System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting energy usage data; means for transmitting the energy usage data; means for storing the transmitted energy usage data; means for analyzing the stored energy usage data; means for generating an energy saving advice based on a result of the analysis; means for displaying the generated energy saving advice; and means for collecting feedback on the energy saving advice.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Problems such as the depletion of energy resources, budget shortfalls for water infrastructure, and rising utility costs due to rising prices are becoming more serious for many homes and businesses. To solve these problems, efficient energy use is essential, and support is needed to accurately grasp energy usage status and implement appropriate energy-saving measures. Conventional energy management systems have limitations in data collection and analysis, making it difficult to provide individualized energy-saving advice. The present invention aims to provide a system that solves these problems and promotes more efficient energy use. [Means for solving the problem]

[0005] The present invention solves the above-mentioned problems with a system that includes the following means: means for collecting energy usage data, means for transmitting the energy usage data, means for storing the transmitted energy usage data, means for analyzing the stored energy usage data, means for generating energy-saving advice based on the analysis results, means for displaying the generated energy-saving advice, and means for collecting feedback on the energy-saving advice. In particular, the system collects electricity, gas, and water usage data, generates energy-saving advice using AI, and notifies the user, thereby promoting efficient energy use. The system is also designed to improve the quality of the advice based on feedback.

[0006] "Energy usage data" refers to data related to energy consumption such as electricity, gas, and water.

[0007] A "collection instrument" is a device or apparatus used to obtain energy usage data for a home or business.

[0008] The "transmitting means" is a communication device for sending the collected energy usage data to a relay point such as a server.

[0009] "Storing means" is a database or storage system for recording and storing transmitted energy usage data.

[0010] "Means of analysis" refers to software or algorithms that detect consumption patterns and abnormal trends based on stored energy usage data.

[0011] "Energy saving advice" is a specific proposal to promote efficient energy use based on the analysis of energy usage data.

[0012] The "means of generation" refers to the generation AI and related algorithms used to create energy-saving advice based on the analysis results.

[0013] The "display means" refers to a display device or application for visually presenting the generated energy saving advice to the user.

[0014] "Feedback" is information about the actions and results of the energy saving advice that the user has taken.

[0015] A "collection means" is a system or device for receiving feedback provided by a user. [Brief explanation of the drawings]

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

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

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

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

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

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

[0022] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0024] [First embodiment]

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

[0026] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

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

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

[0029] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

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

[0031] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0033] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

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

[0035] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

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

[0037] System Overview

[0038] The present invention relates to a system that collects and analyzes energy usage data and provides energy-saving advice based on the results. This enables homes and businesses to use energy more efficiently and promotes reductions in utility costs. The system mainly includes a collection means, transmission means, storage means, analysis means, generation means, display means, and feedback collection means for handling energy usage data.

[0039] Program processing explanation

[0040] 1. Data Collection

[0041] The terminals are connected to the HEMS (Home Energy Management System) to collect electricity, gas, and water usage data from each home or business.

[0042] The terminal periodically (e.g., every three hours) obtains usage data from the HEMS and sends it to the server.

[0043] 2. Data Receipt and Storage

[0044] The server receives the energy usage data transmitted from the terminal.

[0045] The server stores the received data in a database, organizing it by household and business and recording it in chronological order.

[0046] 3. Data Preprocessing

[0047] The server extracts energy usage data for a certain period of time (e.g., the last 30 days) from the database.

[0048] The server performs pre-processing to complement missing values ​​in the extracted data and remove noise. Specifically, it detects and corrects abnormally high-consumption data.

[0049] 4. Data Analysis and Pattern Recognition

[0050] The server inputs the preprocessed data into the generative AI model.

[0051] Generative AI uses past data to recognize patterns of energy consumption and unusual trends.

[0052] 5. Advice Generation

[0053] The server generates specific energy-saving advice based on the output from the generative AI model.

[0054] Generative AI provides advice tailored to individual consumption patterns.

[0055] 6. Advice Delivery

[0056] The server transmits the generated energy saving advice to the terminal.

[0057] The terminal notifies the user of the received advice and displays it within the application.

[0058] 7. Gathering Feedback

[0059] The user receives feedback within the application about the effectiveness of the advice they have implemented, for example, by inputting a specific action such as "I changed the refrigerator temperature setting."

[0060] The terminal transmits the feedback input by the user to the server.

[0061] 8. Feedback Analysis

[0062] The server analyzes the received feedback data and evaluates the effectiveness of the advice.

[0063] The generative AI learns from the feedback data and reflects it in future advice generation.

[0064] Specific examples

[0065] Examples of reducing electricity usage

[0066] The terminal obtains 24 hours of electricity usage data from the HEMS every three hours and sends it to the server.

[0067] The server receives the sent data and stores it in the database for "Home ID_1234."

[0068] The server extracts the electricity usage data for the last 30 days from the database for "household ID_1234" and detects and corrects any invalid data (e.g., extremely high consumption).

[0069] The generation AI analyzes the data for "household ID_1234" and determines that nighttime consumption is higher than average.

[0070] The AI ​​generates advice recommending that "if you are not using home appliances at night, unplug them" and outputs it to the server.

[0071] The server sends the generated advice to the terminal, which notifies the user within the application and displays the details.

[0072] The user inputs into the application that "I unplugged my home appliances overnight," and the device sends feedback to the server.

[0073] The server analyzes the feedback data and confirms that electricity usage has been reduced.

[0074] The generation AI will incorporate this feedback as learning data and reflect it in generating the next piece of advice.

[0075] Example of reducing gas usage

[0076] The terminal periodically obtains gas usage data from the HEMS and sends it to the server.

[0077] The server receives the data and stores it in a database.

[0078] The server preprocesses the data and adjusts for outliers.

[0079] Generative AI analyzes the data and discovers that certain cooking methods increase gas consumption.

[0080] The generative AI generates advice such as "Use a pressure cooker to reduce cooking time" and outputs it to the server.

[0081] The server sends the generated advice to the terminal, which notifies the user within the application and displays the recommendation.

[0082] The user enters "I used a pressure cooker" in the application, and the device sends the feedback to the server.

[0083] The server analyzes the feedback data and confirms that gas usage has been reduced.

[0084] The generation AI will incorporate this feedback as learning data and reflect it in generating the next piece of advice.

[0085] Through the above processing, the present invention can improve the efficiency of energy use and contribute to reducing utility costs for homes and businesses.

[0086] The processing flow will be explained below.

[0087] Program processing steps

[0088] 1. Data Collection

[0089] Step 1:

[0090] The device connects to the HEMS and collects electricity, gas, and water usage data, including energy usage information for homes and businesses.

[0091] Step 2:

[0092] The device sends the collected data to the server at specified intervals (e.g., every 3 hours).

[0093] 2. Data Receipt and Storage

[0094] Step 3:

[0095] The server receives the energy usage data transmitted from the terminal.

[0096] Step 4:

[0097] The server stores the received data in a database and organizes it by household or business. The data is recorded in chronological order.

[0098] 3. Data Preprocessing

[0099] Step 5:

[0100] The server extracts energy usage data for a certain period (e.g., the last 30 days) from the database.

[0101] Step 6:

[0102] The server completes the extracted data, removes noise, and detects abnormal high-consumption data and corrects it as necessary.

[0103] 4. Data Analysis and Pattern Recognition

[0104] Step 7:

[0105] The server inputs the preprocessed data into the generative AI model.

[0106] Step 8:

[0107] Generative AI recognizes patterns and unusual trends in energy consumption, identifying increases or decreases in consumption for specific times of day or dates.

[0108] 5. Advice Generation

[0109] Step 9:

[0110] The server receives the output of the generative AI model and generates specific energy-saving advice.

[0111] Step 10:

[0112] Generative AI provides customized advice based on individual consumption patterns and usage.

[0113] 6. Advice Delivery

[0114] Step 11:

[0115] The server transmits the generated energy saving advice to the terminal.

[0116] Step 12:

[0117] The terminal notifies the user of the received advice and displays it within the application.

[0118] 7. Gathering Feedback

[0119] Step 13:

[0120] Users can provide feedback within the application about the effectiveness of the advice they have implemented, by entering specific actions and results.

[0121] Step 14:

[0122] The terminal transmits the feedback input by the user to the server.

[0123] 8. Feedback Analysis

[0124] Step 15:

[0125] The server analyzes the received feedback data and evaluates the effectiveness of the advice.

[0126] Step 16:

[0127] The generative AI learns from the feedback data and reflects it in the next advice generation, aiming to provide improved advice.

[0128] Specific examples

[0129] Examples of reducing electricity usage

[0130] Step 1:

[0131] The terminal obtains 24 hours of electricity usage data from the HEMS every three hours.

[0132] Step 2:

[0133] The terminal transmits the collected data to the server.

[0134] Step 3:

[0135] The server receives the transmitted data.

[0136] Step 4:

[0137] The server stores the received data in a database.

[0138] Step 5:

[0139] The server extracts the electricity usage data for the last 30 days from the database for "Home ID_1234."

[0140] Step 6:

[0141] The server detects and corrects abnormally high data consumption.

[0142] Step 7:

[0143] The generative AI analyzes the data and identifies higher-than-average consumption at night.

[0144] Step 8:

[0145] The AI ​​generates advice such as "unplug appliances if you are not using them at night."

[0146] Step 9:

[0147] The server transmits the generated advice to the terminal.

[0148] Step 10:

[0149] The device will notify the user within the application and display details.

[0150] Step 11:

[0151] The user inputs into the application that "I unplugged my home appliances overnight."

[0152] Step 12:

[0153] The terminal sends the feedback to the server.

[0154] Step 13:

[0155] The server analyzes the feedback data and verifies that electricity usage has been reduced.

[0156] Step 14:

[0157] The generative AI incorporates the feedback as learning data and reflects it in generating the next piece of advice.

[0158] Example 1

[0159] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0160] Conventional energy management systems are limited to collecting and simply analyzing energy usage data, making it difficult to efficiently provide specific energy-saving advice to users. They also lack a mechanism for collecting feedback and reflecting the results in the analysis. This creates the challenge of being unable to provide timely and appropriate energy-saving advice tailored to the user's energy consumption patterns.

[0161] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0162] In this invention, the server includes means for periodically transmitting energy usage data, means for receiving the transmitted energy usage data, means for storing the received energy usage data, means for extracting and preprocessing the stored energy usage data, means for inputting the preprocessed data into a generative AI model to recognize energy consumption patterns, means for generating energy saving advice based on the recognition results, means for transmitting the generated energy saving advice, means for displaying the transmitted energy saving advice, means for collecting feedback on the energy saving advice, and means for analyzing the collected feedback and updating the generative AI model. This integrates the collection and analysis of energy usage data, advice generation, feedback collection, and model updating, making it possible to provide users with timely and effective energy saving advice.

[0163] "Energy usage data" refers to data showing the amount of electricity, gas, water, etc. used, and is a measurement of the specific energy usage status consumed within homes and businesses.

[0164] "HEMS" is an abbreviation for Home Energy Management System, a system for optimizing and managing energy usage within the home.

[0165] A "generative AI model" refers to an artificial intelligence model that performs statistical analysis and machine learning based on various data, and uses the knowledge gained from this to generate new information.

[0166] A "prompt sentence" is an input sentence for a specific generative AI model, and is a sentence that instructs the model to produce a specific output.

[0167] "Data preprocessing" refers to the process carried out before data analysis, and refers to the process of preparing data by filling in missing values ​​and removing noise, etc.

[0168] "Feedback" is information that records the user's response to and results of implementing advice provided by the system, and reflects this information back into the system.

[0169] The present invention relates to a system that collects and analyzes energy usage data and provides energy conservation advice based on the results. This invention makes it possible to improve the efficiency of energy usage in homes and businesses and promote reductions in utility costs. Specifically, this system operates according to the following steps.

[0170] System Configuration

[0171] The system mainly includes the following components:

[0172] 1. Collection Method

[0173] 2. Transmission Method

[0174] 3. Preservation means

[0175] 4. Pretreatment Methods

[0176] 5. Analysis method

[0177] 6. Advice Generation Methods

[0178] 7. Display means

[0179] 8. Feedback Collection Methods

[0180] Details of each component are shown below.

[0181] Collection Method

[0182] The device collects energy usage data. Specifically, the device connects to a HEMS (Home Energy Management System) and obtains electricity, gas, and water usage data every three hours. For example, the device obtains the data using the HEMS API ("GET / energy_usage"). The device temporarily stores the data in local storage and prepares to send it to the server later.

[0183] Transmission method

[0184] The device sends the temporarily stored energy usage data to the server using MQTT or REST API. For example, the data to be sent is in JSON format and is assigned an identifier such as "Home ID_1234."

[0185] Preservation means

[0186] The server receives the energy usage data sent from the device and stores it in a NoSQL database (for example, MongoDB). The server stores the received data separately for each household and each business, and manages it in chronological order, which allows for efficient data access.

