consumption pattern. The server uses TensorFlow to run the predictive model and identify future consumption spikes.

[0130] Step 4:

[0131] The server generates multiple consumption planning scenarios using generative AI. The input is the predicted consumption pattern, and the output is the generated consumption planning scenario. The server sends a prompt to the generative AI model: "Predict the expected energy consumption spikes for the next week and generate consumption shift scenarios to optimize energy use," and then generates the scenarios.

[0132] Step 5:

[0133] The server evaluates the generated consumption plan scenarios and selects the optimal one. The input is multiple scenarios, and the output is the selected optimal scenario. The server evaluates the efficiency of each scenario and automatically determines the plan best suited to the objective.

[0134] Step 6:

[0135] The server generates automated contracts based on the selected consumption scenario. The input is an optimized scenario, and the output is an executable automated contract. The server uses a contract generation engine to create smart contracts and records the contract details.

[0136] Step 7:

[0137] The terminal notifies the user of the generated consumption shift plan. The input is an automated contract received from the server, and the output is a notification to the user. The terminal displays the details of this contract and notifies the user through the UI / UX so that they can confirm it.

[0138] Step 8:

[0139] The user approves the consumption shift plan through the terminal. The input is the notified consumption scenario, and the output is an approval or modification request. The user reviews the information provided on the terminal and either approves it or instructs the plan to be modified as needed.

[0140] Step 9:

[0141] The server adjusts and executes the equipment's operational plan based on user approval. The input is an approved automated contract, and the output is the adjusted execution schedule. The server signals the equipment control system to implement the schedule in real time.

[0142] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0143] This invention integrates an emotion engine into an energy consumption management system to enable prediction and management of energy consumption spikes that take into account the user's emotional state. The system aims to improve the efficiency of energy use through energy consumption data collection, preprocessing, spike prediction, consumption scenario generation, and smart contract generation and execution. In addition, by using the emotion engine, it optimizes interactions based on the user's emotions.

[0144] First, the server collects energy consumption data from smart meters and IoT sensors. The collected data is preprocessed to impute missing values ​​and remove outliers. The server then uses recurrent neural networks (RNNs) and long-short-term memory (LSTMs) to predict future energy consumption spikes.

[0145] Next, the server uses a generation AI to generate multiple consumption scenarios and evaluates their effects to select the optimal scenario. The selected scenario is automatically generated as a smart contract, and a specific consumption shift plan is formulated.

[0146] Here, the device's role is to use an emotion engine to analyze the user's emotional state and notify them of consumption shift plans in an appropriate manner based on that state. For example, if the user is feeling stressed, the notification tone can be softened or the options increased. The user's emotional state is analyzed based on factors such as voice tone, language selection, and past user response data.

[0147] When a user receives a terminal notification, they review the shift plan and approve it through an interface tailored to their emotional state. Based on the approved plan, the server automatically adjusts the equipment's operating schedule to level out consumption.

[0148] As a specific example, in one household, it was predicted that a consumption spike might occur on Friday evenings, a time when users typically experience stress. The emotion engine notified the user of a simple shift plan to reduce energy consumption during a relaxing break time, presenting flexible options. Suggestions for stress-reducing lighting and music were also offered. As a result, the user agreed to the shift plan without stress, successfully reducing energy consumption effectively.

[0149] The following describes the processing flow.

[0150] Step 1:

[0151] The server collects energy consumption data in real time from smart meters and IoT sensors. The data includes consumption, environmental conditions, and the operating status of each device.

[0152] Step 2:

[0153] The server preprocesses the collected data. Missing data is imputed using the nearest neighbor mean method, and outliers are detected and removed using statistical methods. This ensures the accuracy of the data.

[0154] Step 3:

[0155] The server uses recurrent neural networks (RNNs) and long-short-term memory (LSTMs) to analyze pre-processed data and predict the occurrence of energy consumption spikes. This allows for an understanding of future consumption patterns.

[0156] Step 4:

[0157] The server utilizes generation AI to generate multiple consumption scenarios to mitigate predicted consumption spikes. It evaluates the consumption reduction and cost-effectiveness of each scenario and selects the optimal one.

[0158] Step 5:

[0159] The server automatically generates smart contracts based on the selected consumption scenarios. These contracts include specific consumption shift plans, execution conditions, and triggers.

[0160] Step 6:

[0161] The device uses an emotion engine to analyze the user's emotional state from facial expressions, voice tone, and input text. Based on the analysis results, it notifies the user of a consumption shift plan at the optimal time and in the appropriate manner.

[0162] Step 7:

[0163] Users review the consumption shift plan presented through their device. They have the option to approve or modify the plan through a customized interface that responds to their emotional state.

[0164] Step 8:

[0165] After user approval, the device feeds that information back to the server. The server automatically adjusts the device's operating schedule according to the smart contract. This reduces the power consumption burden during peak hours.

[0166] Step 9:

[0167] The server monitors energy consumption data after execution and evaluates the effectiveness of the plan. The evaluation results are used to predict future consumption and optimize the plan.

[0168] (Example 2)

[0169] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0170] When optimizing energy consumption, mechanically shifting consumption without considering the emotional state of users can lead to the accumulation of user dissatisfaction and stress. Furthermore, conventional energy spike prediction systems have difficulty in designing flexible scenarios that reflect emotional changes, which hinders optimal energy consumption management.

[0171] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0172] In this invention, the server includes means for collecting energy usage information and processing the data, means for predicting energy spikes using a recurrent neural network or a long-term short-term memory network, and means for generating multiple consumption scenarios using a generative model, evaluating their effectiveness, and selecting the optimal scenario. This enables flexible energy consumption management that takes into account the user's emotional state.

[0173] "Energy usage information" refers to data on energy sources such as electricity, gas, and water used by consumers. By collecting and analyzing this data, it forms the basis for understanding consumption patterns.

[0174] "Data processing" refers to a series of techniques that involve collecting raw data and then performing analysis, cleaning, and format conversion to prepare it for subsequent analysis and prediction.

[0175] A "regressive neural network" is an artificial intelligence model that specializes in ordinal and temporal data and is used to predict future outcomes from past data.

[0176] A "long-term short-term memory network" is a type of recurrent neural network that processes short-term data while considering the dependencies between long-term historical data, thereby achieving more accurate predictions.

[0177] A "generative model" is an artificial intelligence technology that can learn from large amounts of input data and create new data, and is used to generate various scenarios and content.

[0178] A "consumption scenario" is an assumption about different situations and patterns of energy consumption, and it forms the basis for developing plans to address predicted energy consumption spikes.

[0179] A "smart contract" is an automated contract based on blockchain technology, which includes a program that automatically executes when certain conditions are met.

[0180] "User emotional state" refers to the user's psychological state and stress level, and analyzing this provides a basis for providing the most suitable notifications and services to the user.

[0181] A mode for carrying out the present invention relates to a system for predicting and optimizing energy consumption. This system functions through the interaction of a server, terminals, and users.

[0182] First, the server collects energy usage information from smart meters and IoT sensors. This information is processed using software such as Python and Pandas. This processing includes imputing missing information and removing outliers, enabling reliable data analysis. Next, the server uses machine learning frameworks such as TensorFlow to run recurrent neural networks and long-term short-term memory networks to predict future energy consumption spikes. Based on these predictions, generative models are used to generate optimal consumption scenarios.

[0183] The generated scenarios are then automatically incorporated as smart contracts. These smart contracts, based on blockchain technology, help automate energy management by being executed when specific consumption conditions are met.

[0184] The device is equipped with an emotion engine that analyzes the user's voice and text data to evaluate their emotional state. For example, it uses a natural language processing library to perform voice tone analysis and text sentiment analysis, and adjusts the notification method based on the results. This allows for the delivery of update information in a way that is less stressful for the user.

[0185] Users can review and approve consumption shift plans by receiving notifications from their devices. For example, users can specify their settings by using prompts to input into the generating AI model, such as "Please propose a shift plan to reduce energy consumption by 10% over the weekend."

[0186] Ultimately, the server adjusts the device's operating schedule based on user approval, resulting in efficient energy consumption. In this way, the present invention enables flexible energy management that responds to the user's emotional state by combining data analysis and generative AI technology.

[0187] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0188] Step 1:

[0189] The server collects energy usage information from smart meters and IoT sensors. The input is real-time data from various sensors, including date and time, consumption amount, and consumption location. The server uses Python and Pandas to store this data in a database, impute missing information, and remove outliers. The output is a cleaned dataset, which is then used for subsequent prediction processes.

[0190] Step 2:

[0191] The server predicts future spikes in energy consumption using a pre-processed dataset. The input is formatted energy data. The server uses TensorFlow to run recurrent neural networks and long-term short-term memory networks, outputting the probability of future consumption spikes numerically. The output is a predicted value for energy spikes at a specific time.

[0192] Step 3:

[0193] The server uses a generative AI model to generate multiple consumption scenarios. The input is the predicted consumption spike value. The server generates a prompt message, "Generate a plan to reduce energy consumption by 5%", and uses this prompt to run the generative AI model. The output is a set of various energy consumption scenarios, each of which is evaluated for effectiveness.

[0194] Step 4:

[0195] The server simulates the effects of the generated scenarios. The input is a set of multiple scenarios. The server evaluates the energy costs and efficiency under different conditions for each scenario and selects the optimal scenario. The output is the consumption scenario deemed optimal and its associated data.

[0196] Step 5:

[0197] The server automatically generates a smart contract based on the selected scenario. The input is the optimal consumption scenario. The server uses blockchain technology to create a contract that automatically executes under the specified conditions. The output is the smart contract, which is stored as part of the automated execution process.

[0198] Step 6:

[0199] The device analyzes the user's emotional state through an emotion engine. Input consists of the user's voice tone and text data. The device uses a natural language processing library to decode the emotional state from the voice and text, and selects the optimal notification method based on the results. The output is a customized notification tailored to the user's emotional state.

[0200] Step 7:

[0201] Users receive notifications via their devices to review their consumption shift plans. Input is a customized notification from the device. Users can evaluate the notification content and approve or reject the plan. Output is the user's plan approval information, which is used for the next steps.

[0202] Step 8:

[0203] The server automatically adjusts the operating schedule of the devices based on user approval. Inputs are user approval information and selected smart contracts. The server coordinates commands to household appliances through the interface, leveling energy consumption. Output is the optimized energy usage schedule.

[0204] (Application Example 2)

[0205] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0206] Conventional energy management systems relied solely on energy consumption data for prediction and management, without considering the emotional state of users. As a result, they failed to adequately address user comfort and stress reduction. This made it difficult to effectively mitigate energy consumption spikes and created situations where users were reluctant to accept the proposed plans. Consequently, efficient energy management across cities was not adequately achieved.

[0207] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0208] In this invention, the server includes means for collecting and initially processing energy usage information, means for predicting energy usage spikes using a recurrent neural network and long- and short-term memory, means for generating multiple usage scenarios using generative AI, evaluating their effectiveness, and selecting the optimal scenario, and means for adjusting notifications based on the user's emotional state. This enables the optimization of energy usage in accordance with the user's emotions, resulting in efficient management of energy consumption and improved user comfort.

[0209] "Energy usage information" refers to data obtained from smart meters and IoT sensors that shows the energy consumption status of individual consumers and devices.

[0210] "Initial processing" refers to the process of filling in missing information and removing outliers in order to make the collected energy usage information ready for analysis.

[0211] A "recurrent neural network" is a machine learning model suitable for handling time-series data, and it predicts future energy usage spikes while taking past data into consideration.

[0212] "Long-short-term memory" is a type of recurrent neural network that is a model designed to predict energy usage spikes with greater accuracy while maintaining long-term dependencies.

[0213] "Generative AI" is an artificial intelligence technology that generates multiple usage scenarios from given input data and then compares and evaluates those scenarios to derive the optimal choice.

[0214] A "self-execution contract" is a contract that is automatically executed based on selected usage scenarios, ensuring that changes in energy use are implemented.

[0215] "User's emotional state" refers to the mental state analyzed based on the user's voice, behavior, and past response patterns, and is taken into consideration by the system in order to respond accordingly.

[0216] "Notification adjustment" is a process to optimize the content and tone of notifications regarding changes in energy use, based on the user's emotional state, in order to improve their acceptability.

[0217] "Device operating time" refers to the operating schedule of home appliances and equipment that is adjusted to mitigate energy usage spikes.

[0218] 1. System Program Overview

[0219] The server collects energy usage information in real time from smart meters and IoT devices. This information is converted into an analyzable state through initial processing. Data reliability is improved by imputing missing information and removing outliers.

[0220] 2. Data Processing and Calculation

[0221] The server uses a recurrent neural network (RNN) and long-short-term memory (LSTM) to accurately predict energy usage spikes based on time-series data. The generative AI generates multiple usage scenarios to address the predicted spikes and evaluates their effectiveness. As a result, the optimal scenario is automatically generated as a self-executing contract.

[0222] 3. Feedback to users via devices

[0223] The device adjusts notification methods based on the user's emotional state. By optimizing voice tone and wording selection, and providing a flexible interface that reduces stress, it presents usage change plans in an acceptable way.