[0187] Pretreatment means

[0188] The server extracts the most recent 30 days' worth of data from the database, imputes missing values, and performs preprocessing to remove noise. Specifically, the data is cleansed using Python's pandas library, and outliers are corrected using algorithms such as K-Nearest Neighbor (KNN).

[0189] Analysis means

[0190] The server inputs the preprocessed data into a generative AI model to recognize energy consumption patterns and abnormal trends. An example of the generative AI model is OpenAI's GPT. An example prompt is "Please identify abnormal patterns and signs of energy saving from the energy usage data for the last 30 days for household ID_1234."

[0191] Advice Generation Method

[0192] Based on the analysis results, the generation AI generates energy-saving advice tailored to individual consumption patterns. For example, it may generate advice recommending that appliances be unplugged if not in use at night. The generated advice is sent back to the server, which then formats the information in JSON format and stores it back in the database.

[0193] Display means

[0194] The server sends the generated energy-saving advice to the device. The device saves the received advice in local storage and notifies the user within the application. For example, a smartphone push notification can be used to notify the user that "new energy-saving advice is available."

[0195] Feedback collection methods

[0196] The user acts on the advice and provides feedback on the results within the application. Specifically, the user inputs something like, "I changed the refrigerator's set temperature." The device receives this feedback and sends it to the server. The sent feedback is stored in a feedback database on the server.

[0197] Feedback Analysis

[0198] The server analyzes the received feedback data and evaluates the effectiveness of the advice. The generative AI model learns from this feedback data and reflects it in generating new advice. For example, it learns how much power consumption was reduced by changing the refrigerator's set temperature, and reflects this in the next advice.

[0199] Specific examples

[0200] A specific example of operation is shown below.

[0201] 1. The device obtains electricity usage data from the HEMS every three hours and sends the data to the server.

[0202] 2. The server receives the data and stores it in the database as "Home ID_1234".

[0203] 3. The server extracts and preprocesses the data for the last 30 days.

[0204] 4. The generation AI performs an analysis based on the prompt: "Please identify any abnormal patterns or signs of energy saving from the last 30 days of energy usage data for household ID_1234."

[0205] 5. The AI ​​generates advice such as "It is recommended to unplug the device at night" and sends it to the server.

[0206] 6. The server sends the advice to the device, and the device notifies the user via push notification that "New energy saving advice is available."

[0207] 7. The user enters into the application that "I unplugged my appliances overnight," and the device sends feedback to the server.

[0208] 8. The server analyzes the feedback, and the AI ​​uses it to generate the next piece of advice.

[0209] As described above, the present invention can improve the efficiency of energy use and contribute to reducing utility costs for homes and businesses.

[0210] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0211] Step 1: Data collection

[0212] The device connects to the HEMS and obtains electricity, gas, and water usage data every three hours. Specifically, it accesses the HEMS API endpoint "GET / energy_usage" to receive the latest energy usage data. The input is sensor data from the HEMS, and the output is usage data saved in the device's temporary storage. For example, data in the format "Electricity usage: 10kWh" is saved.

[0213] Step 2: Send data

[0214] The energy usage data collected by the device is sent to the server. The data is sent periodically (for example, every three hours) using MQTT or REST API. The input is temporarily stored usage data, and the output is the data to be sent to the server. Specifically, data such as "electricity usage: 10kWh, gas usage: 5m3, water usage: 100L" is structured in JSON format and sent to the server.

[0215] Step 3: Data reception and storage

[0216] The server receives the data sent from the device and stores it in a NoSQL database (for example, MongoDB). The input is the JSON data sent from the device, and the output is the energy usage data stored in the database. Specifically, the received data is organized by household and company, and stored using an identifier such as "Home ID_1234."

[0217] Step 4: Preprocessing the data

[0218] The server extracts the last 30 days' worth of energy usage data from the database. It then performs pre-processing to fill in missing values ​​in the extracted data and remove noise. The input is energy usage data from the database (e.g., the last 30 days' worth of data for "household ID_1234"), and the output is pre-processed, clean data. Specifically, it uses Python's pandas library to fill in missing values ​​and the KNN algorithm to detect and correct outliers.

[0219] Step 5: Data analysis and pattern recognition

[0220] The server inputs the preprocessed data into the generative AI model. The input is the preprocessed data (e.g., energy usage data for the last 30 days for "household ID_1234"), and the output is consumption patterns and abnormal trends recognized by the generative AI model. Specifically, the generative AI is prompted with the following prompt: "Please identify abnormal patterns and signs of energy saving from the energy usage data for the last 30 days for household ID_1234."

[0221] Step 6: Advice Generation

[0222] The generative AI generates energy-saving advice based on the analysis results. The input is the output of the generative AI model (e.g., consumption patterns and abnormal trends), and the output is specific energy-saving advice. For example, the generated advice recommends "unplugging appliances when not in use at night." This advice is sent to the server, which then formats the advice content in JSON format and stores it in a database.

[0223] Step 7: Advice Delivery

[0224] The server sends the generated energy-saving advice to the device. The input is the generated advice (e.g., "Unplug appliances if they are not in use at night"), and the output is the data sent to the device. The device saves the received advice in local storage and notifies the user within the application. Specifically, it uses a push notification on the smartphone to notify the user that "new energy-saving advice is available."

[0225] Step 8: Gather feedback

[0226] The results of the user's execution of advice within the application are fed back. The input is information about the user's actions (e.g., "I changed the refrigerator's set temperature"), and the output is the feedback data. The device receives feedback from the user and sends it to the server. Specifically, it formats the user's input in JSON format and sends a POST request to the server.

[0227] Step 9: Analyze feedback

[0228] The server analyzes the received feedback data and evaluates the effectiveness of the advice. The input is the feedback data (e.g., "As a result of changing the set temperature, power consumption decreased by 5%), and the output is the evaluation result. The generation AI learns from this feedback data and reflects it in generating new advice. For example, it learns that "power consumption at night has decreased" and uses this information to generate the next piece of advice.

[0229] The above are the specific processing steps of this system.

[0230] (Application example 1)

[0231] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0232] In modern brick-and-mortar stores, efficient energy use is important from the perspective of environmental protection and cost reduction. However, in reality, there is a lack of means to grasp energy usage patterns in detail and receive efficient energy-saving advice. As a result, store managers are unable to implement appropriate energy-saving measures, and wasteful energy consumption continues. The objective of this invention is to promote efficient energy use and reduce wasteful energy consumption by analyzing energy usage data in brick-and-mortar stores in detail and providing specific energy-saving advice.

[0233] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0234] In this invention, the server includes means for collecting energy usage data, means for transmitting the energy usage data, means for storing the transmitted energy usage data, means for preprocessing the stored energy usage data, means for analyzing the stored energy usage data, means for generating energy saving advice based on the analysis results, means for generating the generated energy saving advice for specific actions in the store, means for displaying the generated energy saving advice, and means for collecting feedback on the energy saving advice. This makes it possible to understand energy usage patterns in the physical store in detail and to specifically implement efficient energy saving measures.

[0235] "Energy usage data" refers to data that records the amount of energy consumed in a physical store, such as electricity, gas, and water.

[0236] The "transmission means" is a function for transmitting collected energy usage data to a server.

[0237] The "storage means" is a function for storing the transmitted energy usage data in a storage device such as a database.

[0238] The "preprocessing means" is a function for performing preprocessing such as complementing missing values ​​and removing outliers on stored energy usage data.

[0239] The "analysis means" is a function for analyzing the pre-processed energy usage data and identifying patterns and anomalies in energy consumption.

[0240] The "generation means" is a function for generating energy saving advice based on the analysis results.

[0241] The "display means" is a function for presenting the generated energy saving advice to the operator of the physical store.

[0242] The "feedback collection means" is a function for collecting feedback regarding the energy saving advice that the operator has implemented.

[0243] "Generative AI" is an artificial intelligence model that analyzes energy usage data and generates specific energy-saving advice.

[0244] A "prompt sentence" is an instruction sentence that provides specific energy-saving advice to the generation AI.

[0245] "In-store energy saving advice" is advice that recommends specific energy saving actions for specific equipment or activities in a physical store.

[0246] The present invention relates to a system that collects and analyzes energy usage data and provides energy conservation advice based on the results, thereby making it possible to improve the efficiency of energy usage in physical stores and promote reductions in utility costs.

[0247] The system includes a collection means, a transmission means, a storage means, a pre-processing means, an analysis means, a generation means, a display means, and a feedback collection means for handling energy usage data.

[0248] Specific hardware and software configuration:

[0249] The terminal, a smartphone or tablet, connects to the HEMS (Home Energy Management System) in the physical store and collects energy usage data.

[0250] The server receives the collected energy usage data and stores it in a database, where it is organized by household and business and recorded in chronological order.

[0251] The preprocessing means extracts energy usage data for a certain period from the database and performs preprocessing such as filling in missing values ​​and removing outliers, thereby ensuring the reliability of the data.

[0252] The analytics tool feeds the pre-processed data into a generative AI model to identify patterns and unusual trends in energy consumption.

[0253] The AI ​​that generates the energy-saving advice is based on the analysis results and is specific to the user's consumption patterns.

[0254] The terminal as a display means notifies the user of the generated energy saving advice and displays it within the application.

[0255] The feedback collection means collects feedback regarding the effect of the energy saving advice implemented by the user and transmits the feedback to the server.

[0256] Software used and detailed data processing:

[0257] Data collection: The device acquires energy usage data from the HEMS. The data is sent to the server periodically (e.g., every three hours).

[0258] Data storage and preprocessing: The server stores the data in a database and preprocesses it, including imputing missing values ​​and removing outliers.

[0259] Data analysis and pattern recognition: Analyze the pre-processed data and recognize consumption patterns and unusual trends from historical data. Generative AI models take on this role and automate the process.

[0260] Generating energy-saving advice: The AI ​​generates energy-saving advice based on the analysis results. For example, it may generate advice such as, "Since power consumption is high at night, we recommend turning off devices outside of business hours."

[0261] User notification and feedback collection: The generated advice is notified to the user and feedback is collected within the application. The feedback is sent back to the server and reflected in subsequent advice generation.

[0262] Examples and prompts:

[0263] Example: Identifying high consumption at night and making specific recommendations on which appliances in a store should be turned off outside of business hours.

[0264] Example prompt: "Our store consumes a lot of electricity at night. Please have your Generative AI suggest which devices in our store should be turned off after hours."

[0265] In this way, the present invention efficiently manages energy usage in physical stores and provides specific energy-saving advice, enabling operators of physical stores to implement appropriate energy-saving measures.

[0266] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0267] Step 1:

[0268] Data collection

[0269] The terminal obtains electricity, gas, and water usage data every three hours from the physical store's HEMS (Home Energy Management System). The input is real-time energy usage data from the HEMS, and the output is collected energy usage data, which is used for further analysis.

[0270] Step 2:

[0271] Data transmission

[0272] The terminal periodically transmits the collected energy usage data to the server. The input is the collected energy usage data, and the output is the energy usage data transmitted to the server. The transmitted data is stored on the server.

[0273] Step 3:

[0274] Data storage

[0275] The server receives the transmitted energy usage data and stores it in a database. The input is the energy usage data transmitted to the server, and the output is the energy usage data stored in the database. This data is used for later analysis.

[0276] Step 4:

[0277] Data Preprocessing

[0278] The server extracts energy usage data for a certain period of time (e.g., the last 30 days) from the database, and completes missing values ​​and removes outliers. The input is the energy usage data stored in the database, and the output is the corrected and preprocessed energy usage data. Specifically, it completes missing values ​​and removes outliers using the 3 sigma rule.

[0279] Step 5:

[0280] Data Analysis and Pattern Recognition

[0281] The server inputs the preprocessed energy usage data into a generative AI model to identify energy consumption patterns and anomalies. The input is the preprocessed energy usage data, and the output is the recognition results of consumption patterns and anomalies. Specifically, it finds consumption patterns from past data and identifies abnormal trends.

[0282] Step 6:

[0283] Energy saving advice generation

[0284] The server uses a generative AI model based on the analysis results to generate energy-saving advice. The input is the consumption pattern and anomaly recognition results, and the output is the generated energy-saving advice. For example, it generates specific advice such as "Since consumption is high at night, it is recommended that you turn off devices outside of business hours."

[0285] Step 7:

[0286] Advice display

[0287] The device notifies the user of the generated energy-saving advice and displays it within the application. The input is the generated energy-saving advice, and the output is the advice displayed to the user. Specifically, a notification is sent to the user's smartphone or tablet.

[0288] Step 8:

[0289] Feedback collection

[0290] The user inputs feedback about the effectiveness of the energy-saving advice they have implemented through the application. The input is the user's feedback, and the output is feedback data. Specifically, the user inputs an action such as "I unplugged my home appliances at night" into the application.

[0291] Step 9:

[0292] Feedback Analysis

[0293] The server analyzes the received feedback data, evaluates the effectiveness of the advice, and provides feedback to the generation AI model. The input is the feedback data sent by the user, and the output is the effectiveness evaluation result. Specifically, the effectiveness of the energy-saving advice is evaluated and reflected in the generation of the next advice.

[0294] By following these steps, energy usage in physical stores can be managed efficiently and specific energy-saving advice can be provided.

[0295] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0296] System Overview

[0297] The present invention combines a system that collects and analyzes energy usage data and provides energy-saving advice based on the results with an emotion engine that recognizes user emotions. This makes it possible to improve the efficiency of energy use in homes and businesses, promote reductions in utility bills, and provide advice that takes user emotions into consideration. The system mainly includes a collection means for handling energy usage data, a transmission means, a storage means, an analysis means, a generation means, a display means, an emotion engine, and a feedback collection means.

[0298] Program processing explanation

[0299] 1. Data Collection

[0300] The terminals are connected to the HEMS (Home Energy Management System) to collect electricity, gas, and water usage data from each home or business.

[0301] The terminal periodically (e.g., every three hours) obtains usage data from the HEMS and sends it to the server.

[0302] 2. Data Receipt and Storage

[0303] The server receives the energy usage data transmitted from the terminal.

[0304] The server stores the received data in a database, organizing it by household and business and recording it in chronological order.

[0305] 3. Data Preprocessing

[0306] The server extracts energy usage data for a certain period (e.g., the last 30 days) from the database.

[0307] The server performs pre-processing to complement missing values ​​in the extracted data and remove noise. Specifically, it detects and corrects abnormally high-consumption data.

[0308] 4. Data Analysis and Pattern Recognition

[0309] The server inputs the preprocessed data into the generative AI model.

[0310] Generative AI uses past data to recognize patterns of energy consumption and unusual trends.

[0311] 5. Advice Generation

[0312] The server generates specific energy-saving advice based on the output from the generative AI model.

[0313] Generative AI provides advice tailored to individual consumption patterns.

[0314] 6. Emotion recognition and advice adjustment

[0315] The device analyzes the user's feedback, voice, and facial expressions to recognize the user's emotions.

[0316] The emotion engine tailors the generated energy-saving advice based on the recognized emotion: for example, if the user is feeling stressed, it generates a message containing encouraging and kind words.

[0317] 7. Advice Delivery

[0318] The server transmits the adjusted energy saving advice to the terminal.

[0319] The terminal notifies the user of the received advice and displays it within the application.

[0320] 8. Collecting Feedback

[0321] Users can provide feedback within the application about the effectiveness of the advice they have implemented, by entering specific actions and results.

[0322] The terminal transmits the feedback input by the user to the server.

[0323] 9. Feedback Analysis

[0324] The server analyzes the received feedback data and evaluates the effectiveness of the advice.

[0325] The generation AI learns from the feedback data and reflects it in generating the next piece of advice.

[0326] Specific examples

[0327] Examples of reducing electricity usage

[0328] System Operation

[0329] The device retrieves 24 hours of electricity usage data from the HEMS every three hours.

[0330] The terminal transmits the collected data to the server.

[0331] The server receives the data and stores it in a database.

[0332] The server extracts the electricity usage data for the last 30 days from the database for "Home ID_1234."

[0333] The server detects and corrects abnormally high data consumption.

[0334] Generative AI analyzes the data and identifies higher-than-average consumption at night.

[0335] The AI ​​generates advice such as "unplug appliances if you are not using them at night."

[0336] The device analyzes the user's feedback, voice, and facial expressions, and uses an emotion engine to check whether the user is feeling stressed.

[0337] The emotion engine tailors advice as needed, adding encouragement and kind words.

[0338] The server sends the tailored advice to the terminal, which notifies the user within the application and displays the details.

[0339] The user inputs into the application that "I unplugged my home appliances overnight."

[0340] The device sends the feedback to the server.

[0341] The server analyzes the feedback data and confirms that electricity usage has been reduced.

[0342] The generation AI will incorporate this feedback as learning data and reflect it in generating the next piece of advice.

[0343] The system not only improves energy efficiency and reduces utility bills, but also enhances the user experience by providing emotionally sensitive advice, making energy management in homes and businesses more effective and satisfying.

[0344] The processing flow will be explained below.

[0345] Program processing steps

[0346] 1. Data Collection

[0347] Step 1:

[0348] The device connects to the HEMS and collects electricity, gas, and water usage data, which is obtained from sensors and meters installed in each home or business.

[0349] Step 2:

[0350] The device sends the collected usage data to the server at specified intervals (e.g., every 3 hours).

[0351] 2. Data Receipt and Storage

[0352] Step 3:

[0353] The server receives the energy usage data transmitted from the terminal.

[0354] Step 4:

[0355] The server stores the received data in a database, organizing it by household and business and recording it in chronological order.

[0356] 3. Data Preprocessing

[0357] Step 5:

[0358] The server detects missing values ​​in the energy usage data and completes them as necessary.

[0359] Step 6:

[0360] The server scrutinizes the data to remove abnormal values ​​and noise, and detects and corrects abnormally high consumption data.

[0361] 4. Data Analysis and Pattern Recognition

[0362] Step 7:

[0363] The server inputs the preprocessed data into the generative AI model.

[0364] Step 8:

[0365] Generative AI recognizes patterns and unusual trends in energy consumption, for example, identifying sudden increases or decreases in consumption on certain days or during certain times of the day.

[0366] 5. Advice Generation

[0367] Step 9:

[0368] The server generates specific advice for energy conservation based on the analysis results of the generation AI.

[0369] Step 10:

[0370] The generative AI provides customized advice based on consumption patterns, such as "unplug appliances if you're not using them overnight" or "use a pressure cooker to reduce cooking time."

[0371] 6. Emotion recognition and advice adjustment

[0372] Step 11:

[0373] The device analyzes the user's feedback, voice, and facial expressions to recognize the user's emotions. Technologies used include voice recognition and facial recognition.

[0374] Step 12:

[0375] The emotion engine adjusts the energy-saving advice based on the recognized emotion. For example, if the user is feeling stressed, the engine generates advice that includes encouraging and kind words.

[0376] 7. Advice Delivery

[0377] Step 13:

[0378] The server transmits the adjusted energy saving advice to the terminal.

[0379] Step 14:

[0380] The device will notify the user and provide detailed advice within the application, which the user can review at any time.

[0381] 8. Collecting Feedback

[0382] Step 15:

[0383] The user provides feedback on the effectiveness of the advice they received within the application and whether they implemented it. For example, they can input information such as "I changed the refrigerator temperature setting" or "I used a pressure cooker."

[0384] Step 16:

[0385] The terminal transmits the feedback input by the user to the server.

[0386] 9. Feedback Analysis

[0387] Step 17:

[0388] The server analyzes the received feedback data and evaluates the effectiveness of the advice, for example, checking whether electricity usage has been reduced.

[0389] Step 18:

[0390] The AI ​​learns from the feedback data and reflects it in the next advice generation, allowing it to provide more effective and tailored advice to the user.

[0391] Specific examples

[0392] Reducing electricity consumption

[0393] Step 1:

[0394] The device retrieves 24 hours of electricity usage data from the HEMS every three hours.

[0395] Step 2:

[0396] The terminal transmits the collected data to the server.

[0397] Step 3:

[0398] The server receives the transmitted data.

[0399] Step 4:

[0400] The server stores the received data in a database.

[0401] Step 5:

[0402] The server extracts the electricity usage data for the last 30 days from the database for "Home ID_1234."

[0403] Step 6:

[0404] The server detects and corrects abnormally high data consumption.

[0405] Step 7:

[0406] Generative AI analyzes the data and identifies higher-than-average consumption at night.

[0407] Step 8:

[0408] The AI ​​generates advice such as "unplug appliances if you are not using them at night."

[0409] Step 9:

[0410] The server transmits the generated advice to the terminal.

[0411] Step 10:

[0412] The device will notify the user within the application and display details.

[0413] Step 11:

[0414] The device analyzes the user's voice and facial expressions and uses an emotion engine to check whether the user is feeling stressed.

[0415] Step 12:

[0416] The emotion engine tailors advice as needed, adding encouragement and kind words.

[0417] Step 13:

[0418] The server sends the adjusted advice to the terminal.

[0419] Step 14:

[0420] The device will notify the user within the application and display details.

[0421] Step 15:

[0422] The user inputs into the application that "I unplugged my home appliances overnight."

[0423] Step 16:

[0424] The device sends the feedback to the server.

[0425] Step 17:

[0426] The server analyzes the feedback data and confirms that electricity usage has been reduced.

[0427] Step 18:

[0428] The generation AI will incorporate this feedback as learning data and reflect it in generating the next piece of advice.

[0429] Example 2

[0430] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0431] Existing systems that collect and analyze energy usage data and provide energy-saving advice rarely provide advice that takes users' emotions into consideration, resulting in insufficient user satisfaction and energy reduction effects. Another problem is that there is a lack of a mechanism for continuously improving the performance of energy-saving advice using collected feedback, which limits the effectiveness of the advice.

[0432] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0433] In this invention, the server includes means for collecting energy usage data, means for transmitting the energy usage data, means for storing the transmitted energy usage data, means for preprocessing the stored energy usage data, means for analyzing the preprocessed data and recognizing energy consumption patterns, means for generating energy saving advice based on the analysis results, means for adjusting the generated energy saving advice based on a user's emotions, means for displaying the adjusted energy saving advice, means for collecting feedback on the energy saving advice, and means for analyzing the collected feedback and improving the performance of the energy saving advice generator. This makes it possible to provide energy saving advice that takes the user's emotions into consideration, thereby improving user satisfaction and maximizing energy reduction effects.

[0434] "Energy usage data" refers to data that indicates the amount of energy consumed, such as electricity, gas, and water, in homes and businesses.

[0435] "Collection means" refers to the means for capturing energy usage data, and specifically includes connection to devices such as HEMS (Home Energy Management System).

[0436] The "transmission means" is a means for sending the collected energy usage data to the server via a network.

[0437] The "storage means" is a means for storing and managing the energy usage data sent to the server by the transmission means in a database.

[0438] The "preprocessing means" is a means for complementing missing values ​​and removing noise from the stored energy usage data.

[0439] "Analysis means" refers to means for using the pre-processed data to recognize energy consumption patterns and detect abnormal trends.

[0440] "Generation means" refers to the means for creating specific energy-saving advice based on the analysis results, and includes the use of a generative AI model.

[0441] The "adjustment means" is a means for modifying the content of the generated energy saving advice based on the user's feelings.

[0442] The "display means" is a means for providing the user with the adjusted energy saving advice, and includes displays and notifications within the application.

[0443] The "feedback collection means" is a means for transmitting to the server the effects of advice taken by the user and their impressions.

[0444] The "feedback analysis means" is a means used to analyze collected feedback data and improve the performance of subsequent advice generation.

[0445] The present invention is a system that collects and analyzes energy usage data and provides energy-saving advice based on the results, and combines it with an emotion engine that recognizes the user's emotions. This system makes it possible to improve the efficiency of energy usage in homes and businesses, promote reductions in utility costs, and provide advice that takes the user's emotions into consideration. An embodiment of the present invention is described in detail below.

[0446] System Overview

[0447] The system mainly includes a collection means, a transmission means, a storage means, a pre-processing means, an analysis means, a generation means, an adjustment means, a display means, a feedback collection means, and a feedback analysis means for handling energy usage data. These means are executed by terminals installed in homes and businesses and a server that collects and analyzes data.

[0448] Data collection

[0449] The device connects to a HEMS (Home Energy Management System) installed in a home or business to collect electricity, gas, and water usage data. This connection is via Wi-Fi or wired LAN. The device encrypts the collected energy usage data and sends it to a server over the network.

[0450] Data Receipt and Storage

[0451] The server receives the energy usage data sent from the device and stores it in a database in real time, so that the data is properly categorized and recorded in chronological order.

[0452] Data Preprocessing

[0453] The server extracts energy usage data for the past 30 days from the database and performs preprocessing such as filling in missing values ​​and removing noise. Specifically, it fills in missing data using linear interpolation and average value interpolation, detects abnormally high consumption data, and corrects it based on past trends.

[0454] Data Analysis and Pattern Recognition

[0455] The server inputs the preprocessed data into a generative AI model to recognize energy consumption patterns and abnormal trends. The generative AI model learns past consumption patterns based on a large amount of data, enabling highly accurate pattern recognition.

[0456] Example prompt: "Generate energy conservation advice based on the past 30 days of energy consumption data, specifically nighttime consumption patterns."

[0457] Advice Generation

[0458] The generator generates specific energy-saving advice based on the analysis results, such as "unplug appliances when not in use at night."