[0224] 4. Specific Examples and Prompts

[0225] For example, if a user is predicted to be prone to stress on Friday evenings, the device will notify them of a concise and flexible energy usage plan for a more relaxing time. The user can review and approve the plan through a smartphone app. Once approved, the server efficiently adjusts the device's operating time to level out energy usage. An example of a prompt message might be, "We have an optimal energy usage plan that takes your emotional state into consideration. Please review the details, which include options to help reduce stress."

[0226] In this way, the system can efficiently manage energy and enhance user comfort.

[0227] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0228] Step 1:

[0229] The server collects energy usage information from smart meters and IoT devices. The input is real-time energy consumption data, and initial processing includes imputing missing data and removing outliers. This ensures that reliable data is output and sent to the next prediction step.

[0230] Step 2:

[0231] The server uses reliable energy usage data to predict energy usage spikes using a recurrent neural network (RNN) and long-short-term memory (LSTM). The input is pre-processed time-series data, and the server performs predictive calculations based on the model, outputting predictions of future consumption spikes. These predictions are then used to generate the next scenario.

[0232] Step 3:

[0233] The server uses a generative AI model to generate multiple usage scenarios and evaluate their effectiveness. The input is predicted energy spike data, and the generative AI model evaluates the effectiveness of each scenario and selects the optimal one. This process outputs new scenarios that will be automatically executed as self-executing contracts.

[0234] Step 4:

[0235] The terminal receives usage scenarios generated as self-execution contracts and notifies the user. The input is the scenario sent from the server, and the notification content is optimized based on the user's emotional state. Specifically, the voice tone and message content are adjusted to be more acceptable to the user. The output is a notification of the energy usage change plan presented to the user.

[0236] Step 5:

[0237] The user reviews the usage plan presented on the device and indicates whether they approve or reject it. The input is optimized notification content, and user feedback is output. Approved plans are fed back to the server.

[0238] Step 6:

[0239] The server adjusts and executes the device's operating time based on the usage plan approved by the user. The input is the approved plan information, specifically adjusting the device's operation to reduce energy use during certain time periods and shift it to other time periods. The output is the optimized energy usage schedule.

[0240] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0241] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0242] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0243] [Second Embodiment]

[0244] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0245] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0246] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0247] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0248] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0249] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0250] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0251] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0252] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0253] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0254] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0255] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0256] The present invention's system supports energy consumption efficiency by using generative AI to predict future energy consumption spikes and automatically executes plans to level out consumption. This system includes a server that collects and analyzes data in real time, a terminal that notifies and approves of consumption plans, and a user that controls equipment based on the plan.

[0257] The server first collects energy consumption data from smart meters and IoT sensors. This data includes time-based consumption, weather conditions, and facility operating status. After collecting the data, the server preprocesses it to improve data quality by imputing missing values ​​and removing outliers.

[0258] Next, the server uses recurrent neural networks (RNNs) and long-short-term memory (LSTMs) to predict future energy consumption patterns. This allows it to detect consumption spikes at specific times. Based on the prediction results, the server uses generative AI to generate multiple consumption scenarios. The server then evaluates each scenario to select the most efficient consumption shift plan from among them.

[0259] The selected scenario is automatically generated as a smart contract. This smart contract includes details of the consumption shift plan, trigger conditions for execution, and specific execution steps. The generated contract is notified to the user via the terminal, and the user reviews and approves the plan.

[0260] Once a device receives user approval, it feeds that information back to the server. The server then automatically adjusts the device's operating schedule based on the approved plan.

[0261] As a concrete example, let's consider implementation in a local commercial facility. The facility's server predicts a consumption spike every Monday morning. Generating AI creates a scenario to shift the operation of air conditioning and elevators to a different time and incorporates it into a smart contract. A terminal notifies the facility manager of this plan, and once the manager approves, the terminal automatically executes the schedule change. As a result, peak energy consumption is effectively reduced, leading to lower operating costs.

[0262] The following describes the processing flow.

[0263] Step 1:

[0264] The server collects energy consumption data from smart meters and IoT sensors. This data includes hourly consumption, environmental conditions (temperature, humidity, etc.), and the operating status of the equipment.

[0265] Step 2:

[0266] The server preprocesses the collected data. Data quality is ensured by imputing missing values ​​using the nearest neighbor mean method and detecting and removing outliers using statistical methods.

[0267] Step 3:

[0268] The server uses a recurrent neural network (RNN) or long-short-term memory (LSTM) to predict future energy consumption spikes from preprocessed data. It trains the model and outputs time-series prediction results.

[0269] Step 4:

[0270] Based on the prediction results, the server uses generative AI to create multiple consumption scenarios. These generated scenarios include strategies for shifting consumption during peak hours and reducing load.

[0271] Step 5:

[0272] The server evaluates the effectiveness of each consumption scenario and selects the most effective one. Evaluation criteria include consumption reduction, cost-effectiveness, and feasibility.

[0273] Step 6:

[0274] The server automatically generates smart contracts based on the selected scenario. These contracts include details of the consumption shift plan, execution conditions, and specific triggers.

[0275] Step 7:

[0276] The terminal notifies the user of the content of the smart contract and prompts the user to confirm and approve the consumption shift plan. The notification is sent via push notifications of the app or emails.

[0277] Step 8:

[0278] When the user approves the shift plan through the terminal, the information is fed back to the server. The user can also modify the plan.

[0279] Step 9:

[0280] Based on the approved plan, the server automatically adjusts the operation schedule of the device via the terminal. As a result, the consumption is leveled at the specified time.

[0281] Step 10:

[0282] The server monitors the consumption data after the executed plan, evaluates the results, and uses them for optimization in subsequent times.

[0283] (Example 1)

[0284] Next, Example 1 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart glasses 214 are referred to as the "terminal".

[0285] In the current energy management system, there is a lack of means to effectively predict and level out rapid fluctuations or spikes in energy consumption. For this reason, the operation cost of the facility increases, and the efficient use of energy is hindered. Furthermore, there is a problem that it is also difficult to provide an optimal energy consumption pattern for each facility and end-user.

[0286] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 1 is realized by the following respective means.

[0287] In this invention, the server includes means for collecting information related to energy consumption and preprocessing it to improve data quality; means for predicting future energy consumption patterns using a time-series forecasting model; and means for generating multiple consumption adjustment scenarios using a generative model, evaluating the effects of each, and selecting the optimal scenario. This enables the automation and optimization of plans to predict and smooth out rapid fluctuations in energy consumption.

[0288] "Information related to energy consumption" refers to all data related to energy use, such as consumption per hour, the impact of the external environment, and the operating status of equipment.

[0289] "Methods for preprocessing data to improve its quality" refers to the process of imputing missing values ​​and removing outliers in order to ensure the accuracy of collected data.

[0290] A "time series forecasting model" is an algorithm used to learn how data changes over time and predict future data patterns.

[0291] A "generative model" is a technique used to create new information from existing data and construct multiple scenarios.

[0292] A "consumption adjustment scenario" refers to a specific consumption plan developed to optimize energy usage patterns.

[0293] A "contract" refers to an official document that details the plan and implementation conditions for energy consumption.

[0294] An "information transmission device" refers to hardware or software that provides information to users and enables interaction.

[0295] "End users" refer to individuals or companies that use this system and ultimately benefit from it.

[0296] "Methods for optimizing equipment operation schedules" refers to the process of adjusting the timing of equipment operation in order to improve energy efficiency.

[0297] This invention relates to a system designed for the purpose of improving energy consumption efficiency. The system consists mainly of a server, terminals, and users, and each component works in cooperation to predict and adjust energy consumption.

[0298] The server collects data in real time through smart meters and IoT sensors and stores it in a database. This data includes hourly energy consumption, weather conditions, and facility operating status. The server cleanses the obtained data to improve its quality. This includes imputing missing values ​​and removing outliers.

[0299] Next, the server uses time-series prediction models such as recurrent neural networks (RNNs) and long-short-term memory (LSTMs) to predict future energy consumption patterns. This makes it possible to identify consumption spikes in advance and take countermeasures.

[0300] Using a generative AI model, the server generates multiple consumption adjustment scenarios. Each scenario attempts to level out energy consumption based on various adjustment proposals. The server evaluates the effectiveness of each scenario and selects the optimal one. The selected scenario is automatically generated in the form of a contract as a concrete, actionable consumption adjustment plan.

[0301] The terminal notifies the user of this generated contract. The notification is made through a visualized interface, allowing the user to review the details of the adjustment plan. Once the user approves, that information is fed back from the terminal to the server.

[0302] The server improves energy efficiency by automatically adjusting and executing the equipment's operating schedule based on an approved consumption adjustment plan.

[0303] As a specific example, in a certain commercial facility, the server predicts the normal consumption spike on Mondays. By leveraging generative AI, a prompt such as "Please predict the peak hours of energy consumption for the next week and propose an optimal consumption shift plan to level off that peak." is used. As a result, a plan to level off the operation of the air conditioning system is generated. The terminal notifies the facility manager of the plan, and after the manager approves it, the schedule is adjusted and executed. This realizes peak reduction of energy consumption and cost reduction.

[0304] The flow of the specific process in Example 1 will be described using FIG. 11.

[0305] Step 1:

[0306] The server collects energy consumption data from smart meters and IoT sensors. The inputs are data such as the consumption amount per time, weather conditions, and the operating status of the facility. These pieces of information are accumulated in a central database. Specifically, data is obtained from each sensor through the network, and the continuously growing dataset is updated in real time. The output is the accumulated raw data.

[0307] Step 2:

[0308] The server preprocesses the collected data. In this step, missing values in the data are complemented and outliers are removed. The input is the raw data obtained in Step 1. By applying data cleansing techniques and performing normalization and filtering, it is shaped into an accurate dataset. The output is the preprocessed clean data.

[0309] Step 3:

[0310] The server predicts energy consumption patterns using a recurrent neural network (RNN) and long-short-term memory (LSTM). The input for this step is the clean data obtained in step 2. Time series analysis is performed to train the model and detect future consumption spikes. Prediction results are generated and output. The output is future energy consumption prediction data.

[0311] Step 4:

[0312] The server generates consumption adjustment scenarios using a generative AI model. In this step, it uses the predicted data obtained in step 3 as input to generate various adjustment options. The generative AI model evaluates the effectiveness of the scenarios using prompt messages. The output is the evaluated multiple scenarios.

[0313] Step 5:

[0314] The server selects the most efficient scenario from the generated scenarios. This step involves comparing multiple input scenarios and determining the optimal scenario based on criteria such as energy efficiency, cost, and feasibility. The output is the selected best-performing scenario.

[0315] Step 6:

[0316] The server generates a smart contract based on the selected scenario. The input for this step is the best-case scenario determined in step 5. It automatically generates a contract that includes details of the specific execution steps, time, and conditions. The output is the generated contract.

[0317] Step 7:

[0318] The terminal notifies the user of the generated contract. The input for this step is the contract generated in step 6. The terminal displays the contract details through the user interface and requests approval. The user reviews the contract and inputs whether they approve or reject it into the terminal. The output is the user's approval result.

[0319] Step 8:

[0320] The server adjusts and executes the equipment's operating schedule based on the user's approval. This step uses the approval result obtained in step 7 as input. The server sends instructions to the equipment to implement specific changes to its operating schedule. The output is the adjusted equipment operating schedule.

[0321] (Application Example 1)

[0322] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0323] Urban energy consumption presents numerous problems during peak hours. For example, it can compromise the stability of the power supply and increase energy costs. Furthermore, inefficient energy use can lead to increased environmental impact. To address these challenges, methods are needed to predict and efficiently adjust urban-wide energy consumption in real time.

[0324] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0325] In this invention, the server includes means for collecting and preprocessing energy usage information, means for predicting increases in energy consumption using a recurrent neural network and long-term short-term memory, and means for generating multiple consumption plans using generative AI, evaluating their effectiveness, and selecting the optimal plan. This enables efficient optimization of urban energy use, mitigation of the impact of energy spikes, and reduction of environmental impact and costs.

[0326] "Energy usage information" refers to data showing the energy consumption situation in cities and facilities, and specifically includes indicators related to electricity consumption and usage.

[0327] "Preprocessing" is the process of removing noise and invalid data from collected energy usage information and preparing it into an analyzable format.

[0328] A "recurrent neural network" is a type of machine learning model designed to handle time-series data, possessing the ability to predict future data patterns while considering past information.

[0329] "Long-term short-term memory" is a part of recurrent neural networks, a special type of neural network with a structure that enables predictions that take long-term dependencies into account.

[0330] A "consumption plan" is a plan that outlines a specific implementation schedule for adjusting the timing and amount of energy consumption in order to efficiently manage energy use.

[0331] An "automated contract" is a contract that is automatically executed under specific conditions based on a generated consumption plan, and is a mechanism for programmatically adjusting energy use.

[0332] A "communication terminal" is an electronic device used by users to check and approve energy consumption plans, such as smartphones and tablet devices.

[0333] "User" refers to an individual or organization that has the authority to receive, approve, or modify information regarding energy consumption plans.

[0334] "Equipment operation plan" refers to a detailed schedule for operating equipment to optimize energy consumption in facilities and urban infrastructure.

[0335] "Optimizing energy use across an entire city" refers to the process of balancing energy supply and demand on a city-wide scale, and adjusting energy to use it efficiently and without waste.