[0459] Emotion Recognition and Advice Adjustment

[0460] The device collects the user's feedback, voice, and facial expressions through a camera and microphone, and the emotion engine analyzes them. If the user is feeling stressed, the emotion engine generates advice with encouraging and kind words.

[0461] Advice Delivery

[0462] The server then sends the adjusted energy saving advice to the device, which then notifies the user within the application, for example, by using a push notification.

[0463] Feedback collection

[0464] The user inputs feedback about the effectiveness of the advice within the application. Specifically, the application records the action of "unplugging home appliances overnight." The device then sends the user-entered feedback to the server.

[0465] Feedback Analysis

[0466] The server analyzes the received feedback data and checks whether energy usage has actually been reduced. The AI ​​learns from this feedback data and reflects it in the generation of next advice.

[0467] This not only improves energy efficiency and reduces utility bills, but also improves the user experience by providing emotionally sensitive advice, making energy management in homes and businesses more effective and satisfying.

[0468] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0469] The flow of this system's program processing

[0470] Step 1: Data collection

[0471] The terminal connects to a HEMS (Home Energy Management System).

[0472] Input: Energy usage data from a HEMS installed in the user's home or business

[0473] Output: Collected energy usage data

[0474] Specific operation: Acquires and collects electricity, gas, and water usage data from the HEMS. This data includes timestamps and consumption amounts.

[0475] Step 2: Send data

[0476] The terminal transmits the collected energy usage data to a server.

[0477] Input: Energy usage data obtained from HEMS

[0478] Output: Energy usage data sent to the server

[0479] What it does: Encrypts data and sends it over the internet to a server via Wi-Fi or wired LAN.

[0480] Step 3: Data reception and storage

[0481] The server receives the energy usage data sent from the terminal and stores it in a database.

[0482] Input: Energy usage data sent from the device

[0483] Output: Energy usage data stored in a database

[0484] Specific operation: The received data is stored in a database in real time and categorized by household or business.

[0485] Step 4: Preprocessing the data

[0486] The server extracts energy usage data for the last 30 days from the database, fills in missing values, and removes noise.

[0487] Input: Energy usage data extracted from a database

[0488] Output: Preprocessed energy usage data

[0489] Specific operation: For the extracted data, missing values ​​are filled in using linear interpolation or mean value interpolation, and abnormal high-consumption data is detected and corrected.

[0490] Step 5: Data analysis and pattern recognition

[0491] The server feeds the pre-processed data into a generative AI model to recognize energy consumption patterns and unusual trends.

[0492] Input: Preprocessed energy usage data

[0493] Output: Recognized energy consumption patterns and unusual trends

[0494] Specific operation: Data is input into the generative AI model to analyze consumption patterns and detect anomalies.

[0495] Example prompt: "Generate energy conservation advice based on the past 30 days of energy consumption data, specifically nighttime consumption patterns."

[0496] Step 6: Advice Generation

[0497] The server generates specific energy-saving advice using the output of the generative AI model.

[0498] Input: Analysis results from a generative AI model

[0499] Output: Specific energy saving advice

[0500] Specific actions: Based on the analysis results, specific advice for energy conservation is generated, such as "unplug appliances at night."

[0501] Step 7: Emotion recognition and advice adjustment

[0502] The device collects the user's feedback, voice, and facial expressions, which are then analyzed by an emotion engine.

[0503] Input: User feedback, voice, facial expressions

[0504] Output: Energy saving advice tailored based on user's emotions

[0505] Specific behavior: The system uses a camera and microphone to recognize the user's emotions and adjusts the content of the energy-saving advice it generates based on the user's emotions. If the user is feeling stressed, it will include encouraging or kind words.

[0506] Step 8: Advice Delivery

[0507] The server sends the adjusted energy saving advice to the terminal, which notifies the user within the application.

[0508] Input: Tailored energy saving advice

[0509] Output: Advice given to the user

[0510] Specific operation: The adjusted advice is sent to the device, and the device notifies the user through the application.

[0511] Step 9: Gather feedback

[0512] The user enters feedback within the application about the effectiveness of the advice they have implemented.

[0513] Input: User feedback on advice taken

[0514] Output: Feedback sent to the device

[0515] Specific actions: Using the application, you can input specific actions and their results. For example, you can input "I unplugged the appliances overnight."

[0516] Step 10: Feedback analysis

[0517] The server analyzes the received feedback data and reflects it in the generation of the next advice.

[0518] Input: Feedback data sent from the device

[0519] Output: More accurate next energy saving advice

[0520] What it does: Analyzes the feedback, checks whether electricity usage has been reduced, and trains the generating AI based on the feedback data, which improves the accuracy of the next advice.

[0521] (Application example 2)

[0522] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0523] Conventional energy management systems were able to collect energy usage data from homes and businesses and provide energy-saving advice, but they did not take user emotions into account when providing advice or collecting and analyzing feedback. This resulted in issues such as a lack of improvement in the user experience and limited effectiveness of advice. Furthermore, in large facilities such as factories, energy consumption patterns are complex, so conventional energy management systems were unable to achieve sufficient energy-saving effects.

[0524] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0525] In this invention, the server includes means for collecting energy usage data, means for transmitting the energy usage data, means for storing the transmitted energy usage data, means for analyzing the stored energy usage data, means for generating energy saving advice based on the analysis results, means for recognizing a user's emotion when generating the energy saving advice and adjusting the content of the advice in accordance with the recognized emotion, and means for collecting feedback on the energy saving advice. This achieves more efficient energy use and reduced utility costs, and also provides advice that takes the user's emotion into consideration, improving the user experience and enabling more satisfying energy management.

[0526] "Energy usage data" refers to data on the amount of energy used, such as electricity, gas, and water, by households, businesses, factories, etc.

[0527] "Means of collection" refers to devices or systems for obtaining energy usage data, specifically sensors and meters.

[0528] The "transmitting means" refers to the communication device or protocol for transferring the collected energy usage data to the server.

[0529] "Means for storage" refers to a database or storage device for retaining and managing the transmitted energy usage data for a long period of time.

[0530] "Means for analysis" refers to software or algorithms used to analyze stored energy usage data and detect consumption patterns and outliers.

[0531] "Means for generating energy-saving advice" refers to a generative AI model or program that generates specific advice recommending energy-efficient ways of using energy based on the analyzed data.

[0532] The "display means" refers to a display, monitor, or application that notifies the user of the generated energy saving advice and visually presents it to the user.

[0533] "Means for recognizing emotions" refers to emotion recognition engines or software that analyze and identify emotions from the user's voice, facial expressions, etc.

[0534] The "means for collecting feedback" refers to an interface or system for collecting information about the results and effects of the energy conservation advice that the user has implemented.

[0535] "In-factory energy usage data" refers to data on the amount of energy consumed by the factory's production lines and each device and equipment.

[0536] A "generative AI model" is an artificial intelligence model that generates new energy-saving advice based on past energy usage data.

[0537] A "prompt sentence" is an input sentence that instructs the generative AI model on what kind of analysis or advice to generate.

[0538] The system that realizes this application example is a combination of specific hardware and software for energy management and emotion recognition within a factory. The main components are sensors and meters for collecting energy usage data, a server for managing and analyzing the data, a robot terminal that provides an interface with the user, and an emotion engine for recognizing emotions. A detailed example of the system is described below.

[0539] Hardware and Software

[0540] 1. Sensors and meters

[0541] Sensors: Installed on each production line in the factory, they collect real-time data on electricity, gas, water, etc. For example, smart meters and environmental monitoring sensors are used.

[0542] Meter: A meter that measures the energy consumption of each device or equipment, such as a smart electricity meter.

[0543] 2. Server

[0544] Database: A database for centrally storing collected energy usage data. For example, MySQL can be used.

[0545] Analysis software: Analyzes the stored data to detect energy consumption patterns and outliers, for example using data analysis libraries such as Pandas and Scikit-learn.

[0546] Generative AI model: A generative AI model for generating energy-saving advice based on data analysis results. For example, the latest generative AI models such as GPT-4 can be used.

[0547] 3. Robot terminal

[0548] Display: A display to inform and visually present energy saving advice to the user.

[0549] Voice input device: A microphone to transmit the user's voice to the emotion recognition engine.

[0550] Emotion recognition engine: Analyzes the user's emotions and adjusts the advice content. For example, OpenFace or Emotion API is used.

[0551] Data processing details

[0552] The server uses sensors and meters to collect energy usage data from each production line in the factory. The collected data is sent to the server and stored in a database. The stored data is then pre-processed using analytical software to detect energy consumption patterns and outliers.

[0553] The analyzed data is input into a generative AI model, which generates specific energy-saving advice. Furthermore, during this generation process, an emotion recognition engine is used to analyze the user's emotions and tailor the advice content based on the recognized emotions. For example, if the user is feeling stressed, encouraging or kind words may be added. This tailored advice is then displayed on the robot terminal's display, informing the user.

[0554] Specific examples

[0555] For example, if analysis reveals that a factory's production line consumes a lot of energy at night, the generative AI model generates advice recommending "replacing nighttime lighting with LEDs." Furthermore, if the emotion recognition engine determines that the operator is stressed, the advice message will include a gentle note: "You don't have to make all the changes at once." This tailored advice is then displayed on the robot terminal's display and provided to the operator.

[0556] Prompt Sentence Examples

[0557] "Please suggest energy-saving measures based on the energy usage data from the last 30 days. Please identify any abnormal consumption patterns and provide specific advice accordingly. If the user is feeling stressed, please offer encouragement and kind words."

[0558] By using this system, factories can achieve more efficient energy use, reducing utility costs and improving the user experience. Furthermore, by analyzing feedback data and reflecting it in the next advice generation, the system can continuously learn and improve the accuracy of its advice.

[0559] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0560] Step 1:

[0561] The terminal collects energy usage data from sensors and meters within the factory. The input is real-time data from the sensors, and the output is the collected energy data. This data includes the electricity, gas, and water usage of each production line and piece of equipment. Specifically, the terminal reads data from the sensors at regular intervals (e.g., every three hours) and temporarily stores the data.

[0562] Step 2:

[0563] The terminal sends the collected energy usage data to the server. The input is the energy data collected in step 1 above, and the output is the data sent to the server. Specifically, the terminal periodically packets the collected data and sends it to the server via the network.

[0564] Step 3:

[0565] The server receives energy data sent from the device and stores it in a database. The input is the sent energy data, and the output is the data organized and stored in the database. Specifically, the server verifies the data it receives, organizes it in chronological order, and stores it.

[0566] Step 4:

[0567] The server preprocesses the stored energy data. The input is the energy data stored in the database, and the output is the preprocessed data. This preprocessing includes filling in missing values ​​and removing noise. Specifically, the server cleans the data using a data processing library such as Pandas.

[0568] Step 5:

[0569] The server inputs the preprocessed data into a generative AI model for analysis. The input is the preprocessed energy data, and the output is the analysis results. Specifically, the server uses a machine learning library such as Scikit-learn to pass the data to the model and identify energy consumption patterns and anomalies.

[0570] Step 6:

[0571] The server creates energy-saving advice generated by the generative AI model. The input is the analysis results, and the output is specific energy-saving advice. Specifically, the server uses the generative AI model to generate specific measures for saving electricity and water. For example, advice such as "replace nighttime lighting with LEDs" is generated.

[0572] Step 7:

[0573] The device analyzes the user's voice and facial expressions and uses an emotion recognition engine to recognize the user's emotions. The input is the user's voice and facial expression data, and the output is the recognized emotion. Specifically, the device analyzes the voice and images using the Emotion API, etc., to identify the emotion.

[0574] Step 8:

[0575] The server adjusts the advice content based on the recognized emotion. The input is energy-saving advice and the recognized emotion data, and the output is the adjusted advice. Specifically, the server adds encouraging or kind words to the advice message. For example, if the user is feeling stressed, the server adds the message, "You don't need to make all the changes at once."

[0576] Step 9:

[0577] The device notifies the user of the adjusted energy-saving advice and displays details. The input is the adjusted advice, and the output is the state notified to the user. Specifically, the device displays the advice on the display and notifies the user in a format that is easy for the user to understand.

[0578] Step 10:

[0579] The user provides feedback on the results of the energy-saving advice they have implemented to the terminal. The input is the user's feedback information, and the output is feedback data. Specifically, the user inputs the results of implementing the advice through the application.

[0580] Step 11:

[0581] The terminal sends feedback from the user to the server. The input is the feedback data, and the output is the data sent to the server. Specifically, the terminal packetizes the feedback information and sends it to the server via the network.

[0582] Step 12:

[0583] The server analyzes the feedback data and evaluates the effectiveness of the advice. The input is the sent feedback data, and the output is the evaluation result. Specifically, the server analyzes the feedback using a data analysis tool and evaluates the effectiveness of the advice. The result is reflected in the next advice generation.

[0584] In this way, a series of steps will result in a system that will improve energy efficiency within the factory and also enhance the user experience.

[0585] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0586] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0587] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0588] [Second embodiment]

[0589] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0590] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0591] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0592] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0593] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0594] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0595] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0596] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0597] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0598] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0599] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0600] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0601] System Overview

[0602] The present invention relates to a system that collects and analyzes energy usage data and provides energy-saving advice based on the results. This enables homes and businesses to use energy more efficiently and promotes reductions in utility costs. The system mainly includes a collection means, transmission means, storage means, analysis means, generation means, display means, and feedback collection means for handling energy usage data.

[0603] Program processing explanation

[0604] 1. Data Collection

[0605] The terminals are connected to the HEMS (Home Energy Management System) to collect electricity, gas, and water usage data from each home or business.

[0606] The terminal periodically (e.g., every three hours) obtains usage data from the HEMS and sends it to the server.

[0607] 2. Data Receipt and Storage

[0608] The server receives the energy usage data transmitted from the terminal.

[0609] The server stores the received data in a database, organizing it by household and business and recording it in chronological order.

[0610] 3. Data Preprocessing

[0611] The server extracts energy usage data for a certain period of time (e.g., the last 30 days) from the database.

[0612] The server performs pre-processing to complement missing values ​​in the extracted data and remove noise. Specifically, it detects and corrects abnormally high-consumption data.

[0613] 4. Data Analysis and Pattern Recognition

[0614] The server inputs the preprocessed data into the generative AI model.

[0615] Generative AI uses past data to recognize patterns of energy consumption and unusual trends.

[0616] 5. Advice Generation

[0617] The server generates specific energy-saving advice based on the output from the generative AI model.

[0618] Generative AI provides advice tailored to individual consumption patterns.

[0619] 6. Advice Delivery

[0620] The server transmits the generated energy saving advice to the terminal.

[0621] The terminal notifies the user of the received advice and displays it within the application.

[0622] 7. Gathering Feedback

[0623] The user receives feedback within the application about the effectiveness of the advice they have implemented, for example, by inputting a specific action such as "I changed the refrigerator temperature setting."

[0624] The terminal transmits the feedback input by the user to the server.

[0625] 8. Feedback Analysis

[0626] The server analyzes the received feedback data and evaluates the effectiveness of the advice.

[0627] The generative AI learns from the feedback data and reflects it in future advice generation.

[0628] Specific examples

[0629] Examples of reducing electricity usage

[0630] The terminal obtains 24 hours of electricity usage data from the HEMS every three hours and sends it to the server.

[0631] The server receives the sent data and stores it in the database for "Home ID_1234."

[0632] The server extracts the electricity usage data for the last 30 days from the database for "household ID_1234" and detects and corrects any invalid data (e.g., extremely high consumption).

[0633] The generation AI analyzes the data for "household ID_1234" and determines that nighttime consumption is higher than average.

[0634] The AI ​​generates advice recommending that "if you are not using home appliances at night, unplug them" and outputs it to the server.

[0635] The server sends the generated advice to the terminal, which notifies the user within the application and displays the details.

[0636] The user inputs into the application that "I unplugged my home appliances overnight," and the device sends feedback to the server.

[0637] The server analyzes the feedback data and confirms that electricity usage has been reduced.

[0638] The generation AI will incorporate this feedback as learning data and reflect it in generating the next piece of advice.

[0639] Example of reducing gas usage

[0640] The terminal periodically obtains gas usage data from the HEMS and sends it to the server.

[0641] The server receives the data and stores it in a database.

[0642] The server preprocesses the data and adjusts for outliers.

[0643] Generative AI analyzes the data and discovers that certain cooking methods increase gas consumption.

[0644] The generative AI generates advice such as "Use a pressure cooker to reduce cooking time" and outputs it to the server.

[0645] The server sends the generated advice to the terminal, which notifies the user within the application and displays the recommendation.

[0646] The user enters "I used a pressure cooker" in the application, and the device sends the feedback to the server.

[0647] The server analyzes the feedback data and confirms that gas usage has been reduced.

[0648] The generation AI will incorporate this feedback as learning data and reflect it in generating the next piece of advice.

[0649] Through the above processing, the present invention can improve the efficiency of energy use and contribute to reducing utility costs for homes and businesses.

[0650] The processing flow will be explained below.

[0651] Program processing steps

[0652] 1. Data Collection

[0653] Step 1:

[0654] The device connects to the HEMS and collects electricity, gas, and water usage data, including energy usage information for homes and businesses.

[0655] Step 2:

[0656] The device sends the collected data to the server at specified intervals (e.g., every 3 hours).

[0657] 2. Data Receipt and Storage

[0658] Step 3:

[0659] The server receives the energy usage data transmitted from the terminal.

[0660] Step 4:

[0661] The server stores the received data in a database and organizes it by household or business. The data is recorded in chronological order.

[0662] 3. Data Preprocessing

[0663] Step 5:

[0664] The server extracts energy usage data for a certain period (e.g., the last 30 days) from the database.

[0665] Step 6:

[0666] The server completes the extracted data, removes noise, and detects abnormal high-consumption data and corrects it as necessary.

[0667] 4. Data Analysis and Pattern Recognition

[0668] Step 7:

[0669] The server inputs the preprocessed data into the generative AI model.

[0670] Step 8:

[0671] Generative AI recognizes patterns and unusual trends in energy consumption, identifying increases or decreases in consumption for specific times of day or dates.

[0672] 5. Advice Generation

[0673] Step 9:

[0674] The server receives the output of the generative AI model and generates specific energy-saving advice.

[0675] Step 10:

[0676] Generative AI provides customized advice based on individual consumption patterns and usage.

[0677] 6. Advice Delivery

[0678] Step 11:

[0679] The server transmits the generated energy saving advice to the terminal.

[0680] Step 12:

[0681] The terminal notifies the user of the received advice and displays it within the application.

[0682] 7. Gathering Feedback

[0683] Step 13:

[0684] Users can provide feedback within the application about the effectiveness of the advice they have implemented, by entering specific actions and results.

[0685] Step 14:

[0686] The terminal transmits the feedback input by the user to the server.

[0687] 8. Feedback Analysis

[0688] Step 15:

[0689] The server analyzes the received feedback data and evaluates the effectiveness of the advice.

[0690] Step 16:

[0691] The generative AI learns from the feedback data and reflects it in the next advice generation, aiming to provide improved advice.

[0692] Specific examples

[0693] Examples of reducing electricity usage

[0694] Step 1:

[0695] The terminal obtains 24 hours of electricity usage data from the HEMS every three hours.

[0696] Step 2:

[0697] The terminal transmits the collected data to the server.

[0698] Step 3:

[0699] The server receives the transmitted data.

[0700] Step 4:

[0701] The server stores the received data in a database.

[0702] Step 5:

[0703] The server extracts the electricity usage data for the last 30 days from the database for "Home ID_1234."

[0704] Step 6:

[0705] The server detects and corrects abnormally high data consumption.

[0706] Step 7:

[0707] The generative AI analyzes the data and identifies higher-than-average consumption at night.

[0708] Step 8:

[0709] The AI ​​generates advice such as "unplug appliances if you are not using them at night."

[0710] Step 9:

[0711] The server transmits the generated advice to the terminal.

[0712] Step 10:

[0713] The device will notify the user within the application and display details.

[0714] Step 11:

[0715] The user inputs into the application that "I unplugged my home appliances overnight."

[0716] Step 12:

[0717] The terminal sends the feedback to the server.

[0718] Step 13:

[0719] The server analyzes the feedback data and verifies that electricity usage has been reduced.

[0720] Step 14:

[0721] The generative AI incorporates the feedback as learning data and reflects it in generating the next piece of advice.

[0722] Example 1

[0723] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0724] Conventional energy management systems are limited to collecting and simply analyzing energy usage data, making it difficult to efficiently provide specific energy-saving advice to users. They also lack a mechanism for collecting feedback and reflecting the results in the analysis. This creates the challenge of being unable to provide timely and appropriate energy-saving advice tailored to the user's energy consumption patterns.

[0725] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0726] In this invention, the server includes means for periodically transmitting energy usage data, means for receiving the transmitted energy usage data, means for storing the received energy usage data, means for extracting and preprocessing the stored energy usage data, means for inputting the preprocessed data into a generative AI model to recognize energy consumption patterns, means for generating energy saving advice based on the recognition results, means for transmitting the generated energy saving advice, means for displaying the transmitted energy saving advice, means for collecting feedback on the energy saving advice, and means for analyzing the collected feedback and updating the generative AI model. This integrates the collection and analysis of energy usage data, advice generation, feedback collection, and model updating, making it possible to provide users with timely and effective energy saving advice.

[0727] "Energy usage data" refers to data showing the amount of electricity, gas, water, etc. used, and is a measurement of the specific energy usage status consumed within homes and businesses.

[0728] "HEMS" is an abbreviation for Home Energy Management System, a system for optimizing and managing energy usage within the home.

[0729] A "generative AI model" refers to an artificial intelligence model that performs statistical analysis and machine learning based on various data, and uses the knowledge gained from this to generate new information.

[0730] A "prompt sentence" is an input sentence for a specific generative AI model, and is a sentence that instructs the model to produce a specific output.

[0731] "Data preprocessing" refers to the process carried out before data analysis, and refers to the process of preparing data by filling in missing values ​​and removing noise, etc.

[0732] "Feedback" is information that records the user's response to and results of implementing advice provided by the system, and reflects this information back into the system.

[0733] The present invention relates to a system that collects and analyzes energy usage data and provides energy conservation advice based on the results. This invention makes it possible to improve the efficiency of energy usage in homes and businesses and promote reductions in utility costs. Specifically, this system operates according to the following steps.

[0734] System Configuration

[0735] The system mainly includes the following components:

[0736] 1. Collection Method

[0737] 2. Transmission Method

[0738] 3. Preservation means

[0739] 4. Pretreatment Methods

[0740] 5. Analysis method

[0741] 6. Advice Generation Methods

[0742] 7. Display means

[0743] 8. Feedback Collection Methods

[0744] Details of each component are shown below.

[0745] Collection Method

[0746] The device collects energy usage data. Specifically, the device connects to a HEMS (Home Energy Management System) and obtains electricity, gas, and water usage data every three hours. For example, the device obtains the data using the HEMS API ("GET / energy_usage"). The device temporarily stores the data in local storage and prepares to send it to the server later.

[0747] Transmission method

[0748] The device sends the temporarily stored energy usage data to the server using MQTT or REST API. For example, the data to be sent is in JSON format and is assigned an identifier such as "Home ID_1234."

[0749] Preservation means

[0750] The server receives the energy usage data sent from the device and stores it in a NoSQL database (for example, MongoDB). The server stores the received data separately for each household and each business, and manages it in chronological order, which allows for efficient data access.

[0751] Pretreatment means

[0752] The server extracts the most recent 30 days' worth of data from the database, imputes missing values, and performs preprocessing to remove noise. Specifically, the data is cleansed using Python's pandas library, and outliers are corrected using algorithms such as K-Nearest Neighbor (KNN).

[0753] Analysis means

[0754] The server inputs the preprocessed data into a generative AI model to recognize energy consumption patterns and abnormal trends. An example of the generative AI model is OpenAI's GPT. An example prompt is "Please identify abnormal patterns and signs of energy saving from the energy usage data for the last 30 days for household ID_1234."

[0755] Advice Generation Method

[0756] Based on the analysis results, the generation AI generates energy-saving advice tailored to individual consumption patterns. For example, it may generate advice recommending that appliances be unplugged if not in use at night. The generated advice is sent back to the server, which then formats the information in JSON format and stores it back in the database.

[0757] Display means

[0758] The server sends the generated energy-saving advice to the device. The device saves the received advice in local storage and notifies the user within the application. For example, a smartphone push notification can be used to notify the user that "new energy-saving advice is available."

[0759] Feedback collection methods

[0760] The user acts on the advice and provides feedback on the results within the application. Specifically, the user inputs something like, "I changed the refrigerator's set temperature." The device receives this feedback and sends it to the server. The sent feedback is stored in a feedback database on the server.

[0761] Feedback Analysis

[0762] The server analyzes the received feedback data and evaluates the effectiveness of the advice. The generative AI model learns from this feedback data and reflects it in generating new advice. For example, it learns how much power consumption was reduced by changing the refrigerator's set temperature, and reflects this in the next advice.

[0763] Specific examples

[0764] A specific example of operation is shown below.

[0765] 1. The device obtains electricity usage data from the HEMS every three hours and sends the data to the server.

[0766] 2. The server receives the data and stores it in the database as "Home ID_1234".

[0767] 3. The server extracts and preprocesses the data for the last 30 days.

[0768] 4. The generation AI performs an analysis based on the prompt: "Please identify any abnormal patterns or signs of energy saving from the last 30 days of energy usage data for household ID_1234."