[0336] This invention is realized through the cooperation of servers, communication terminals, and users in order to optimize energy consumption across the entire city.

[0337] The server collects energy usage information in real time from smart meters and IoT sensors. The collected data is preprocessed to impute missing information and remove outliers. During this process, programming languages ​​such as Python are used to improve data quality. The server also uses recurrent neural networks and long-term short-term memory (LSTM) to predict future increases in energy consumption. This prediction is performed using machine learning libraries such as TensorFlow.

[0338] Using a generation AI, multiple consumption plans are generated on the server. These plans are evaluated for their effectiveness, and the optimal plan is selected. Based on the selected plan, an automated contract is generated. This contract includes the specific elements of the consumption shift plan, the timing of its execution, and the conditions.

[0339] The communication terminal is responsible for notifying users of the generated consumption shift plan. Users receive the notification via smartphone or tablet and approve or modify the plan. After approval, the server automatically adjusts and executes the equipment operation plan based on the user's approval.

[0340] As a concrete example, the server generates a scenario that suggests adjusting the air conditioning schedule and optimizing lighting brightness in response to a predicted energy consumption spike on Monday morning. This scenario is then automatically executed after user approval.

[0341] An example of a prompt for a generative AI model is, "Predict the expected energy consumption spikes for the next week and generate consumption shift scenarios to optimize energy use." This prompt prepares the AI ​​model to generate the necessary scenarios.

[0342] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0343] Step 1:

[0344] The server collects energy usage information from smart meters and IoT sensors. The input is real-time consumption data, and the output is a dataset containing this data. The server periodically retrieves data from these devices via an API.

[0345] Step 2:

[0346] The server preprocesses the collected energy usage information. It performs data imputation and removes outliers to generate a high-quality dataset. The input is the raw data obtained in step 1, and the output is the preprocessed data. The server uses Python libraries to perform noise reduction and data cleaning.

[0347] Step 3:

[0348] The server uses pre-processed data to predict increases in energy consumption based on recurrent neural networks and long short-term memory (LSTM) models. The input is pre-processed data, and the output is the predicted consumption pattern. The server uses TensorFlow to run the predictive model and identify future consumption spikes.

[0349] Step 4:

[0350] The server generates multiple consumption planning scenarios using generative AI. The input is the predicted consumption pattern, and the output is the generated consumption planning scenario. The server sends a prompt to the generative AI model: "Predict the expected energy consumption spikes for the next week and generate consumption shift scenarios to optimize energy use," and then generates the scenarios.

[0351] Step 5:

[0352] The server evaluates the generated consumption plan scenarios and selects the optimal one. The input is multiple scenarios, and the output is the selected optimal scenario. The server evaluates the efficiency of each scenario and automatically determines the plan best suited to the objective.

[0353] Step 6:

[0354] The server generates automated contracts based on the selected consumption scenario. The input is an optimized scenario, and the output is an executable automated contract. The server uses a contract generation engine to create smart contracts and records the contract details.

[0355] Step 7:

[0356] The terminal notifies the user of the generated consumption shift plan. The input is an automated contract received from the server, and the output is a notification to the user. The terminal displays the details of this contract and notifies the user through the UI / UX so that they can confirm it.

[0357] Step 8:

[0358] The user approves the consumption shift plan through the terminal. The input is the notified consumption scenario, and the output is an approval or modification request. The user reviews the information provided on the terminal and either approves it or instructs the plan to be modified as needed.

[0359] Step 9:

[0360] The server adjusts and executes the equipment's operational plan based on user approval. The input is an approved automated contract, and the output is the adjusted execution schedule. The server signals the equipment control system to implement the schedule in real time.

[0361] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0362] This invention integrates an emotion engine into an energy consumption management system to enable prediction and management of energy consumption spikes that take into account the user's emotional state. The system aims to improve the efficiency of energy use through energy consumption data collection, preprocessing, spike prediction, consumption scenario generation, and smart contract generation and execution. In addition, by using the emotion engine, it optimizes interactions based on the user's emotions.

[0363] First, the server collects energy consumption data from smart meters and IoT sensors. The collected data is preprocessed to impute missing values ​​and remove outliers. The server then uses recurrent neural networks (RNNs) and long-short-term memory (LSTMs) to predict future energy consumption spikes.

[0364] Next, the server uses a generation AI to generate multiple consumption scenarios and evaluates their effects to select the optimal scenario. The selected scenario is automatically generated as a smart contract, and a specific consumption shift plan is formulated.

[0365] Here, the device's role is to use an emotion engine to analyze the user's emotional state and notify them of consumption shift plans in an appropriate manner based on that state. For example, if the user is feeling stressed, the notification tone can be softened or the options increased. The user's emotional state is analyzed based on factors such as voice tone, language selection, and past user response data.

[0366] When a user receives a terminal notification, they review the shift plan and approve it through an interface tailored to their emotional state. Based on the approved plan, the server automatically adjusts the equipment's operating schedule to level out consumption.

[0367] As a specific example, in one household, it was predicted that a consumption spike might occur on Friday evenings, a time when users typically experience stress. The emotion engine notified the user of a simple shift plan to reduce energy consumption during a relaxing break time, presenting flexible options. Suggestions for stress-reducing lighting and music were also offered. As a result, the user agreed to the shift plan without stress, successfully reducing energy consumption effectively.

[0368] The following describes the processing flow.

[0369] Step 1:

[0370] The server collects energy consumption data in real time from smart meters and IoT sensors. The data includes consumption, environmental conditions, and the operating status of each device.

[0371] Step 2:

[0372] The server preprocesses the collected data. Missing data is imputed using the nearest neighbor mean method, and outliers are detected and removed using statistical methods. This ensures the accuracy of the data.

[0373] Step 3:

[0374] The server uses recurrent neural networks (RNNs) and long-short-term memory (LSTMs) to analyze pre-processed data and predict the occurrence of energy consumption spikes. This allows for an understanding of future consumption patterns.

[0375] Step 4:

[0376] The server utilizes generation AI to generate multiple consumption scenarios to mitigate predicted consumption spikes. It evaluates the consumption reduction and cost-effectiveness of each scenario and selects the optimal one.

[0377] Step 5:

[0378] The server automatically generates smart contracts based on the selected consumption scenarios. These contracts include specific consumption shift plans, execution conditions, and triggers.

[0379] Step 6:

[0380] The device uses an emotion engine to analyze the user's emotional state from facial expressions, voice tone, and input text. Based on the analysis results, it notifies the user of a consumption shift plan at the optimal time and in the appropriate manner.

[0381] Step 7:

[0382] Users review the consumption shift plan presented through their device. They have the option to approve or modify the plan through a customized interface that responds to their emotional state.

[0383] Step 8:

[0384] After user approval, the device feeds that information back to the server. The server automatically adjusts the device's operating schedule according to the smart contract. This reduces the power consumption burden during peak hours.

[0385] Step 9:

[0386] The server monitors energy consumption data after execution and evaluates the effectiveness of the plan. The evaluation results are used to predict future consumption and optimize the plan.

[0387] (Example 2)

[0388] Next, we will describe Example 2. 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".

[0389] When optimizing energy consumption, mechanically shifting consumption without considering the emotional state of users can lead to the accumulation of user dissatisfaction and stress. Furthermore, conventional energy spike prediction systems have difficulty in designing flexible scenarios that reflect emotional changes, which hinders optimal energy consumption management.

[0390] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0391] In this invention, the server includes means for collecting energy usage information and processing the data, means for predicting energy spikes using a recurrent neural network or a long-term short-term memory network, and means for generating multiple consumption scenarios using a generative model, evaluating their effectiveness, and selecting the optimal scenario. This enables flexible energy consumption management that takes into account the user's emotional state.

[0392] "Energy usage information" refers to data on energy sources such as electricity, gas, and water used by consumers. By collecting and analyzing this data, it forms the basis for understanding consumption patterns.

[0393] "Data processing" refers to a series of techniques that involve collecting raw data and then performing analysis, cleaning, and format conversion to prepare it for subsequent analysis and prediction.

[0394] A "regressive neural network" is an artificial intelligence model that specializes in ordinal and temporal data and is used to predict future outcomes from past data.

[0395] A "long-term short-term memory network" is a type of recurrent neural network that processes short-term data while considering the dependencies between long-term historical data, thereby achieving more accurate predictions.

[0396] A "generative model" is an artificial intelligence technology that can learn from large amounts of input data and create new data, and is used to generate various scenarios and content.

[0397] A "consumption scenario" is an assumption about different situations and patterns of energy consumption, and it forms the basis for developing plans to address predicted energy consumption spikes.

[0398] A "smart contract" is an automated contract based on blockchain technology, which includes a program that automatically executes when certain conditions are met.

[0399] "User emotional state" refers to the user's psychological state and stress level, and analyzing this provides a basis for providing the most suitable notifications and services to the user.

[0400] A mode for carrying out the present invention relates to a system for predicting and optimizing energy consumption. This system functions through the interaction of a server, terminals, and users.

[0401] First, the server collects energy usage information from smart meters and IoT sensors. This information is processed using software such as Python and Pandas. This processing includes imputing missing information and removing outliers, enabling reliable data analysis. Next, the server uses machine learning frameworks such as TensorFlow to run recurrent neural networks and long-term short-term memory networks to predict future energy consumption spikes. Based on these predictions, generative models are used to generate optimal consumption scenarios.

[0402] The generated scenarios are then automatically incorporated as smart contracts. These smart contracts, based on blockchain technology, help automate energy management by being executed when specific consumption conditions are met.

[0403] The device is equipped with an emotion engine that analyzes the user's voice and text data to evaluate their emotional state. For example, it uses a natural language processing library to perform voice tone analysis and text sentiment analysis, and adjusts the notification method based on the results. This allows for the delivery of update information in a way that is less stressful for the user.

[0404] Users can review and approve consumption shift plans by receiving notifications from their devices. For example, users can specify their settings by using prompts to input into the generating AI model, such as "Please propose a shift plan to reduce energy consumption by 10% over the weekend."

[0405] Ultimately, the server adjusts the device's operating schedule based on user approval, resulting in efficient energy consumption. In this way, the present invention enables flexible energy management that responds to the user's emotional state by combining data analysis and generative AI technology.

[0406] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0407] Step 1:

[0408] The server collects energy usage information from smart meters and IoT sensors. The input is real-time data from various sensors, including date and time, consumption amount, and consumption location. The server uses Python and Pandas to store this data in a database, impute missing information, and remove outliers. The output is a cleaned dataset, which is then used for subsequent prediction processes.

[0409] Step 2:

[0410] The server predicts future spikes in energy consumption using a pre-processed dataset. The input is formatted energy data. The server uses TensorFlow to run recurrent neural networks and long-term short-term memory networks, outputting the probability of future consumption spikes numerically. The output is a predicted value for energy spikes at a specific time.

[0411] Step 3:

[0412] The server uses a generative AI model to generate multiple consumption scenarios. The input is the predicted consumption spike value. The server generates a prompt message, "Generate a plan to reduce energy consumption by 5%", and uses this prompt to run the generative AI model. The output is a set of various energy consumption scenarios, each of which is evaluated for effectiveness.

[0413] Step 4:

[0414] The server simulates the effects of the generated scenarios. The input is a set of multiple scenarios. The server evaluates the energy costs and efficiency under different conditions for each scenario and selects the optimal scenario. The output is the consumption scenario deemed optimal and its associated data.

[0415] Step 5:

[0416] The server automatically generates a smart contract based on the selected scenario. The input is the optimal consumption scenario. The server uses blockchain technology to create a contract that automatically executes under the specified conditions. The output is the smart contract, which is stored as part of the automated execution process.

[0417] Step 6:

[0418] The device analyzes the user's emotional state through an emotion engine. Input consists of the user's voice tone and text data. The device uses a natural language processing library to decode the emotional state from the voice and text, and selects the optimal notification method based on the results. The output is a customized notification tailored to the user's emotional state.

[0419] Step 7:

[0420] Users receive notifications via their devices to review their consumption shift plans. Input is a customized notification from the device. Users can evaluate the notification content and approve or reject the plan. Output is the user's plan approval information, which is used for the next steps.

[0421] Step 8:

[0422] The server automatically adjusts the operating schedule of the devices based on user approval. Inputs are user approval information and selected smart contracts. The server coordinates commands to household appliances through the interface, leveling energy consumption. Output is the optimized energy usage schedule.

[0423] (Application Example 2)

[0424] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0425] Conventional energy management systems relied solely on energy consumption data for prediction and management, without considering the emotional state of users. As a result, they failed to adequately address user comfort and stress reduction. This made it difficult to effectively mitigate energy consumption spikes and created situations where users were reluctant to accept the proposed plans. Consequently, efficient energy management across cities was not adequately achieved.

[0426] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0427] In this invention, the server includes means for collecting and initially processing energy usage information, means for predicting energy usage spikes using a recurrent neural network and long- and short-term memory, means for generating multiple usage scenarios using generative AI, evaluating their effectiveness, and selecting the optimal scenario, and means for adjusting notifications based on the user's emotional state. This enables the optimization of energy usage in accordance with the user's emotions, resulting in efficient management of energy consumption and improved user comfort.

[0428] "Energy usage information" refers to data obtained from smart meters and IoT sensors that shows the energy consumption status of individual consumers and devices.