[0769] 5. The AI ​​generates advice such as "It is recommended to unplug the device at night" and sends it to the server.

[0770] 6. The server sends the advice to the device, and the device notifies the user via push notification that "New energy saving advice is available."

[0771] 7. The user enters into the application that "I unplugged my appliances overnight," and the device sends feedback to the server.

[0772] 8. The server analyzes the feedback, and the AI ​​uses it to generate the next piece of advice.

[0773] As described above, the present invention can improve the efficiency of energy use and contribute to reducing utility costs for homes and businesses.

[0774] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0775] Step 1: Data collection

[0776] The device connects to the HEMS and obtains electricity, gas, and water usage data every three hours. Specifically, it accesses the HEMS API endpoint "GET / energy_usage" to receive the latest energy usage data. The input is sensor data from the HEMS, and the output is usage data saved in the device's temporary storage. For example, data in the format "Electricity usage: 10kWh" is saved.

[0777] Step 2: Send data

[0778] The energy usage data collected by the device is sent to the server. The data is sent periodically (for example, every three hours) using MQTT or REST API. The input is temporarily stored usage data, and the output is the data to be sent to the server. Specifically, data such as "electricity usage: 10kWh, gas usage: 5m3, water usage: 100L" is structured in JSON format and sent to the server.

[0779] Step 3: Data reception and storage

[0780] The server receives the data sent from the device and stores it in a NoSQL database (for example, MongoDB). The input is the JSON data sent from the device, and the output is the energy usage data stored in the database. Specifically, the received data is organized by household and company, and stored using an identifier such as "Home ID_1234."

[0781] Step 4: Preprocessing the data

[0782] The server extracts the last 30 days' worth of energy usage data from the database. It then performs pre-processing to fill in missing values ​​in the extracted data and remove noise. The input is energy usage data from the database (e.g., the last 30 days' worth of data for "household ID_1234"), and the output is pre-processed, clean data. Specifically, it uses Python's pandas library to fill in missing values ​​and the KNN algorithm to detect and correct outliers.

[0783] Step 5: Data analysis and pattern recognition

[0784] The server inputs the preprocessed data into the generative AI model. The input is the preprocessed data (e.g., energy usage data for the last 30 days for "household ID_1234"), and the output is consumption patterns and abnormal trends recognized by the generative AI model. Specifically, the generative AI is prompted with the following prompt: "Please identify abnormal patterns and signs of energy saving from the energy usage data for the last 30 days for household ID_1234."

[0785] Step 6: Advice Generation

[0786] The generative AI generates energy-saving advice based on the analysis results. The input is the output of the generative AI model (e.g., consumption patterns and abnormal trends), and the output is specific energy-saving advice. For example, the generated advice recommends "unplugging appliances when not in use at night." This advice is sent to the server, which then formats the advice content in JSON format and stores it in a database.

[0787] Step 7: Advice Delivery

[0788] The server sends the generated energy-saving advice to the device. The input is the generated advice (e.g., "Unplug appliances if they are not in use at night"), and the output is the data sent to the device. The device saves the received advice in local storage and notifies the user within the application. Specifically, it uses a push notification on the smartphone to notify the user that "new energy-saving advice is available."

[0789] Step 8: Gather feedback

[0790] The results of the user's execution of advice within the application are fed back. The input is information about the user's actions (e.g., "I changed the refrigerator's set temperature"), and the output is the feedback data. The device receives feedback from the user and sends it to the server. Specifically, it formats the user's input in JSON format and sends a POST request to the server.

[0791] Step 9: Analyze feedback

[0792] The server analyzes the received feedback data and evaluates the effectiveness of the advice. The input is the feedback data (e.g., "As a result of changing the set temperature, power consumption decreased by 5%), and the output is the evaluation result. The generation AI learns from this feedback data and reflects it in generating new advice. For example, it learns that "power consumption at night has decreased" and uses this information to generate the next piece of advice.

[0793] The above are the specific processing steps of this system.

[0794] (Application example 1)

[0795] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0796] In modern brick-and-mortar stores, efficient energy use is important from the perspective of environmental protection and cost reduction. However, in reality, there is a lack of means to grasp energy usage patterns in detail and receive efficient energy-saving advice. As a result, store managers are unable to implement appropriate energy-saving measures, and wasteful energy consumption continues. The objective of this invention is to promote efficient energy use and reduce wasteful energy consumption by analyzing energy usage data in brick-and-mortar stores in detail and providing specific energy-saving advice.

[0797] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0798] In this invention, the server includes means for collecting energy usage data, means for transmitting the energy usage data, means for storing the transmitted energy usage data, means for preprocessing the stored energy usage data, means for analyzing the stored energy usage data, means for generating energy saving advice based on the analysis results, means for generating the generated energy saving advice for specific actions in the store, means for displaying the generated energy saving advice, and means for collecting feedback on the energy saving advice. This makes it possible to understand energy usage patterns in the physical store in detail and to specifically implement efficient energy saving measures.

[0799] "Energy usage data" refers to data that records the amount of energy consumed in a physical store, such as electricity, gas, and water.

[0800] The "transmission means" is a function for transmitting collected energy usage data to a server.

[0801] The "storage means" is a function for storing the transmitted energy usage data in a storage device such as a database.

[0802] The "preprocessing means" is a function for performing preprocessing such as complementing missing values ​​and removing outliers on stored energy usage data.

[0803] The "analysis means" is a function for analyzing the pre-processed energy usage data and identifying patterns and anomalies in energy consumption.

[0804] The "generation means" is a function for generating energy saving advice based on the analysis results.

[0805] The "display means" is a function for presenting the generated energy saving advice to the operator of the physical store.

[0806] The "feedback collection means" is a function for collecting feedback regarding the energy saving advice that the operator has implemented.

[0807] "Generative AI" is an artificial intelligence model that analyzes energy usage data and generates specific energy-saving advice.

[0808] A "prompt sentence" is an instruction sentence that provides specific energy-saving advice to the generation AI.

[0809] "In-store energy saving advice" is advice that recommends specific energy saving actions for specific equipment or activities in a physical store.

[0810] The present invention relates to a system that collects and analyzes energy usage data and provides energy conservation advice based on the results, thereby making it possible to improve the efficiency of energy usage in physical stores and promote reductions in utility costs.

[0811] The system includes a collection means, a transmission means, a storage means, a pre-processing means, an analysis means, a generation means, a display means, and a feedback collection means for handling energy usage data.

[0812] Specific hardware and software configuration:

[0813] The terminal, a smartphone or tablet, connects to the HEMS (Home Energy Management System) in the physical store and collects energy usage data.

[0814] The server receives the collected energy usage data and stores it in a database, where it is organized by household and business and recorded in chronological order.

[0815] The preprocessing means extracts energy usage data for a certain period from the database and performs preprocessing such as filling in missing values ​​and removing outliers, thereby ensuring the reliability of the data.

[0816] The analytics tool feeds the pre-processed data into a generative AI model to identify patterns and unusual trends in energy consumption.

[0817] The AI ​​that generates the energy-saving advice is based on the analysis results and is specific to the user's consumption patterns.

[0818] The terminal as a display means notifies the user of the generated energy saving advice and displays it within the application.

[0819] The feedback collection means collects feedback regarding the effect of the energy saving advice implemented by the user and transmits the feedback to the server.

[0820] Software used and detailed data processing:

[0821] Data collection: The device acquires energy usage data from the HEMS. The data is sent to the server periodically (e.g., every three hours).

[0822] Data storage and preprocessing: The server stores the data in a database and preprocesses it, including imputing missing values ​​and removing outliers.

[0823] Data analysis and pattern recognition: Analyze the pre-processed data and recognize consumption patterns and unusual trends from historical data. Generative AI models take on this role and automate the process.

[0824] Generating energy-saving advice: The AI ​​generates energy-saving advice based on the analysis results. For example, it may generate advice such as, "Since power consumption is high at night, we recommend turning off devices outside of business hours."

[0825] User notification and feedback collection: The generated advice is notified to the user and feedback is collected within the application. The feedback is sent back to the server and reflected in subsequent advice generation.

[0826] Examples and prompts:

[0827] Example: Identifying high consumption at night and making specific recommendations on which appliances in a store should be turned off outside of business hours.

[0828] Example prompt: "Our store consumes a lot of electricity at night. Please have your Generative AI suggest which devices in our store should be turned off after hours."

[0829] In this way, the present invention efficiently manages energy usage in physical stores and provides specific energy-saving advice, enabling operators of physical stores to implement appropriate energy-saving measures.

[0830] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0831] Step 1:

[0832] Data collection

[0833] The terminal obtains electricity, gas, and water usage data every three hours from the physical store's HEMS (Home Energy Management System). The input is real-time energy usage data from the HEMS, and the output is collected energy usage data, which is used for further analysis.

[0834] Step 2:

[0835] Data transmission

[0836] The terminal periodically transmits the collected energy usage data to the server. The input is the collected energy usage data, and the output is the energy usage data transmitted to the server. The transmitted data is stored on the server.

[0837] Step 3:

[0838] Data storage

[0839] The server receives the transmitted energy usage data and stores it in a database. The input is the energy usage data transmitted to the server, and the output is the energy usage data stored in the database. This data is used for later analysis.

[0840] Step 4:

[0841] Data Preprocessing

[0842] The server extracts energy usage data for a certain period of time (e.g., the last 30 days) from the database, and completes missing values ​​and removes outliers. The input is the energy usage data stored in the database, and the output is the corrected and preprocessed energy usage data. Specifically, it completes missing values ​​and removes outliers using the 3 sigma rule.

[0843] Step 5:

[0844] Data Analysis and Pattern Recognition

[0845] The server inputs the preprocessed energy usage data into a generative AI model to identify energy consumption patterns and anomalies. The input is the preprocessed energy usage data, and the output is the recognition results of consumption patterns and anomalies. Specifically, it finds consumption patterns from past data and identifies abnormal trends.

[0846] Step 6:

[0847] Energy saving advice generation

[0848] The server uses a generative AI model based on the analysis results to generate energy-saving advice. The input is the consumption pattern and anomaly recognition results, and the output is the generated energy-saving advice. For example, it generates specific advice such as "Since consumption is high at night, it is recommended that you turn off devices outside of business hours."

[0849] Step 7:

[0850] Advice display

[0851] The device notifies the user of the generated energy-saving advice and displays it within the application. The input is the generated energy-saving advice, and the output is the advice displayed to the user. Specifically, a notification is sent to the user's smartphone or tablet.

[0852] Step 8:

[0853] Feedback collection

[0854] The user inputs feedback about the effectiveness of the energy-saving advice they have implemented through the application. The input is the user's feedback, and the output is feedback data. Specifically, the user inputs an action such as "I unplugged my home appliances at night" into the application.

[0855] Step 9:

[0856] Feedback Analysis

[0857] The server analyzes the received feedback data, evaluates the effectiveness of the advice, and provides feedback to the generation AI model. The input is the feedback data sent by the user, and the output is the effectiveness evaluation result. Specifically, the effectiveness of the energy-saving advice is evaluated and reflected in the generation of the next advice.

[0858] By following these steps, energy usage in physical stores can be managed efficiently and specific energy-saving advice can be provided.

[0859] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0860] System Overview

[0861] The present invention combines a system that collects and analyzes energy usage data and provides energy-saving advice based on the results with an emotion engine that recognizes user emotions. This makes it possible to improve the efficiency of energy use in homes and businesses, promote reductions in utility bills, and provide advice that takes user emotions into consideration. The system mainly includes a collection means for handling energy usage data, a transmission means, a storage means, an analysis means, a generation means, a display means, an emotion engine, and a feedback collection means.

[0862] Program processing explanation

[0863] 1. Data Collection

[0864] The terminals are connected to the HEMS (Home Energy Management System) to collect electricity, gas, and water usage data from each home or business.

[0865] The terminal periodically (e.g., every three hours) obtains usage data from the HEMS and sends it to the server.

[0866] 2. Data Receipt and Storage

[0867] The server receives the energy usage data transmitted from the terminal.

[0868] The server stores the received data in a database, organizing it by household and business and recording it in chronological order.

[0869] 3. Data Preprocessing

[0870] The server extracts energy usage data for a certain period (e.g., the last 30 days) from the database.

[0871] The server performs pre-processing to complement missing values ​​in the extracted data and remove noise. Specifically, it detects and corrects abnormally high-consumption data.

[0872] 4. Data Analysis and Pattern Recognition

[0873] The server inputs the preprocessed data into the generative AI model.

[0874] Generative AI uses past data to recognize patterns of energy consumption and unusual trends.

[0875] 5. Advice Generation

[0876] The server generates specific energy-saving advice based on the output from the generative AI model.

[0877] Generative AI provides advice tailored to individual consumption patterns.

[0878] 6. Emotion recognition and advice adjustment

[0879] The device analyzes the user's feedback, voice, and facial expressions to recognize the user's emotions.