[0429] "Initial processing" refers to the process of filling in missing information and removing outliers in order to make the collected energy usage information ready for analysis.

[0430] A "recurrent neural network" is a machine learning model suitable for handling time-series data, and it predicts future energy usage spikes while taking past data into consideration.

[0431] "Long-short-term memory" is a type of recurrent neural network that is a model designed to predict energy usage spikes with greater accuracy while maintaining long-term dependencies.

[0432] "Generative AI" is an artificial intelligence technology that generates multiple usage scenarios from given input data and then compares and evaluates those scenarios to derive the optimal choice.

[0433] A "self-execution contract" is a contract that is automatically executed based on selected usage scenarios, ensuring that changes in energy use are implemented.

[0434] "User's emotional state" refers to the mental state analyzed based on the user's voice, behavior, and past response patterns, and is taken into consideration by the system in order to respond accordingly.

[0435] "Notification adjustment" is a process to optimize the content and tone of notifications regarding changes in energy use, based on the user's emotional state, in order to improve their acceptability.

[0436] "Device operating time" refers to the operating schedule of home appliances and equipment that is adjusted to mitigate energy usage spikes.

[0437] 1. System Program Overview

[0438] The server collects energy usage information in real time from smart meters and IoT devices. This information is converted into an analyzable state through initial processing. Data reliability is improved by imputing missing information and removing outliers.

[0439] 2. Data Processing and Calculation

[0440] The server uses a recurrent neural network (RNN) and long-short-term memory (LSTM) to accurately predict energy usage spikes based on time-series data. The generative AI generates multiple usage scenarios to address the predicted spikes and evaluates their effectiveness. As a result, the optimal scenario is automatically generated as a self-executing contract.

[0441] 3. Feedback to users via devices

[0442] The device adjusts notification methods based on the user's emotional state. By optimizing voice tone and wording selection, and providing a flexible interface that reduces stress, it presents usage change plans in an acceptable way.

[0443] 4. Specific Examples and Prompts

[0444] For example, if a user is predicted to be prone to stress on Friday evenings, the device will notify them of a concise and flexible energy usage plan for a more relaxing time. The user can review and approve the plan through a smartphone app. Once approved, the server efficiently adjusts the device's operating time to level out energy usage. An example of a prompt message might be, "We have an optimal energy usage plan that takes your emotional state into consideration. Please review the details, which include options to help reduce stress."

[0445] In this way, the system can efficiently manage energy and enhance user comfort.

[0446] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0447] Step 1:

[0448] The server collects energy usage information from smart meters and IoT devices. The input is real-time energy consumption data, and initial processing includes imputing missing data and removing outliers. This ensures that reliable data is output and sent to the next prediction step.

[0449] Step 2:

[0450] The server uses reliable energy usage data to predict energy usage spikes using a recurrent neural network (RNN) and long-short-term memory (LSTM). The input is pre-processed time-series data, and the server performs predictive calculations based on the model, outputting predictions of future consumption spikes. These predictions are then used to generate the next scenario.

[0451] Step 3:

[0452] The server uses a generative AI model to generate multiple usage scenarios and evaluate their effectiveness. The input is predicted energy spike data, and the generative AI model evaluates the effectiveness of each scenario and selects the optimal one. This process outputs new scenarios that will be automatically executed as self-executing contracts.

[0453] Step 4:

[0454] The terminal receives usage scenarios generated as self-execution contracts and notifies the user. The input is the scenario sent from the server, and the notification content is optimized based on the user's emotional state. Specifically, the voice tone and message content are adjusted to be more acceptable to the user. The output is a notification of the energy usage change plan presented to the user.

[0455] Step 5:

[0456] The user reviews the usage plan presented on the device and indicates whether they approve or reject it. The input is optimized notification content, and user feedback is output. Approved plans are fed back to the server.

[0457] Step 6:

[0458] The server adjusts and executes the device's operating time based on the usage plan approved by the user. The input is the approved plan information, specifically adjusting the device's operation to reduce energy use during certain time periods and shift it to other time periods. The output is the optimized energy usage schedule.

[0459] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0460] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0461] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0462] [Third Embodiment]

[0463] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0464] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0465] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0466] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0467] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0468] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0469] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0470] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0471] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0472] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0473] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0474] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0475] The present invention's system supports energy consumption efficiency by using generative AI to predict future energy consumption spikes and automatically executes plans to level out consumption. This system includes a server that collects and analyzes data in real time, a terminal that notifies and approves of consumption plans, and a user that controls equipment based on the plan.

[0476] The server first collects energy consumption data from smart meters and IoT sensors. This data includes time-based consumption, weather conditions, and facility operating status. After collecting the data, the server preprocesses it to improve data quality by imputing missing values ​​and removing outliers.

[0477] Next, the server uses recurrent neural networks (RNNs) and long-short-term memory (LSTMs) to predict future energy consumption patterns. This allows it to detect consumption spikes at specific times. Based on the prediction results, the server uses generative AI to generate multiple consumption scenarios. The server then evaluates each scenario to select the most efficient consumption shift plan from among them.

[0478] The selected scenario is automatically generated as a smart contract. This smart contract includes details of the consumption shift plan, trigger conditions for execution, and specific execution steps. The generated contract is notified to the user via the terminal, and the user reviews and approves the plan.

[0479] Once a device receives user approval, it feeds that information back to the server. The server then automatically adjusts the device's operating schedule based on the approved plan.

[0480] As a concrete example, let's consider implementation in a local commercial facility. The facility's server predicts a consumption spike every Monday morning. Generating AI creates a scenario to shift the operation of air conditioning and elevators to a different time and incorporates it into a smart contract. A terminal notifies the facility manager of this plan, and once the manager approves, the terminal automatically executes the schedule change. As a result, peak energy consumption is effectively reduced, leading to lower operating costs.

[0481] The following describes the processing flow.

[0482] Step 1:

[0483] The server collects energy consumption data from smart meters and IoT sensors. This data includes hourly consumption, environmental conditions (temperature, humidity, etc.), and the operating status of the equipment.

[0484] Step 2:

[0485] The server preprocesses the collected data. Data quality is ensured by imputing missing values ​​using the nearest neighbor mean method and detecting and removing outliers using statistical methods.

[0486] Step 3:

[0487] The server uses a recurrent neural network (RNN) or long-short-term memory (LSTM) to predict future energy consumption spikes from preprocessed data. It trains the model and outputs time-series prediction results.

[0488] Step 4:

[0489] Based on the prediction results, the server uses generative AI to create multiple consumption scenarios. These generated scenarios include strategies for shifting consumption during peak hours and reducing load.

[0490] Step 5:

[0491] The server evaluates the effectiveness of each consumption scenario and selects the most effective one. Evaluation criteria include consumption reduction, cost-effectiveness, and feasibility.

[0492] Step 6:

[0493] The server automatically generates smart contracts based on the selected scenario. These contracts include details of the consumption shift plan, execution conditions, and specific triggers.

[0494] Step 7:

[0495] The device will notify the user of the details of their smart contract and prompt them to review and approve their consumption shift plan. Notifications will be sent via app push notifications or email.

[0496] Step 8:

[0497] When a user approves a shift plan via their terminal, that information is fed back to the server. Users can also modify the plan.

[0498] Step 9:

[0499] The server automatically adjusts the equipment's operating schedule via the terminal based on the approved plan. This ensures that consumption is leveled out over specified times.

[0500] Step 10:

[0501] The server monitors consumption data after the planned execution, evaluates the results, and uses them to optimize future plans.

[0502] (Example 1)

[0503] Next, we will describe Example 1. 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."

[0504] Current energy management systems lack the means to effectively predict and smooth out rapid fluctuations and spikes in energy consumption. This leads to increased facility operating costs and hinders efficient energy use. Furthermore, providing optimal energy consumption patterns for each facility and end-user presents a challenge.

[0505] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0506] In this invention, the server includes means for collecting information related to energy consumption and preprocessing it to improve data quality; means for predicting future energy consumption patterns using a time-series forecasting model; and means for generating multiple consumption adjustment scenarios using a generative model, evaluating the effects of each, and selecting the optimal scenario. This enables the automation and optimization of plans to predict and smooth out rapid fluctuations in energy consumption.

[0507] "Information related to energy consumption" refers to all data related to energy use, such as consumption per hour, the impact of the external environment, and the operating status of equipment.

[0508] "Methods for preprocessing data to improve its quality" refers to the process of imputing missing values ​​and removing outliers in order to ensure the accuracy of collected data.

[0509] A "time series forecasting model" is an algorithm used to learn how data changes over time and predict future data patterns.

[0510] A "generative model" is a technique used to create new information from existing data and construct multiple scenarios.

[0511] A "consumption adjustment scenario" refers to a specific consumption plan developed to optimize energy usage patterns.

[0512] A "contract" refers to an official document that details the plan and implementation conditions for energy consumption.

[0513] An "information transmission device" refers to hardware or software that provides information to users and enables interaction.

[0514] "End users" refer to individuals or companies that use this system and ultimately benefit from it.

[0515] "Methods for optimizing equipment operation schedules" refers to the process of adjusting the timing of equipment operation in order to improve energy efficiency.

[0516] This invention relates to a system designed for the purpose of improving energy consumption efficiency. The system consists mainly of a server, terminals, and users, and each component works in cooperation to predict and adjust energy consumption.

[0517] The server collects data in real time through smart meters and IoT sensors and stores it in a database. This data includes hourly energy consumption, weather conditions, and facility operating status. The server cleanses the obtained data to improve its quality. This includes imputing missing values ​​and removing outliers.

[0518] Next, the server uses time-series prediction models such as recurrent neural networks (RNNs) and long-short-term memory (LSTMs) to predict future energy consumption patterns. This makes it possible to identify consumption spikes in advance and take countermeasures.

[0519] Using a generative AI model, the server generates multiple consumption adjustment scenarios. Each scenario attempts to level out energy consumption based on various adjustment proposals. The server evaluates the effectiveness of each scenario and selects the optimal one. The selected scenario is automatically generated in the form of a contract as a concrete, actionable consumption adjustment plan.

[0520] The terminal notifies the user of this generated contract. The notification is made through a visualized interface, allowing the user to review the details of the adjustment plan. Once the user approves, that information is fed back from the terminal to the server.

[0521] The server improves energy efficiency by automatically adjusting and executing the equipment's operating schedule based on an approved consumption adjustment plan.

[0522] As a concrete example, in one commercial facility, a server typically predicts a consumption spike on Mondays. Using a generation AI, prompts such as, "Predict the peak time for energy consumption over the next week and propose an optimal consumption shift plan to level out that peak," are used. This results in a plan to level out the operation of the air conditioning system. A terminal then notifies the facility manager of this plan, and after the manager approves it, the schedule is adjusted and implemented. This achieves both peak energy consumption reduction and cost savings.

[0523] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0524] Step 1:

[0525] The server collects energy consumption data from smart meters and IoT sensors. Inputs include hourly consumption, weather conditions, and facility operating status. This information is stored in a central database. Specifically, it acquires data from each sensor via the network and updates the ever-growing dataset in real time. The output is the stored raw data.

[0526] Step 2:

[0527] The server preprocesses the collected data. In this step, missing values ​​are imputed and outliers are removed. The input is the raw data obtained in step 1. Data cleansing techniques are applied, and normalization and filtering are performed to shape the data into an accurate dataset. The output is the preprocessed, clean data.

[0528] Step 3:

[0529] The server predicts energy consumption patterns using a recurrent neural network (RNN) and long-short-term memory (LSTM). The input for this step is the clean data obtained in step 2. Time series analysis is performed to train the model and detect future consumption spikes. Prediction results are generated and output. The output is future energy consumption prediction data.

[0530] Step 4:

[0531] The server generates consumption adjustment scenarios using a generative AI model. In this step, it uses the predicted data obtained in step 3 as input to generate various adjustment options. The generative AI model evaluates the effectiveness of the scenarios using prompt messages. The output is the evaluated multiple scenarios.

[0532] Step 5:

[0533] The server selects the most efficient scenario from the generated scenarios. This step involves comparing multiple input scenarios and determining the optimal scenario based on criteria such as energy efficiency, cost, and feasibility. The output is the selected best-performing scenario.

[0534] Step 6:

[0535] The server generates a smart contract based on the selected scenario. The input for this step is the best-case scenario determined in step 5. It automatically generates a contract that includes details of the specific execution steps, time, and conditions. The output is the generated contract.

[0536] Step 7:

[0537] The terminal notifies the user of the generated contract. The input for this step is the contract generated in step 6. The terminal displays the contract details through the user interface and requests approval. The user reviews the contract and inputs whether they approve or reject it into the terminal. The output is the user's approval result.

[0538] Step 8:

[0539] The server adjusts and executes the equipment's operating schedule based on the user's approval. This step uses the approval result obtained in step 7 as input. The server sends instructions to the equipment to implement specific changes to its operating schedule. The output is the adjusted equipment operating schedule.

[0540] (Application Example 1)

[0541] Next, we will explain Application Example 1. In the following explanation, 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."

[0542] Urban energy consumption presents numerous problems during peak hours. For example, it can compromise the stability of the power supply and increase energy costs. Furthermore, inefficient energy use can lead to increased environmental impact. To address these challenges, methods are needed to predict and efficiently adjust urban-wide energy consumption in real time.