[0880] The emotion engine tailors the generated energy-saving advice based on the recognized emotion: for example, if the user is feeling stressed, it generates a message containing encouraging and kind words.

[0881] 7. Advice Delivery

[0882] The server transmits the adjusted energy saving advice to the terminal.

[0883] The terminal notifies the user of the received advice and displays it within the application.

[0884] 8. Collecting Feedback

[0885] Users can provide feedback within the application about the effectiveness of the advice they have implemented, by entering specific actions and results.

[0886] The terminal transmits the feedback input by the user to the server.

[0887] 9. Feedback Analysis

[0888] The server analyzes the received feedback data and evaluates the effectiveness of the advice.

[0889] The generation AI learns from the feedback data and reflects it in generating the next piece of advice.

[0890] Specific examples

[0891] Examples of reducing electricity usage

[0892] System Operation

[0893] The device retrieves 24 hours of electricity usage data from the HEMS every three hours.

[0894] The terminal transmits the collected data to the server.

[0895] The server receives the data and stores it in a database.

[0896] The server extracts the electricity usage data for the last 30 days from the database for "Home ID_1234."

[0897] The server detects and corrects abnormally high data consumption.

[0898] Generative AI analyzes the data and identifies higher-than-average consumption at night.

[0899] The AI ​​generates advice such as "unplug appliances if you are not using them at night."

[0900] The device analyzes the user's feedback, voice, and facial expressions, and uses an emotion engine to check whether the user is feeling stressed.

[0901] The emotion engine tailors advice as needed, adding encouragement and kind words.

[0902] The server sends the tailored advice to the terminal, which notifies the user within the application and displays the details.

[0903] The user inputs into the application that "I unplugged my home appliances overnight."

[0904] The device sends the feedback to the server.

[0905] The server analyzes the feedback data and confirms that electricity usage has been reduced.

[0906] The generation AI will incorporate this feedback as learning data and reflect it in generating the next piece of advice.

[0907] The system not only improves energy efficiency and reduces utility bills, but also enhances the user experience by providing emotionally sensitive advice, making energy management in homes and businesses more effective and satisfying.

[0908] The processing flow will be explained below.

[0909] Program processing steps

[0910] 1. Data Collection

[0911] Step 1:

[0912] The device connects to the HEMS and collects electricity, gas, and water usage data, which is obtained from sensors and meters installed in each home or business.

[0913] Step 2:

[0914] The device sends the collected usage data to the server at specified intervals (e.g., every 3 hours).

[0915] 2. Data Receipt and Storage

[0916] Step 3:

[0917] The server receives the energy usage data transmitted from the terminal.

[0918] Step 4:

[0919] The server stores the received data in a database, organizing it by household and business and recording it in chronological order.

[0920] 3. Data Preprocessing

[0921] Step 5:

[0922] The server detects missing values ​​in the energy usage data and completes them as necessary.

[0923] Step 6:

[0924] The server scrutinizes the data to remove abnormal values ​​and noise, and detects and corrects abnormally high consumption data.

[0925] 4. Data Analysis and Pattern Recognition

[0926] Step 7:

[0927] The server inputs the preprocessed data into the generative AI model.

[0928] Step 8:

[0929] Generative AI recognizes patterns and unusual trends in energy consumption, for example, identifying sudden increases or decreases in consumption on certain days or during certain times of the day.

[0930] 5. Advice Generation

[0931] Step 9:

[0932] The server generates specific advice for energy conservation based on the analysis results of the generation AI.

[0933] Step 10:

[0934] The generative AI provides customized advice based on consumption patterns, such as "unplug appliances if you're not using them overnight" or "use a pressure cooker to reduce cooking time."

[0935] 6. Emotion recognition and advice adjustment

[0936] Step 11:

[0937] The device analyzes the user's feedback, voice, and facial expressions to recognize the user's emotions. Technologies used include voice recognition and facial recognition.

[0938] Step 12:

[0939] The emotion engine adjusts the energy-saving advice based on the recognized emotion. For example, if the user is feeling stressed, the engine generates advice that includes encouraging and kind words.

[0940] 7. Advice Delivery

[0941] Step 13:

[0942] The server transmits the adjusted energy saving advice to the terminal.

[0943] Step 14:

[0944] The device will notify the user and provide detailed advice within the application, which the user can review at any time.

[0945] 8. Collecting Feedback

[0946] Step 15:

[0947] The user provides feedback on the effectiveness of the advice they received within the application and whether they implemented it. For example, they can input information such as "I changed the refrigerator temperature setting" or "I used a pressure cooker."

[0948] Step 16:

[0949] The terminal transmits the feedback input by the user to the server.

[0950] 9. Feedback Analysis

[0951] Step 17:

[0952] The server analyzes the received feedback data and evaluates the effectiveness of the advice, for example, checking whether electricity usage has been reduced.

[0953] Step 18:

[0954] The AI ​​learns from the feedback data and reflects it in the next advice generation, allowing it to provide more effective and tailored advice to the user.

[0955] Specific examples

[0956] Reducing electricity consumption

[0957] Step 1:

[0958] The device retrieves 24 hours of electricity usage data from the HEMS every three hours.

[0959] Step 2:

[0960] The terminal transmits the collected data to the server.

[0961] Step 3:

[0962] The server receives the transmitted data.

[0963] Step 4:

[0964] The server stores the received data in a database.

[0965] Step 5:

[0966] The server extracts the electricity usage data for the last 30 days from the database for "Home ID_1234."

[0967] Step 6:

[0968] The server detects and corrects abnormally high data consumption.

[0969] Step 7:

[0970] Generative AI analyzes the data and identifies higher-than-average consumption at night.

[0971] Step 8:

[0972] The AI ​​generates advice such as "unplug appliances if you are not using them at night."

[0973] Step 9:

[0974] The server transmits the generated advice to the terminal.

[0975] Step 10:

[0976] The device will notify the user within the application and display details.

[0977] Step 11:

[0978] The device analyzes the user's voice and facial expressions and uses an emotion engine to check whether the user is feeling stressed.

[0979] Step 12:

[0980] The emotion engine tailors advice as needed, adding encouragement and kind words.

[0981] Step 13:

[0982] The server sends the adjusted advice to the terminal.

[0983] Step 14:

[0984] The device will notify the user within the application and display details.

[0985] Step 15:

[0986] The user inputs into the application that "I unplugged my home appliances overnight."

[0987] Step 16:

[0988] The device sends the feedback to the server.

[0989] Step 17:

[0990] The server analyzes the feedback data and confirms that electricity usage has been reduced.

[0991] Step 18:

[0992] The generation AI will incorporate this feedback as learning data and reflect it in generating the next piece of advice.

[0993] Example 2

[0994] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0995] Existing systems that collect and analyze energy usage data and provide energy-saving advice rarely provide advice that takes users' emotions into consideration, resulting in insufficient user satisfaction and energy reduction effects. Another problem is that there is a lack of a mechanism for continuously improving the performance of energy-saving advice using collected feedback, which limits the effectiveness of the advice.

[0996] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0997] In this invention, the server includes means for collecting energy usage data, means for transmitting the energy usage data, means for storing the transmitted energy usage data, means for preprocessing the stored energy usage data, means for analyzing the preprocessed data and recognizing energy consumption patterns, means for generating energy saving advice based on the analysis results, means for adjusting the generated energy saving advice based on a user's emotions, means for displaying the adjusted energy saving advice, means for collecting feedback on the energy saving advice, and means for analyzing the collected feedback and improving the performance of the energy saving advice generator. This makes it possible to provide energy saving advice that takes the user's emotions into consideration, thereby improving user satisfaction and maximizing energy reduction effects.

[0998] "Energy usage data" refers to data that indicates the amount of energy consumed, such as electricity, gas, and water, in homes and businesses.

[0999] "Collection means" refers to the means for capturing energy usage data, and specifically includes connection to devices such as HEMS (Home Energy Management System).

[1000] The "transmission means" is a means for sending the collected energy usage data to the server via a network.

[1001] The "storage means" is a means for storing and managing the energy usage data sent to the server by the transmission means in a database.

[1002] The "preprocessing means" is a means for complementing missing values ​​and removing noise from the stored energy usage data.

[1003] "Analysis means" refers to means for using the pre-processed data to recognize energy consumption patterns and detect abnormal trends.

[1004] "Generation means" refers to the means for creating specific energy-saving advice based on the analysis results, and includes the use of a generative AI model.

[1005] The "adjustment means" is a means for modifying the content of the generated energy saving advice based on the user's feelings.

[1006] The "display means" is a means for providing the user with the adjusted energy saving advice, and includes displays and notifications within the application.

[1007] The "feedback collection means" is a means for transmitting to the server the effects of advice taken by the user and their impressions.

[1008] The "feedback analysis means" is a means used to analyze collected feedback data and improve the performance of subsequent advice generation.

[1009] The present invention is a system that collects and analyzes energy usage data and provides energy-saving advice based on the results, and combines it with an emotion engine that recognizes the user's emotions. This system makes it possible to improve the efficiency of energy usage in homes and businesses, promote reductions in utility costs, and provide advice that takes the user's emotions into consideration. An embodiment of the present invention is described in detail below.

[1010] System Overview

[1011] The system mainly includes a collection means, a transmission means, a storage means, a pre-processing means, an analysis means, a generation means, an adjustment means, a display means, a feedback collection means, and a feedback analysis means for handling energy usage data. These means are executed by terminals installed in homes and businesses and a server that collects and analyzes data.

[1012] Data collection

[1013] The device connects to a HEMS (Home Energy Management System) installed in a home or business to collect electricity, gas, and water usage data. This connection is via Wi-Fi or wired LAN. The device encrypts the collected energy usage data and sends it to a server over the network.

[1014] Data Receipt and Storage

[1015] The server receives the energy usage data sent from the device and stores it in a database in real time, so that the data is properly categorized and recorded in chronological order.

[1016] Data Preprocessing

[1017] The server extracts energy usage data for the past 30 days from the database and performs preprocessing such as filling in missing values ​​and removing noise. Specifically, it fills in missing data using linear interpolation and average value interpolation, detects abnormally high consumption data, and corrects it based on past trends.

[1018] Data Analysis and Pattern Recognition

[1019] The server inputs the preprocessed data into a generative AI model to recognize energy consumption patterns and abnormal trends. The generative AI model learns past consumption patterns based on a large amount of data, enabling highly accurate pattern recognition.

[1020] Example prompt: "Generate energy conservation advice based on the past 30 days of energy consumption data, specifically nighttime consumption patterns."

[1021] Advice Generation

[1022] The generator generates specific energy-saving advice based on the analysis results, such as "unplug appliances when not in use at night."

[1023] Emotion Recognition and Advice Adjustment

[1024] The device collects the user's feedback, voice, and facial expressions through a camera and microphone, and the emotion engine analyzes them. If the user is feeling stressed, the emotion engine generates advice with encouraging and kind words.

[1025] Advice Delivery

[1026] The server then sends the adjusted energy saving advice to the device, which then notifies the user within the application, for example, by using a push notification.

[1027] Feedback collection

[1028] The user inputs feedback about the effectiveness of the advice within the application. Specifically, the application records the action of "unplugging home appliances overnight." The device then sends the user-entered feedback to the server.

[1029] Feedback Analysis

[1030] The server analyzes the received feedback data and checks whether energy usage has actually been reduced. The AI ​​learns from this feedback data and reflects it in the generation of next advice.

[1031] This not only improves energy efficiency and reduces utility bills, but also improves the user experience by providing emotionally sensitive advice, making energy management in homes and businesses more effective and satisfying.

[1032] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1033] The flow of this system's program processing

[1034] Step 1: Data collection

[1035] The terminal connects to a HEMS (Home Energy Management System).

[1036] Input: Energy usage data from a HEMS installed in the user's home or business

[1037] Output: Collected energy usage data

[1038] Specific operation: Acquires and collects electricity, gas, and water usage data from the HEMS. This data includes timestamps and consumption amounts.

[1039] Step 2: Send data

[1040] The terminal transmits the collected energy usage data to a server.

[1041] Input: Energy usage data obtained from HEMS

[1042] Output: Energy usage data sent to the server

[1043] What it does: Encrypts data and sends it over the internet to a server via Wi-Fi or wired LAN.

[1044] Step 3: Data reception and storage

[1045] The server receives the energy usage data sent from the terminal and stores it in a database.

[1046] Input: Energy usage data sent from the device

[1047] Output: Energy usage data stored in a database

[1048] Specific operation: The received data is stored in a database in real time and categorized by household or business.

[1049] Step 4: Preprocessing the data

[1050] The server extracts energy usage data for the last 30 days from the database, fills in missing values, and removes noise.

[1051] Input: Energy usage data extracted from a database

[1052] Output: Preprocessed energy usage data

[1053] Specific operation: For the extracted data, missing values ​​are filled in using linear interpolation or mean value interpolation, and abnormal high-consumption data is detected and corrected.