[0543] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0544] In this invention, the server includes means for collecting and preprocessing energy usage information, means for predicting increases in energy consumption using a recurrent neural network and long-term short-term memory, and means for generating multiple consumption plans using generative AI, evaluating their effectiveness, and selecting the optimal plan. This enables efficient optimization of urban energy use, mitigation of the impact of energy spikes, and reduction of environmental impact and costs.

[0545] "Energy usage information" refers to data showing the energy consumption situation in cities and facilities, and specifically includes indicators related to electricity consumption and usage.

[0546] "Preprocessing" is the process of removing noise and invalid data from collected energy usage information and preparing it into an analyzable format.

[0547] A "recurrent neural network" is a type of machine learning model designed to handle time-series data, possessing the ability to predict future data patterns while considering past information.

[0548] "Long-term short-term memory" is a part of recurrent neural networks, a special type of neural network with a structure that enables predictions that take long-term dependencies into account.

[0549] A "consumption plan" is a plan that outlines a specific implementation schedule for adjusting the timing and amount of energy consumption in order to efficiently manage energy use.

[0550] An "automated contract" is a contract that is automatically executed under specific conditions based on a generated consumption plan, and is a mechanism for programmatically adjusting energy use.

[0551] A "communication terminal" is an electronic device used by users to check and approve energy consumption plans, such as smartphones and tablet devices.

[0552] "User" refers to an individual or organization that has the authority to receive, approve, or modify information regarding energy consumption plans.

[0553] "Equipment operation plan" refers to a detailed schedule for operating equipment to optimize energy consumption in facilities and urban infrastructure.

[0554] "Optimizing energy use across an entire city" refers to the process of balancing energy supply and demand on a city-wide scale, and adjusting energy to use it efficiently and without waste.

[0555] This invention is realized through the cooperation of servers, communication terminals, and users in order to optimize energy consumption across the entire city.

[0556] The server collects energy usage information in real time from smart meters and IoT sensors. The collected data is preprocessed to impute missing information and remove outliers. During this process, programming languages ​​such as Python are used to improve data quality. The server also uses recurrent neural networks and long-term short-term memory (LSTM) to predict future increases in energy consumption. This prediction is performed using machine learning libraries such as TensorFlow.

[0557] Using a generation AI, multiple consumption plans are generated on the server. These plans are evaluated for their effectiveness, and the optimal plan is selected. Based on the selected plan, an automated contract is generated. This contract includes the specific elements of the consumption shift plan, the timing of its execution, and the conditions.

[0558] The communication terminal is responsible for notifying users of the generated consumption shift plan. Users receive the notification via smartphone or tablet and approve or modify the plan. After approval, the server automatically adjusts and executes the equipment operation plan based on the user's approval.

[0559] As a concrete example, the server generates a scenario that suggests adjusting the air conditioning schedule and optimizing lighting brightness in response to a predicted energy consumption spike on Monday morning. This scenario is then automatically executed after user approval.

[0560] An example of a prompt for a generative AI model is, "Predict the expected energy consumption spikes for the next week and generate consumption shift scenarios to optimize energy use." This prompt prepares the AI ​​model to generate the necessary scenarios.

[0561] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0562] Step 1:

[0563] The server collects energy usage information from smart meters and IoT sensors. The input is real-time consumption data, and the output is a dataset containing this data. The server periodically retrieves data from these devices via an API.

[0564] Step 2:

[0565] The server preprocesses the collected energy usage information. It performs data imputation and removes outliers to generate a high-quality dataset. The input is the raw data obtained in step 1, and the output is the preprocessed data. The server uses Python libraries to perform noise reduction and data cleaning.

[0566] Step 3:

[0567] The server uses pre-processed data to predict increases in energy consumption based on recurrent neural networks and long short-term memory (LSTM) models. The input is pre-processed data, and the output is the predicted consumption pattern. The server uses TensorFlow to run the predictive model and identify future consumption spikes.

[0568] Step 4:

[0569] The server generates multiple consumption planning scenarios using generative AI. The input is the predicted consumption pattern, and the output is the generated consumption planning scenario. The server sends a prompt to the generative AI model: "Predict the expected energy consumption spikes for the next week and generate consumption shift scenarios to optimize energy use," and then generates the scenarios.

[0570] Step 5:

[0571] The server evaluates the generated consumption plan scenarios and selects the optimal one. The input is multiple scenarios, and the output is the selected optimal scenario. The server evaluates the efficiency of each scenario and automatically determines the plan best suited to the objective.

[0572] Step 6:

[0573] The server generates automated contracts based on the selected consumption scenario. The input is an optimized scenario, and the output is an executable automated contract. The server uses a contract generation engine to create smart contracts and records the contract details.

[0574] Step 7:

[0575] The terminal notifies the user of the generated consumption shift plan. The input is an automated contract received from the server, and the output is a notification to the user. The terminal displays the details of this contract and notifies the user through the UI / UX so that they can confirm it.

[0576] Step 8:

[0577] The user approves the consumption shift plan through the terminal. The input is the notified consumption scenario, and the output is an approval or modification request. The user reviews the information provided on the terminal and either approves it or instructs the plan to be modified as needed.

[0578] Step 9:

[0579] The server adjusts and executes the equipment's operational plan based on user approval. The input is an approved automated contract, and the output is the adjusted execution schedule. The server signals the equipment control system to implement the schedule in real time.

[0580] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0581] This invention integrates an emotion engine into an energy consumption management system to enable prediction and management of energy consumption spikes that take into account the user's emotional state. The system aims to improve the efficiency of energy use through energy consumption data collection, preprocessing, spike prediction, consumption scenario generation, and smart contract generation and execution. In addition, by using the emotion engine, it optimizes interactions based on the user's emotions.

[0582] First, the server collects energy consumption data from smart meters and IoT sensors. The collected data is preprocessed to impute missing values ​​and remove outliers. The server then uses recurrent neural networks (RNNs) and long-short-term memory (LSTMs) to predict future energy consumption spikes.

[0583] Next, the server uses a generation AI to generate multiple consumption scenarios and evaluates their effects to select the optimal scenario. The selected scenario is automatically generated as a smart contract, and a specific consumption shift plan is formulated.

[0584] Here, the device's role is to use an emotion engine to analyze the user's emotional state and notify them of consumption shift plans in an appropriate manner based on that state. For example, if the user is feeling stressed, the notification tone can be softened or the options increased. The user's emotional state is analyzed based on factors such as voice tone, language selection, and past user response data.

[0585] When a user receives a terminal notification, they review the shift plan and approve it through an interface tailored to their emotional state. Based on the approved plan, the server automatically adjusts the equipment's operating schedule to level out consumption.

[0586] As a specific example, in one household, it was predicted that a consumption spike might occur on Friday evenings, a time when users typically experience stress. The emotion engine notified the user of a simple shift plan to reduce energy consumption during a relaxing break time, presenting flexible options. Suggestions for stress-reducing lighting and music were also offered. As a result, the user agreed to the shift plan without stress, successfully reducing energy consumption effectively.

[0587] The following describes the processing flow.

[0588] Step 1:

[0589] The server collects energy consumption data in real time from smart meters and IoT sensors. The data includes consumption, environmental conditions, and the operating status of each device.

[0590] Step 2:

[0591] The server preprocesses the collected data. Missing data is imputed using the nearest neighbor mean method, and outliers are detected and removed using statistical methods. This ensures the accuracy of the data.

[0592] Step 3:

[0593] The server uses recurrent neural networks (RNNs) and long-short-term memory (LSTMs) to analyze pre-processed data and predict the occurrence of energy consumption spikes. This allows for an understanding of future consumption patterns.

[0594] Step 4:

[0595] The server utilizes generation AI to generate multiple consumption scenarios to mitigate predicted consumption spikes. It evaluates the consumption reduction and cost-effectiveness of each scenario and selects the optimal one.

[0596] Step 5:

[0597] The server automatically generates smart contracts based on the selected consumption scenarios. These contracts include specific consumption shift plans, execution conditions, and triggers.

[0598] Step 6:

[0599] The device uses an emotion engine to analyze the user's emotional state from facial expressions, voice tone, and input text. Based on the analysis results, it notifies the user of a consumption shift plan at the optimal time and in the appropriate manner.

[0600] Step 7:

[0601] Users review the consumption shift plan presented through their device. They have the option to approve or modify the plan through a customized interface that responds to their emotional state.

[0602] Step 8:

[0603] After user approval, the device feeds that information back to the server. The server automatically adjusts the device's operating schedule according to the smart contract. This reduces the power consumption burden during peak hours.

[0604] Step 9:

[0605] The server monitors energy consumption data after execution and evaluates the effectiveness of the plan. The evaluation results are used to predict future consumption and optimize the plan.

[0606] (Example 2)

[0607] Next, we will describe Example 2. 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."

[0608] When optimizing energy consumption, mechanically shifting consumption without considering the emotional state of users can lead to the accumulation of user dissatisfaction and stress. Furthermore, conventional energy spike prediction systems have difficulty in designing flexible scenarios that reflect emotional changes, which hinders optimal energy consumption management.

[0609] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0610] In this invention, the server includes means for collecting energy usage information and processing the data, means for predicting energy spikes using a recurrent neural network or a long-term short-term memory network, and means for generating multiple consumption scenarios using a generative model, evaluating their effectiveness, and selecting the optimal scenario. This enables flexible energy consumption management that takes into account the user's emotional state.

[0611] "Energy usage information" refers to data on energy sources such as electricity, gas, and water used by consumers. By collecting and analyzing this data, it forms the basis for understanding consumption patterns.

[0612] "Data processing" refers to a series of techniques that involve collecting raw data and then performing analysis, cleaning, and format conversion to prepare it for subsequent analysis and prediction.

[0613] A "regressive neural network" is an artificial intelligence model that specializes in ordinal and temporal data and is used to predict future outcomes from past data.

[0614] A "long-term short-term memory network" is a type of recurrent neural network that processes short-term data while considering the dependencies between long-term historical data, thereby achieving more accurate predictions.

[0615] A "generative model" is an artificial intelligence technology that can learn from large amounts of input data and create new data, and is used to generate various scenarios and content.

[0616] A "consumption scenario" is an assumption about different situations and patterns of energy consumption, and it forms the basis for developing plans to address predicted energy consumption spikes.

[0617] A "smart contract" is an automated contract based on blockchain technology, which includes a program that automatically executes when certain conditions are met.

[0618] "User emotional state" refers to the user's psychological state and stress level, and analyzing this provides a basis for providing the most suitable notifications and services to the user.

[0619] A mode for carrying out the present invention relates to a system for predicting and optimizing energy consumption. This system functions through the interaction of a server, terminals, and users.

[0620] First, the server collects energy usage information from smart meters and IoT sensors. This information is processed using software such as Python and Pandas. This processing includes imputing missing information and removing outliers, enabling reliable data analysis. Next, the server uses machine learning frameworks such as TensorFlow to run recurrent neural networks and long-term short-term memory networks to predict future energy consumption spikes. Based on these predictions, generative models are used to generate optimal consumption scenarios.

[0621] The generated scenarios are then automatically incorporated as smart contracts. These smart contracts, based on blockchain technology, help automate energy management by being executed when specific consumption conditions are met.

[0622] The device is equipped with an emotion engine that analyzes the user's voice and text data to evaluate their emotional state. For example, it uses a natural language processing library to perform voice tone analysis and text sentiment analysis, and adjusts the notification method based on the results. This allows for the delivery of update information in a way that is less stressful for the user.

[0623] Users can review and approve consumption shift plans by receiving notifications from their devices. For example, users can specify their settings by using prompts to input into the generating AI model, such as "Please propose a shift plan to reduce energy consumption by 10% over the weekend."

[0624] Ultimately, the server adjusts the device's operating schedule based on user approval, resulting in efficient energy consumption. In this way, the present invention enables flexible energy management that responds to the user's emotional state by combining data analysis and generative AI technology.

[0625] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0626] Step 1:

[0627] The server collects energy usage information from smart meters and IoT sensors. The input is real-time data from various sensors, including date and time, consumption amount, and consumption location. The server uses Python and Pandas to store this data in a database, impute missing information, and remove outliers. The output is a cleaned dataset, which is then used for subsequent prediction processes.

[0628] Step 2:

[0629] The server predicts future spikes in energy consumption using a pre-processed dataset. The input is formatted energy data. The server uses TensorFlow to run recurrent neural networks and long-term short-term memory networks, outputting the probability of future consumption spikes numerically. The output is a predicted value for energy spikes at a specific time.

[0630] Step 3:

[0631] The server uses a generative AI model to generate multiple consumption scenarios. The input is the predicted consumption spike value. The server generates a prompt message, "Generate a plan to reduce energy consumption by 5%", and uses this prompt to run the generative AI model. The output is a set of various energy consumption scenarios, each of which is evaluated for effectiveness.

[0632] Step 4:

[0633] The server simulates the effects of the generated scenarios. The input is a set of multiple scenarios. The server evaluates the energy costs and efficiency under different conditions for each scenario and selects the optimal scenario. The output is the consumption scenario deemed optimal and its associated data.

[0634] Step 5:

[0635] The server automatically generates a smart contract based on the selected scenario. The input is the optimal consumption scenario. The server uses blockchain technology to create a contract that automatically executes under the specified conditions. The output is the smart contract, which is stored as part of the automated execution process.

[0636] Step 6:

[0637] The device analyzes the user's emotional state through an emotion engine. Input consists of the user's voice tone and text data. The device uses a natural language processing library to decode the emotional state from the voice and text, and selects the optimal notification method based on the results. The output is a customized notification tailored to the user's emotional state.

[0638] Step 7:

[0639] Users receive notifications via their devices to review their consumption shift plans. Input is a customized notification from the device. Users can evaluate the notification content and approve or reject the plan. Output is the user's plan approval information, which is used for the next steps.

[0640] Step 8:

[0641] The server automatically adjusts the operating schedule of the devices based on user approval. Inputs are user approval information and selected smart contracts. The server coordinates commands to household appliances through the interface, leveling energy consumption. Output is the optimized energy usage schedule.

[0642] (Application Example 2)

[0643] Next, we will explain application example 2. In the following explanation, 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."

[0644] Conventional energy management systems relied solely on energy consumption data for prediction and management, without considering the emotional state of users. As a result, they failed to adequately address user comfort and stress reduction. This made it difficult to effectively mitigate energy consumption spikes and created situations where users were reluctant to accept the proposed plans. Consequently, efficient energy management across cities was not adequately achieved.

[0645] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0646] In this invention, the server includes means for collecting and initially processing energy usage information, means for predicting energy usage spikes using a recurrent neural network and long- and short-term memory, means for generating multiple usage scenarios using generative AI, evaluating their effectiveness, and selecting the optimal scenario, and means for adjusting notifications based on the user's emotional state. This enables the optimization of energy usage in accordance with the user's emotions, resulting in efficient management of energy consumption and improved user comfort.

[0647] "Energy usage information" refers to data obtained from smart meters and IoT sensors that shows the energy consumption status of individual consumers and devices.

[0648] "Initial processing" refers to the process of filling in missing information and removing outliers in order to make the collected energy usage information ready for analysis.

[0649] A "recurrent neural network" is a machine learning model suitable for handling time-series data, and it predicts future energy usage spikes while taking past data into consideration.

[0650] "Long-short-term memory" is a type of recurrent neural network that is a model designed to predict energy usage spikes with greater accuracy while maintaining long-term dependencies.

[0651] "Generative AI" is an artificial intelligence technology that generates multiple usage scenarios from given input data and then compares and evaluates those scenarios to derive the optimal choice.

[0652] A "self-execution contract" is a contract that is automatically executed based on selected usage scenarios, ensuring that changes in energy use are implemented.

[0653] "User's emotional state" refers to the mental state analyzed based on the user's voice, behavior, and past response patterns, and is taken into consideration by the system in order to respond accordingly.

[0654] "Notification adjustment" is a process to optimize the content and tone of notifications regarding changes in energy use, based on the user's emotional state, in order to improve their acceptability.

[0655] "Device operating time" refers to the operating schedule of home appliances and equipment that is adjusted to mitigate energy usage spikes.

[0656] 1. System Program Overview

[0657] The server collects energy usage information in real time from smart meters and IoT devices. This information is converted into an analyzable state through initial processing. Data reliability is improved by imputing missing information and removing outliers.

[0658] 2. Data Processing and Calculation

[0659] The server uses a recurrent neural network (RNN) and long-short-term memory (LSTM) to accurately predict energy usage spikes based on time-series data. The generative AI generates multiple usage scenarios to address the predicted spikes and evaluates their effectiveness. As a result, the optimal scenario is automatically generated as a self-executing contract.

[0660] 3. Feedback to users via devices

[0661] The device adjusts notification methods based on the user's emotional state. By optimizing voice tone and wording selection, and providing a flexible interface that reduces stress, it presents usage change plans in an acceptable way.

[0662] 4. Specific Examples and Prompts

[0663] For example, if a user is predicted to be prone to stress on Friday evenings, the device will notify them of a concise and flexible energy usage plan for a more relaxing time. The user can review and approve the plan through a smartphone app. Once approved, the server efficiently adjusts the device's operating time to level out energy usage. An example of a prompt message might be, "We have an optimal energy usage plan that takes your emotional state into consideration. Please review the details, which include options to help reduce stress."

[0664] In this way, the system can efficiently manage energy and enhance user comfort.

[0665] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0666] Step 1:

[0667] The server collects energy usage information from smart meters and IoT devices. The input is real-time energy consumption data, and initial processing includes imputing missing data and removing outliers. This ensures that reliable data is output and sent to the next prediction step.

[0668] Step 2:

[0669] The server uses reliable energy usage data to predict energy usage spikes using a recurrent neural network (RNN) and long-short-term memory (LSTM). The input is pre-processed time-series data, and the server performs predictive calculations based on the model, outputting predictions of future consumption spikes. These predictions are then used to generate the next scenario.

[0670] Step 3:

[0671] The server uses a generative AI model to generate multiple usage scenarios and evaluate their effectiveness. The input is predicted energy spike data, and the generative AI model evaluates the effectiveness of each scenario and selects the optimal one. This process outputs new scenarios that will be automatically executed as self-executing contracts.

[0672] Step 4:

[0673] The terminal receives usage scenarios generated as self-execution contracts and notifies the user. The input is the scenario sent from the server, and the notification content is optimized based on the user's emotional state. Specifically, the voice tone and message content are adjusted to be more acceptable to the user. The output is a notification of the energy usage change plan presented to the user.

[0674] Step 5:

[0675] The user reviews the usage plan presented on the device and indicates whether they approve or reject it. The input is optimized notification content, and user feedback is output. Approved plans are fed back to the server.

[0676] Step 6:

[0677] The server adjusts and executes the device's operating time based on the usage plan approved by the user. The input is the approved plan information, specifically adjusting the device's operation to reduce energy use during certain time periods and shift it to other time periods. The output is the optimized energy usage schedule.

[0678] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0679] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0680] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0681] [Fourth Embodiment]

[0682] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0683] As shown in Figure 7, the 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.

[0684] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0685] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0686] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the 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 image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0688] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0689] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0690] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0691] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0692] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0693] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0694] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0695] The present invention's system supports energy consumption efficiency by using generative AI to predict future energy consumption spikes and automatically executes plans to level out consumption. This system includes a server that collects and analyzes data in real time, a terminal that notifies and approves of consumption plans, and a user that controls equipment based on the plan.

[0696] The server first collects energy consumption data from smart meters and IoT sensors. This data includes time-based consumption, weather conditions, and facility operating status. After collecting the data, the server preprocesses it to improve data quality by imputing missing values ​​and removing outliers.

[0697] Next, the server uses recurrent neural networks (RNNs) and long-short-term memory (LSTMs) to predict future energy consumption patterns. This allows it to detect consumption spikes at specific times. Based on the prediction results, the server uses generative AI to generate multiple consumption scenarios. The server then evaluates each scenario to select the most efficient consumption shift plan from among them.

[0698] The selected scenario is automatically generated as a smart contract. This smart contract includes details of the consumption shift plan, trigger conditions for execution, and specific execution steps. The generated contract is notified to the user via the terminal, and the user reviews and approves the plan.

[0699] Once a device receives user approval, it feeds that information back to the server. The server then automatically adjusts the device's operating schedule based on the approved plan.

[0700] As a concrete example, let's consider implementation in a local commercial facility. The facility's server predicts a consumption spike every Monday morning. Generating AI creates a scenario to shift the operation of air conditioning and elevators to a different time and incorporates it into a smart contract. A terminal notifies the facility manager of this plan, and once the manager approves, the terminal automatically executes the schedule change. As a result, peak energy consumption is effectively reduced, leading to lower operating costs.

[0701] The following describes the processing flow.

[0702] Step 1:

[0703] The server collects energy consumption data from smart meters and IoT sensors. This data includes hourly consumption, environmental conditions (temperature, humidity, etc.), and the operating status of the equipment.

[0704] Step 2:

[0705] The server preprocesses the collected data. Data quality is ensured by imputing missing values ​​using the nearest neighbor mean method and detecting and removing outliers using statistical methods.

[0706] Step 3:

[0707] The server uses a recurrent neural network (RNN) or long-short-term memory (LSTM) to predict future energy consumption spikes from preprocessed data. It trains the model and outputs time-series prediction results.

[0708] Step 4:

[0709] Based on the prediction results, the server uses generative AI to create multiple consumption scenarios. These generated scenarios include strategies for shifting consumption during peak hours and reducing load.

[0710] Step 5:

[0711] The server evaluates the effectiveness of each consumption scenario and selects the most effective one. Evaluation criteria include consumption reduction, cost-effectiveness, and feasibility.

[0712] Step 6:

[0713] The server automatically generates smart contracts based on the selected scenario. These contracts include details of the consumption shift plan, execution conditions, and specific triggers.

[0714] Step 7:

[0715] The device will notify the user of the details of their smart contract and prompt them to review and approve their consumption shift plan. Notifications will be sent via app push notifications or email.

[0716] Step 8:

[0717] When a user approves a shift plan via their terminal, that information is fed back to the server. Users can also modify the plan.

[0718] Step 9:

[0719] The server automatically adjusts the equipment's operating schedule via the terminal based on the approved plan. This ensures that consumption is leveled out over specified times.

[0720] Step 10:

[0721] The server monitors consumption data after the planned execution, evaluates the results, and uses them to optimize future plans.

[0722] (Example 1)

[0723] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0724] Current energy management systems lack the means to effectively predict and smooth out rapid fluctuations and spikes in energy consumption. This leads to increased facility operating costs and hinders efficient energy use. Furthermore, providing optimal energy consumption patterns for each facility and end-user presents a challenge.

[0725] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0726] In this invention, the server includes means for collecting information related to energy consumption and preprocessing it to improve data quality; means for predicting future energy consumption patterns using a time-series forecasting model; and means for generating multiple consumption adjustment scenarios using a generative model, evaluating the effects of each, and selecting the optimal scenario. This enables the automation and optimization of plans to predict and smooth out rapid fluctuations in energy consumption.

[0727] "Information related to energy consumption" refers to all data related to energy use, such as consumption per hour, the impact of the external environment, and the operating status of equipment.

[0728] "Methods for preprocessing data to improve its quality" refers to the process of imputing missing values ​​and removing outliers in order to ensure the accuracy of collected data.

[0729] A "time series forecasting model" is an algorithm used to learn how data changes over time and predict future data patterns.

[0730] A "generative model" is a technique used to create new information from existing data and construct multiple scenarios.

[0731] A "consumption adjustment scenario" refers to a specific consumption plan developed to optimize energy usage patterns.

[0732] A "contract" refers to an official document that details the plan and implementation conditions for energy consumption.

[0733] An "information transmission device" refers to hardware or software that provides information to users and enables interaction.

[0734] "End users" refer to individuals or companies that use this system and ultimately benefit from it.

[0735] "Methods for optimizing equipment operation schedules" refers to the process of adjusting the timing of equipment operation in order to improve energy efficiency.

[0736] This invention relates to a system designed for the purpose of improving energy consumption efficiency. The system consists mainly of a server, terminals, and users, and each component works in cooperation to predict and adjust energy consumption.

[0737] The server collects data in real time through smart meters and IoT sensors and stores it in a database. This data includes hourly energy consumption, weather conditions, and facility operating status. The server cleanses the obtained data to improve its quality. This includes imputing missing values ​​and removing outliers.

[0738] Next, the server uses time-series prediction models such as recurrent neural networks (RNNs) and long-short-term memory (LSTMs) to predict future energy consumption patterns. This makes it possible to identify consumption spikes in advance and take countermeasures.

[0739] Using a generative AI model, the server generates multiple consumption adjustment scenarios. Each scenario attempts to level out energy consumption based on various adjustment proposals. The server evaluates the effectiveness of each scenario and selects the optimal one. The selected scenario is automatically generated in the form of a contract as a concrete, actionable consumption adjustment plan.

[0740] The terminal notifies the user of this generated contract. The notification is made through a visualized interface, allowing the user to review the details of the adjustment plan. Once the user approves, that information is fed back from the terminal to the server.

[0741] The server improves energy efficiency by automatically adjusting and executing the equipment's operating schedule based on an approved consumption adjustment plan.

[0742] As a concrete example, in one commercial facility, a server typically predicts a consumption spike on Mondays. Using a generation AI, prompts such as, "Predict the peak time for energy consumption over the next week and propose an optimal consumption shift plan to level out that peak," are used. This results in a plan to level out the operation of the air conditioning system. A terminal then notifies the facility manager of this plan, and after the manager approves it, the schedule is adjusted and implemented. This achieves both peak energy consumption reduction and cost savings.

[0743] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0744] Step 1:

[0745] The server collects energy consumption data from smart meters and IoT sensors. Inputs include hourly consumption, weather conditions, and facility operating status. This information is stored in a central database. Specifically, it acquires data from each sensor via the network and updates the ever-growing dataset in real time. The output is the stored raw data.

[0746] Step 2:

[0747] The server preprocesses the collected data. In this step, missing values ​​are imputed and outliers are removed. The input is the raw data obtained in step 1. Data cleansing techniques are applied, and normalization and filtering are performed to shape the data into an accurate dataset. The output is the preprocessed, clean data.

[0748] Step 3:

[0749] The server predicts energy consumption patterns using a recurrent neural network (RNN) and long-short-term memory (LSTM). The input for this step is the clean data obtained in step 2. Time series analysis is performed to train the model and detect future consumption spikes. Prediction results are generated and output. The output is future energy consumption prediction data.

[0750] Step 4:

[0751] The server generates consumption adjustment scenarios using a generative AI model. In this step, it uses the predicted data obtained in step 3 as input to generate various adjustment options. The generative AI model evaluates the effectiveness of the scenarios using prompt messages. The output is the evaluated multiple scenarios.

[0752] Step 5:

[0753] The server selects the most efficient scenario from the generated scenarios. This step involves comparing multiple input scenarios and determining the optimal scenario based on criteria such as energy efficiency, cost, and feasibility. The output is the selected best-performing scenario.

[0754] Step 6:

[0755] The server generates a smart contract based on the selected scenario. The input for this step is the best-case scenario determined in step 5. It automatically generates a contract that includes details of the specific execution steps, time, and conditions. The output is the generated contract.

[0756] Step 7:

[0757] The terminal notifies the user of the generated contract. The input for this step is the contract generated in step 6. The terminal displays the contract details through the user interface and requests approval. The user reviews the contract and inputs whether they approve or reject it into the terminal. The output is the user's approval result.

[0758] Step 8:

[0759] The server adjusts and executes the equipment's operating schedule based on the user's approval. This step uses the approval result obtained in step 7 as input. The server sends instructions to the equipment to implement specific changes to its operating schedule. The output is the adjusted equipment operating schedule.

[0760] (Application Example 1)

[0761] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0762] Urban energy consumption presents numerous problems during peak hours. For example, it can compromise the stability of the power supply and increase energy costs. Furthermore, inefficient energy use can lead to increased environmental impact. To address these challenges, methods are needed to predict and efficiently adjust urban-wide energy consumption in real time.

[0763] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0764] In this invention, the server includes means for collecting and preprocessing energy usage information, means for predicting increases in energy consumption using a recurrent neural network and long-term short-term memory, and means for generating multiple consumption plans using generative AI, evaluating their effectiveness, and selecting the optimal plan. This enables efficient optimization of urban energy use, mitigation of the impact of energy spikes, and reduction of environmental impact and costs.

[0765] "Energy usage information" refers to data showing the energy consumption situation in cities and facilities, and specifically includes indicators related to electricity consumption and usage.

[0766] "Preprocessing" is the process of removing noise and invalid data from collected energy usage information and preparing it into an analyzable format.

[0767] A "recurrent neural network" is a type of machine learning model designed to handle time-series data, possessing the ability to predict future data patterns while considering past information.

[0768] "Long-term short-term memory" is a part of recurrent neural networks, a special type of neural network with a structure that enables predictions that take long-term dependencies into account.

[0769] A "consumption plan" is a plan that outlines a specific implementation schedule for adjusting the timing and amount of energy consumption in order to efficiently manage energy use.

[0770] An "automated contract" is a contract that is automatically executed under specific conditions based on a generated consumption plan, and is a mechanism for programmatically adjusting energy use.

[0771] A "communication terminal" is an electronic device used by users to check and approve energy consumption plans, such as smartphones and tablet devices.

[0772] "User" refers to an individual or organization that has the authority to receive, approve, or modify information regarding energy consumption plans.

[0773] "Equipment operation plan" refers to a detailed schedule for operating equipment to optimize energy consumption in facilities and urban infrastructure.

[0774] "Optimizing energy use across an entire city" refers to the process of balancing energy supply and demand on a city-wide scale, and adjusting energy to use it efficiently and without waste.

[0775] This invention is realized through the cooperation of servers, communication terminals, and users in order to optimize energy consumption across the entire city.

[0776] The server collects energy usage information in real time from smart meters and IoT sensors. The collected data is preprocessed to impute missing information and remove outliers. During this process, programming languages ​​such as Python are used to improve data quality. The server also uses recurrent neural networks and long-term short-term memory (LSTM) to predict future increases in energy consumption. This prediction is performed using machine learning libraries such as TensorFlow.

[0777] Using a generation AI, multiple consumption plans are generated on the server. These plans are evaluated for their effectiveness, and the optimal plan is selected. Based on the selected plan, an automated contract is generated. This contract includes the specific elements of the consumption shift plan, the timing of its execution, and the conditions.

[0778] The communication terminal is responsible for notifying users of the generated consumption shift plan. Users receive the notification via smartphone or tablet and approve or modify the plan. After approval, the server automatically adjusts and executes the equipment operation plan based on the user's approval.

[0779] As a concrete example, the server generates a scenario that suggests adjusting the air conditioning schedule and optimizing lighting brightness in response to a predicted energy consumption spike on Monday morning. This scenario is then automatically executed after user approval.

[0780] An example of a prompt for a generative AI model is, "Predict the expected energy consumption spikes for the next week and generate consumption shift scenarios to optimize energy use." This prompt prepares the AI ​​model to generate the necessary scenarios.

[0781] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0782] Step 1:

[0783] The server collects energy usage information from smart meters and IoT sensors. The input is real-time consumption data, and the output is a dataset containing this data. The server periodically retrieves data from these devices via an API.

[0784] Step 2:

[0785] The server preprocesses the collected energy usage information. It performs data imputation and removes outliers to generate a high-quality dataset. The input is the raw data obtained in step 1, and the output is the preprocessed data. The server uses Python libraries to perform noise reduction and data cleaning.

[0786] Step 3:

[0787] The server uses pre-processed data to predict increases in energy consumption based on recurrent neural networks and long short-term memory (LSTM) models. The input is pre-processed data, and the output is the predicted consumption pattern. The server uses TensorFlow to run the predictive model and identify future consumption spikes.

[0788] Step 4:

[0789] The server generates multiple consumption planning scenarios using generative AI. The input is the predicted consumption pattern, and the output is the generated consumption planning scenario. The server sends a prompt to the generative AI model: "Predict the expected energy consumption spikes for the next week and generate consumption shift scenarios to optimize energy use," and then generates the scenarios.

[0790] Step 5:

[0791] The server evaluates the generated consumption plan scenarios and selects the optimal one. The input is multiple scenarios, and the output is the selected optimal scenario. The server evaluates the efficiency of each scenario and automatically determines the plan best suited to the objective.

[0792] Step 6:

[0793] The server generates automated contracts based on the selected consumption scenario. The input is an optimized scenario, and the output is an executable automated contract. The server uses a contract generation engine to create smart contracts and records the contract details.

[0794] Step 7:

[0795] The terminal notifies the user of the generated consumption shift plan. The input is an automated contract received from the server, and the output is a notification to the user. The terminal displays the details of this contract and notifies the user through the UI / UX so that they can confirm it.

[0796] Step 8:

[0797] The user approves the consumption shift plan through the terminal. The input is the notified consumption scenario, and the output is an approval or modification request. The user reviews the information provided on the terminal and either approves it or instructs the plan to be modified as needed.

[0798] Step 9:

[0799] The server adjusts and executes the equipment's operational plan based on user approval. The input is an approved automated contract, and the output is the adjusted execution schedule. The server signals the equipment control system to implement the schedule in real time.

[0800] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0801] This invention integrates an emotion engine into an energy consumption management system to enable prediction and management of energy consumption spikes that take into account the user's emotional state. The system aims to improve the efficiency of energy use through energy consumption data collection, preprocessing, spike prediction, consumption scenario generation, and smart contract generation and execution. In addition, by using the emotion engine, it optimizes interactions based on the user's emotions.

[0802] First, the server collects energy consumption data from smart meters and IoT sensors. The collected data is preprocessed to impute missing values ​​and remove outliers. The server then uses recurrent neural networks (RNNs) and long-short-term memory (LSTMs) to predict future energy consumption spikes.

[0803] Next, the server uses a generation AI to generate multiple consumption scenarios and evaluates their effects to select the optimal scenario. The selected scenario is automatically generated as a smart contract, and a specific consumption shift plan is formulated.

[0804] Here, the device's role is to use an emotion engine to analyze the user's emotional state and notify them of consumption shift plans in an appropriate manner based on that state. For example, if the user is feeling stressed, the notification tone can be softened or the options increased. The user's emotional state is analyzed based on factors such as voice tone, language selection, and past user response data.

[0805] When a user receives a terminal notification, they review the shift plan and approve it through an interface tailored to their emotional state. Based on the approved plan, the server automatically adjusts the equipment's operating schedule to level out consumption.

[0806] As a specific example, in one household, it was predicted that a consumption spike might occur on Friday evenings, a time when users typically experience stress. The emotion engine notified the user of a simple shift plan to reduce energy consumption during a relaxing break time, presenting flexible options. Suggestions for stress-reducing lighting and music were also offered. As a result, the user agreed to the shift plan without stress, successfully reducing energy consumption effectively.

[0807] The following describes the processing flow.

[0808] Step 1:

[0809] The server collects energy consumption data in real time from smart meters and IoT sensors. The data includes consumption, environmental conditions, and the operating status of each device.

[0810] Step 2:

[0811] The server preprocesses the collected data. Missing data is imputed using the nearest neighbor mean method, and outliers are detected and removed using statistical methods. This ensures the accuracy of the data.

[0812] Step 3:

[0813] The server uses recurrent neural networks (RNNs) and long-short-term memory (LSTMs) to analyze pre-processed data and predict the occurrence of energy consumption spikes. This allows for an understanding of future consumption patterns.

[0814] Step 4:

[0815] The server utilizes generation AI to generate multiple consumption scenarios to mitigate predicted consumption spikes. It evaluates the consumption reduction and cost-effectiveness of each scenario and selects the optimal one.

[0816] Step 5:

[0817] The server automatically generates smart contracts based on the selected consumption scenarios. These contracts include specific consumption shift plans, execution conditions, and triggers.

[0818] Step 6:

[0819] The device uses an emotion engine to analyze the user's emotional state from facial expressions, voice tone, and input text. Based on the analysis results, it notifies the user of a consumption shift plan at the optimal time and in the appropriate manner.

[0820] Step 7:

[0821] Users review the consumption shift plan presented through their device. They have the option to approve or modify the plan through a customized interface that responds to their emotional state.

[0822] Step 8:

[0823] After user approval, the device feeds that information back to the server. The server automatically adjusts the device's operating schedule according to the smart contract. This reduces the power consumption burden during peak hours.

[0824] Step 9:

[0825] The server monitors energy consumption data after execution and evaluates the effectiveness of the plan. The evaluation results are used to predict future consumption and optimize the plan.

[0826] (Example 2)

[0827] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0828] When optimizing energy consumption, mechanically shifting consumption without considering the emotional state of users can lead to the accumulation of user dissatisfaction and stress. Furthermore, conventional energy spike prediction systems have difficulty in designing flexible scenarios that reflect emotional changes, which hinders optimal energy consumption management.

[0829] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0830] In this invention, the server includes means for collecting energy usage information and processing the data, means for predicting energy spikes using a recurrent neural network or a long-term short-term memory network, and means for generating multiple consumption scenarios using a generative model, evaluating their effectiveness, and selecting the optimal scenario. This enables flexible energy consumption management that takes into account the user's emotional state.

[0831] "Energy usage information" refers to data on energy sources such as electricity, gas, and water used by consumers. By collecting and analyzing this data, it forms the basis for understanding consumption patterns.

[0832] "Data processing" refers to a series of techniques that involve collecting raw data and then performing analysis, cleaning, and format conversion to prepare it for subsequent analysis and prediction.

[0833] A "regressive neural network" is an artificial intelligence model that specializes in ordinal and temporal data and is used to predict future outcomes from past data.

[0834] A "long-term short-term memory network" is a type of recurrent neural network that processes short-term data while considering the dependencies between long-term historical data, thereby achieving more accurate predictions.

[0835] A "generative model" is an artificial intelligence technology that can learn from large amounts of input data and create new data, and is used to generate various scenarios and content.

[0836] A "consumption scenario" is an assumption about different situations and patterns of energy consumption, and it forms the basis for developing plans to address predicted energy consumption spikes.

[0837] A "smart contract" is an automated contract based on blockchain technology, which includes a program that automatically executes when certain conditions are met.

[0838] "User emotional state" refers to the user's psychological state and stress level, and analyzing this provides a basis for providing the most suitable notifications and services to the user.

[0839] A mode for carrying out the present invention relates to a system for predicting and optimizing energy consumption. This system functions through the interaction of a server, terminals, and users.

[0840] First, the server collects energy usage information from smart meters and IoT sensors. This information is processed using software such as Python and Pandas. This processing includes imputing missing information and removing outliers, enabling reliable data analysis. Next, the server uses machine learning frameworks such as TensorFlow to run recurrent neural networks and long-term short-term memory networks to predict future energy consumption spikes. Based on these predictions, generative models are used to generate optimal consumption scenarios.

[0841] The generated scenarios are then automatically incorporated as smart contracts. These smart contracts, based on blockchain technology, help automate energy management by being executed when specific consumption conditions are met.

[0842] The device is equipped with an emotion engine that analyzes the user's voice and text data to evaluate their emotional state. For example, it uses a natural language processing library to perform voice tone analysis and text sentiment analysis, and adjusts the notification method based on the results. This allows for the delivery of update information in a way that is less stressful for the user.

[0843] Users can review and approve consumption shift plans by receiving notifications from their devices. For example, users can specify their settings by using prompts to input into the generating AI model, such as "Please propose a shift plan to reduce energy consumption by 10% over the weekend."

[0844] Ultimately, the server adjusts the device's operating schedule based on user approval, resulting in efficient energy consumption. In this way, the present invention enables flexible energy management that responds to the user's emotional state by combining data analysis and generative AI technology.

[0845] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0846] Step 1:

[0847] The server collects energy usage information from smart meters and IoT sensors. The input is real-time data from various sensors, including date and time, consumption amount, and consumption location. The server uses Python and Pandas to store this data in a database, impute missing information, and remove outliers. The output is a cleaned dataset, which is then used for subsequent prediction processes.

[0848] Step 2:

[0849] The server predicts future spikes in energy consumption using a pre-processed dataset. The input is formatted energy data. The server uses TensorFlow to run recurrent neural networks and long-term short-term memory networks, outputting the probability of future consumption spikes numerically. The output is a predicted value for energy spikes at a specific time.

[0850] Step 3:

[0851] The server uses a generative AI model to generate multiple consumption scenarios. The input is the predicted consumption spike value. The server generates a prompt message, "Generate a plan to reduce energy consumption by 5%", and uses this prompt to run the generative AI model. The output is a set of various energy consumption scenarios, each of which is evaluated for effectiveness.

[0852] Step 4:

[0853] The server simulates the effects of the generated scenarios. The input is a set of multiple scenarios. The server evaluates the energy costs and efficiency under different conditions for each scenario and selects the optimal scenario. The output is the consumption scenario deemed optimal and its associated data.

[0854] Step 5:

[0855] The server automatically generates a smart contract based on the selected scenario. The input is the optimal consumption scenario. The server uses blockchain technology to create a contract that automatically executes under the specified conditions. The output is the smart contract, which is stored as part of the automated execution process.

[0856] Step 6:

[0857] The device analyzes the user's emotional state through an emotion engine. Input consists of the user's voice tone and text data. The device uses a natural language processing library to decode the emotional state from the voice and text, and selects the optimal notification method based on the results. The output is a customized notification tailored to the user's emotional state.

[0858] Step 7:

[0859] Users receive notifications via their devices to review their consumption shift plans. Input is a customized notification from the device. Users can evaluate the notification content and approve or reject the plan. Output is the user's plan approval information, which is used for the next steps.

[0860] Step 8:

[0861] The server automatically adjusts the operating schedule of the devices based on user approval. Inputs are user approval information and selected smart contracts. The server coordinates commands to household appliances through the interface, leveling energy consumption. Output is the optimized energy usage schedule.

[0862] (Application Example 2)

[0863] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0864] Conventional energy management systems relied solely on energy consumption data for prediction and management, without considering the emotional state of users. As a result, they failed to adequately address user comfort and stress reduction. This made it difficult to effectively mitigate energy consumption spikes and created situations where users were reluctant to accept the proposed plans. Consequently, efficient energy management across cities was not adequately achieved.

[0865] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0866] In this invention, the server includes means for collecting and initially processing energy usage information, means for predicting energy usage spikes using a recurrent neural network and long- and short-term memory, means for generating multiple usage scenarios using generative AI, evaluating their effectiveness, and selecting the optimal scenario, and means for adjusting notifications based on the user's emotional state. This enables the optimization of energy usage in accordance with the user's emotions, resulting in efficient management of energy consumption and improved user comfort.

[0867] "Energy usage information" refers to data obtained from smart meters and IoT sensors that shows the energy consumption status of individual consumers and devices.

[0868] "Initial processing" refers to the process of filling in missing information and removing outliers in order to make the collected energy usage information ready for analysis.

[0869] A "recurrent neural network" is a machine learning model suitable for handling time-series data, and it predicts future energy usage spikes while taking past data into consideration.

[0870] "Long-short-term memory" is a type of recurrent neural network that is a model designed to predict energy usage spikes with greater accuracy while maintaining long-term dependencies.

[0871] "Generative AI" is an artificial intelligence technology that generates multiple usage scenarios from given input data and then compares and evaluates those scenarios to derive the optimal choice.

[0872] A "self-execution contract" is a contract that is automatically executed based on selected usage scenarios, ensuring that changes in energy use are implemented.

[0873] "User's emotional state" refers to the mental state analyzed based on the user's voice, behavior, and past response patterns, and is taken into consideration by the system in order to respond accordingly.

[0874] "Notification adjustment" is a process to optimize the content and tone of notifications regarding changes in energy use, based on the user's emotional state, in order to improve their acceptability.

[0875] "Device operating time" refers to the operating schedule of home appliances and equipment that is adjusted to mitigate energy usage spikes.

[0876] 1. System Program Overview

[0877] The server collects energy usage information in real time from smart meters and IoT devices. This information is converted into an analyzable state through initial processing. Data reliability is improved by imputing missing information and removing outliers.

[0878] 2. Data Processing and Calculation

[0879] The server uses a recurrent neural network (RNN) and long-short-term memory (LSTM) to accurately predict energy usage spikes based on time-series data. The generative AI generates multiple usage scenarios to address the predicted spikes and evaluates their effectiveness. As a result, the optimal scenario is automatically generated as a self-executing contract.

[0880] 3. Feedback to users via devices

[0881] The device adjusts notification methods based on the user's emotional state. By optimizing voice tone and wording selection, and providing a flexible interface that reduces stress, it presents usage change plans in an acceptable way.

[0882] 4. Specific Examples and Prompts

[0883] For example, if a user is predicted to be prone to stress on Friday evenings, the device will notify them of a concise and flexible energy usage plan for a more relaxing time. The user can review and approve the plan through a smartphone app. Once approved, the server efficiently adjusts the device's operating time to level out energy usage. An example of a prompt message might be, "We have an optimal energy usage plan that takes your emotional state into consideration. Please review the details, which include options to help reduce stress."

[0884] In this way, the system can efficiently manage energy and enhance user comfort.

[0885] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0886] Step 1:

[0887] The server collects energy usage information from smart meters and IoT devices. The input is real-time energy consumption data, and initial processing includes imputing missing data and removing outliers. This ensures that reliable data is output and sent to the next prediction step.

[0888] Step 2:

[0889] The server uses reliable energy usage data to predict energy usage spikes using a recurrent neural network (RNN) and long-short-term memory (LSTM). The input is pre-processed time-series data, and the server performs predictive calculations based on the model, outputting predictions of future consumption spikes. These predictions are then used to generate the next scenario.

[0890] Step 3:

[0891] The server uses a generative AI model to generate multiple usage scenarios and evaluate their effectiveness. The input is predicted energy spike data, and the generative AI model evaluates the effectiveness of each scenario and selects the optimal one. This process outputs new scenarios that will be automatically executed as self-executing contracts.

[0892] Step 4:

[0893] The terminal receives usage scenarios generated as self-execution contracts and notifies the user. The input is the scenario sent from the server, and the notification content is optimized based on the user's emotional state. Specifically, the voice tone and message content are adjusted to be more acceptable to the user. The output is a notification of the energy usage change plan presented to the user.

[0894] Step 5:

[0895] The user reviews the usage plan presented on the device and indicates whether they approve or reject it. The input is optimized notification content, and user feedback is output. Approved plans are fed back to the server.

[0896] Step 6:

[0897] The server adjusts and executes the device's operating time based on the usage plan approved by the user. The input is the approved plan information, specifically adjusting the device's operation to reduce energy use during certain time periods and shift it to other time periods. The output is the optimized energy usage schedule.

[0898] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0899] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0900] 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 this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0901] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0902] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0903] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0904] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0905] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0906] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0907] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0908] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0909] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0910] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0911] 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.

[0912] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0913] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0914] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0915] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0916] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0917] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0918] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0919] The following is further disclosed regarding the embodiments described above.

[0920] (Claim 1)

[0921] A means for collecting and preprocessing energy consumption data,

[0922] A method for predicting energy consumption spikes using recurrent neural networks and long- and short-term memory,

[0923] A means of generating multiple consumption scenarios using generative AI, evaluating their effectiveness, and selecting the optimal scenario,

[0924] A means of automatically generating smart contracts based on selected consumption scenarios,

[0925] A means of notifying users of consumption shift plans via a terminal and obtaining their approval,

[0926] A system that includes means for adjusting and executing the operating schedule of equipment based on user approval.

[0927] (Claim 2)

[0928] The system according to claim 1, which performs an algorithm to impute missing data and remove outliers in the preprocessing of energy consumption data.

[0929] (Claim 3)

[0930] The system according to claim 1, wherein the generated smart contract includes the specific details, timing, and conditions of a consumption shift plan.

[0931] "Example 1"

[0932] (Claim 1)

[0933] A means for collecting information related to energy consumption and preprocessing it to improve the quality of the data,

[0934] A method for predicting future energy consumption patterns using time series forecasting models,

[0935] A method for generating multiple consumption adjustment scenarios using a generative model, evaluating the effects of each, and selecting the optimal scenario,

[0936] A means for automatically generating contracts based on selected consumption adjustment scenarios,

[0937] A means of notifying end users of a consumption adjustment plan via an information transmission device and obtaining their approval,

[0938] A system that includes means for optimizing the operation schedule of equipment based on end-user approval.

[0939] (Claim 2)

[0940] The system according to claim 1, which performs processing to fill in missing information and remove abnormal data in the preprocessing of information related to energy consumption.

[0941] (Claim 3)

[0942] The system according to claim 1, wherein the generated contract includes a specific consumption adjustment plan, timing of implementation, and conditions.

[0943] "Application Example 1"

[0944] (Claim 1)

[0945] A means for collecting and pre-processing energy usage information,

[0946] A method for predicting increased energy consumption using recurrent neural networks and long-term short-term memory,

[0947] A means for generating multiple consumption plans using generative AI, evaluating their effectiveness, and selecting the optimal plan,

[0948] A means for generating an automated contract based on the selected consumption plan,

[0949] A means of notifying users of proposed consumption shifts via communication terminals and obtaining their approval,

[0950] A means of adjusting and executing the equipment operation plan based on user approval,

[0951] A means to propose and automatically implement approved plans for optimizing energy use across the entire city,

[0952] A system that includes this.

[0953] (Claim 2)

[0954] The system according to claim 1, which performs an analysis procedure to fill in missing information and remove outliers in the preprocessing of energy usage information.

[0955] (Claim 3)

[0956] The system according to claim 1, wherein the generated automated contract includes specific elements of a consumption shift plan, timing of implementation, and conditions, and aims to optimize energy use across the city.

[0957] "Example 2 of combining an emotion engine"

[0958] (Claim 1)

[0959] A means for collecting and processing energy usage information,

[0960] Methods for predicting energy spikes using recurrent neural networks and long-term short-term memory networks,

[0961] A means for generating multiple consumption scenarios using a generative model, evaluating their effectiveness, and selecting the optimal scenario,

[0962] A means of automatically generating smart contracts based on selected consumption scenarios,

[0963] A means of notifying users of consumption change plans via a terminal and obtaining their approval,

[0964] A means of analyzing the user's emotional state and selecting the optimal notification method,

[0965] A system that includes means for adjusting and executing the operation schedule of a device based on user approval.

[0966] (Claim 2)

[0967] The system according to claim 1, which executes an algorithm to impute missing information and remove outliers in data processing.

[0968] (Claim 3)

[0969] The system according to claim 1, wherein the generated smart contract includes the specific details, timing, and conditions of a consumption change plan.

[0970] "Application example 2 when combining with an emotional engine"

[0971] (Claim 1)

[0972] A means for collecting and initially processing energy usage information,

[0973] A method for predicting energy usage spikes using recurrent neural networks and long- and short-term memory,

[0974] A means of generating multiple usage scenarios using generative AI, evaluating their effectiveness, and selecting the optimal scenario,

[0975] A means for automatically generating a self-execution contract based on the selected usage scenario,

[0976] A means of notifying users of the usage change plan via a terminal and obtaining their approval,

[0977] A means of adjusting notifications based on the user's emotional state,

[0978] A system that includes means for adjusting and executing the operating time of a device based on user approval.

[0979] (Claim 2)

[0980] The system according to claim 1, which executes an algorithm to fill in missing information and remove outliers in the initial processing of energy usage information.

[0981] (Claim 3)

[0982] The system according to claim 1, wherein the generated self-execution contract includes the specific details, timing, and conditions of the usage change plan, and further includes options based on the user's emotional state. [Explanation of Symbols]

[0983] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for collecting and preprocessing energy consumption data, A method for predicting energy consumption spikes using recurrent neural networks and long- and short-term memory, A means of generating multiple consumption scenarios using generative AI, evaluating their effectiveness, and selecting the optimal scenario, A means of automatically generating smart contracts based on selected consumption scenarios, A means of notifying users of consumption shift plans via a terminal and obtaining their approval, A system that includes means for adjusting and executing the operating schedule of equipment based on user approval.

2. The system according to claim 1, which executes an algorithm to impute missing data and remove outliers in the preprocessing of energy consumption data.

3. The system according to claim 1, wherein the generated smart contract includes the specific details, execution timing, and conditions of a consumption shift plan.

Images