[1054] Step 5: Data analysis and pattern recognition

[1055] The server feeds the pre-processed data into a generative AI model to recognize energy consumption patterns and unusual trends.

[1056] Input: Preprocessed energy usage data

[1057] Output: Recognized energy consumption patterns and unusual trends

[1058] Specific operation: Data is input into the generative AI model to analyze consumption patterns and detect anomalies.

[1059] Example prompt: "Generate energy conservation advice based on the past 30 days of energy consumption data, specifically nighttime consumption patterns."

[1060] Step 6: Advice Generation

[1061] The server generates specific energy-saving advice using the output of the generative AI model.

[1062] Input: Analysis results from a generative AI model

[1063] Output: Specific energy saving advice

[1064] Specific actions: Based on the analysis results, specific advice for energy conservation is generated, such as "unplug appliances at night."

[1065] Step 7: Emotion recognition and advice adjustment

[1066] The device collects the user's feedback, voice, and facial expressions, which are then analyzed by an emotion engine.

[1067] Input: User feedback, voice, facial expressions

[1068] Output: Energy saving advice tailored based on user's emotions

[1069] Specific behavior: The system uses a camera and microphone to recognize the user's emotions and adjusts the content of the energy-saving advice it generates based on the user's emotions. If the user is feeling stressed, it will include encouraging or kind words.

[1070] Step 8: Advice Delivery

[1071] The server sends the adjusted energy saving advice to the terminal, which notifies the user within the application.

[1072] Input: Tailored energy saving advice

[1073] Output: Advice given to the user

[1074] Specific operation: The adjusted advice is sent to the device, and the device notifies the user through the application.

[1075] Step 9: Gather feedback

[1076] The user enters feedback within the application about the effectiveness of the advice they have implemented.

[1077] Input: User feedback on advice taken

[1078] Output: Feedback sent to the device

[1079] Specific actions: Using the application, you can input specific actions and their results. For example, you can input "I unplugged the appliances overnight."

[1080] Step 10: Feedback analysis

[1081] The server analyzes the received feedback data and reflects it in the generation of the next advice.

[1082] Input: Feedback data sent from the device

[1083] Output: More accurate next energy saving advice

[1084] What it does: Analyzes the feedback, checks whether electricity usage has been reduced, and trains the generating AI based on the feedback data, which improves the accuracy of the next advice.

[1085] (Application example 2)

[1086] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1087] Conventional energy management systems were able to collect energy usage data from homes and businesses and provide energy-saving advice, but they did not take user emotions into account when providing advice or collecting and analyzing feedback. This resulted in issues such as a lack of improvement in the user experience and limited effectiveness of advice. Furthermore, in large facilities such as factories, energy consumption patterns are complex, so conventional energy management systems were unable to achieve sufficient energy-saving effects.

[1088] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1089] In this invention, the server includes means for collecting energy usage data, means for transmitting the energy usage data, means for storing the transmitted energy usage data, means for analyzing the stored energy usage data, means for generating energy saving advice based on the analysis results, means for recognizing a user's emotion when generating the energy saving advice and adjusting the content of the advice in accordance with the recognized emotion, and means for collecting feedback on the energy saving advice. This achieves more efficient energy use and reduced utility costs, and also provides advice that takes the user's emotion into consideration, improving the user experience and enabling more satisfying energy management.

[1090] "Energy usage data" refers to data on the amount of energy used, such as electricity, gas, and water, by households, businesses, factories, etc.

[1091] "Means of collection" refers to devices or systems for obtaining energy usage data, specifically sensors and meters.

[1092] The "transmitting means" refers to the communication device or protocol for transferring the collected energy usage data to the server.

[1093] "Means for storage" refers to a database or storage device for retaining and managing the transmitted energy usage data for a long period of time.

[1094] "Means for analysis" refers to software or algorithms used to analyze stored energy usage data and detect consumption patterns and outliers.

[1095] "Means for generating energy-saving advice" refers to a generative AI model or program that generates specific advice recommending energy-efficient ways of using energy based on the analyzed data.

[1096] The "display means" refers to a display, monitor, or application that notifies the user of the generated energy saving advice and visually presents it to the user.

[1097] "Means for recognizing emotions" refers to emotion recognition engines or software that analyze and identify emotions from the user's voice, facial expressions, etc.

[1098] The "means for collecting feedback" refers to an interface or system for collecting information about the results and effects of the energy conservation advice that the user has implemented.

[1099] "In-factory energy usage data" refers to data on the amount of energy consumed by the factory's production lines and each device and equipment.

[1100] A "generative AI model" is an artificial intelligence model that generates new energy-saving advice based on past energy usage data.

[1101] A "prompt sentence" is an input sentence that instructs the generative AI model on what kind of analysis or advice to generate.

[1102] The system that realizes this application example is a combination of specific hardware and software for energy management and emotion recognition within a factory. The main components are sensors and meters for collecting energy usage data, a server for managing and analyzing the data, a robot terminal that provides an interface with the user, and an emotion engine for recognizing emotions. A detailed example of the system is described below.

[1103] Hardware and Software

[1104] 1. Sensors and meters

[1105] Sensors: Installed on each production line in the factory, they collect real-time data on electricity, gas, water, etc. For example, smart meters and environmental monitoring sensors are used.

[1106] Meter: A meter that measures the energy consumption of each device or equipment, such as a smart electricity meter.

[1107] 2. Server

[1108] Database: A database for centrally storing collected energy usage data. For example, MySQL can be used.

[1109] Analysis software: Analyzes the stored data to detect energy consumption patterns and outliers, for example using data analysis libraries such as Pandas and Scikit-learn.

[1110] Generative AI model: A generative AI model for generating energy-saving advice based on data analysis results. For example, the latest generative AI models such as GPT-4 can be used.

[1111] 3. Robot terminal

[1112] Display: A display to inform and visually present energy saving advice to the user.

[1113] Voice input device: A microphone to transmit the user's voice to the emotion recognition engine.

[1114] Emotion recognition engine: Analyzes the user's emotions and adjusts the advice content. For example, OpenFace or Emotion API is used.

[1115] Data processing details

[1116] The server uses sensors and meters to collect energy usage data from each production line in the factory. The collected data is sent to the server and stored in a database. The stored data is then pre-processed using analytical software to detect energy consumption patterns and outliers.

[1117] The analyzed data is input into a generative AI model, which generates specific energy-saving advice. Furthermore, during this generation process, an emotion recognition engine is used to analyze the user's emotions and tailor the advice content based on the recognized emotions. For example, if the user is feeling stressed, encouraging or kind words may be added. This tailored advice is then displayed on the robot terminal's display, informing the user.

[1118] Specific examples

[1119] For example, if analysis reveals that a factory's production line consumes a lot of energy at night, the generative AI model generates advice recommending "replacing nighttime lighting with LEDs." Furthermore, if the emotion recognition engine determines that the operator is stressed, the advice message will include a gentle note: "You don't have to make all the changes at once." This tailored advice is then displayed on the robot terminal's display and provided to the operator.

[1120] Prompt Sentence Examples

[1121] "Please suggest energy-saving measures based on the energy usage data from the last 30 days. Please identify any abnormal consumption patterns and provide specific advice accordingly. If the user is feeling stressed, please offer encouragement and kind words."

[1122] By using this system, factories can achieve more efficient energy use, reducing utility costs and improving the user experience. Furthermore, by analyzing feedback data and reflecting it in the next advice generation, the system can continuously learn and improve the accuracy of its advice.

[1123] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1124] Step 1:

[1125] The terminal collects energy usage data from sensors and meters within the factory. The input is real-time data from the sensors, and the output is the collected energy data. This data includes the electricity, gas, and water usage of each production line and piece of equipment. Specifically, the terminal reads data from the sensors at regular intervals (e.g., every three hours) and temporarily stores the data.

[1126] Step 2:

[1127] The terminal sends the collected energy usage data to the server. The input is the energy data collected in step 1 above, and the output is the data sent to the server. Specifically, the terminal periodically packets the collected data and sends it to the server via the network.

[1128] Step 3:

[1129] The server receives energy data sent from the device and stores it in a database. The input is the sent energy data, and the output is the data organized and stored in the database. Specifically, the server verifies the data it receives, organizes it in chronological order, and stores it.

[1130] Step 4:

[1131] The server preprocesses the stored energy data. The input is the energy data stored in the database, and the output is the preprocessed data. This preprocessing includes filling in missing values ​​and removing noise. Specifically, the server cleans the data using a data processing library such as Pandas.

[1132] Step 5:

[1133] The server inputs the preprocessed data into a generative AI model for analysis. The input is the preprocessed energy data, and the output is the analysis results. Specifically, the server uses a machine learning library such as Scikit-learn to pass the data to the model and identify energy consumption patterns and anomalies.

[1134] Step 6:

[1135] The server creates energy-saving advice generated by the generative AI model. The input is the analysis results, and the output is specific energy-saving advice. Specifically, the server uses the generative AI model to generate specific measures for saving electricity and water. For example, advice such as "replace nighttime lighting with LEDs" is generated.

[1136] Step 7:

[1137] The device analyzes the user's voice and facial expressions and uses an emotion recognition engine to recognize the user's emotions. The input is the user's voice and facial expression data, and the output is the recognized emotion. Specifically, the device analyzes the voice and images using the Emotion API, etc., to identify the emotion.

[1138] Step 8:

[1139] The server adjusts the advice content based on the recognized emotion. The input is energy-saving advice and the recognized emotion data, and the output is the adjusted advice. Specifically, the server adds encouraging or kind words to the advice message. For example, if the user is feeling stressed, the server adds the message, "You don't need to make all the changes at once."

[1140] Step 9:

[1141] The device notifies the user of the adjusted energy-saving advice and displays details. The input is the adjusted advice, and the output is the state notified to the user. Specifically, the device displays the advice on the display and notifies the user in a format that is easy for the user to understand.

[1142] Step 10:

[1143] The user provides feedback on the results of the energy-saving advice they have implemented to the terminal. The input is the user's feedback information, and the output is feedback data. Specifically, the user inputs the results of implementing the advice through the application.

[1144] Step 11:

[1145] The terminal sends feedback from the user to the server. The input is the feedback data, and the output is the data sent to the server. Specifically, the terminal packetizes the feedback information and sends it to the server via the network.

[1146] Step 12:

[1147] The server analyzes the feedback data and evaluates the effectiveness of the advice. The input is the sent feedback data, and the output is the evaluation result. Specifically, the server analyzes the feedback using a data analysis tool and evaluates the effectiveness of the advice. The result is reflected in the next advice generation.

[1148] In this way, a series of steps will result in a system that will improve energy efficiency within the factory and also enhance the user experience.

[1149] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1150] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1151] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[1152] [Third embodiment]

[1153] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1154] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[1155] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1156] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[1157] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1158] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1159] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1160] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1161] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1162] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1163] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1164] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[1165] System Overview

[1166] The present invention relates to a system that collects and analyzes energy usage data and provides energy-saving advice based on the results. This enables homes and businesses to use energy more efficiently and promotes reductions in utility costs. The system mainly includes a collection means, transmission means, storage means, analysis means, generation means, display means, and feedback collection means for handling energy usage data.

[1167] Program processing explanation

[1168] 1. Data Collection

[1169] The terminals are connected to the HEMS (Home Energy Management System) to collect electricity, gas, and water usage data from each home or business.

[1170] The terminal periodically (e.g., every three hours) obtains usage data from the HEMS and sends it to the server.

[1171] 2. Data Receipt and Storage

[1172] The server receives the energy usage data transmitted from the terminal.

[1173] The server stores the received data in a database, organizing it by household and business and recording it in chronological order.

[1174] 3. Data Preprocessing

[1175] The server extracts energy usage data for a certain period of time (e.g., the last 30 days) from the database.

[1176] The server performs pre-processing to complement missing values ​​in the extracted data and remove noise. Specifically, it detects and corrects abnormally high-consumption data.

[1177] 4. Data Analysis and Pattern Recognition

[1178] The server inputs the preprocessed data into the generative AI model.

[1179] Generative AI uses past data to recognize patterns of energy consumption and unusual trends.

[1180] 5. Advice Generation

[1181] The server generates specific energy-saving advice based on the output from the generative AI model.

[1182] Generative AI provides advice tailored to individual consumption patterns.

[1183] 6. Advice Delivery

[1184] The server transmits the generated energy saving advice to the terminal.

[1185] The terminal notifies the user of the received advice and displays it within the application.

[1186] 7. Gathering Feedback

[1187] The user receives feedback within the application about the effectiveness of the advice they have implemented, for example, by inputting a specific action such as "I changed the refrigerator temperature setting."

[1188] The terminal transmits the feedback input by the user to the server.

[1189] 8. Feedback Analysis

[1190] The server analyzes the received feedback data and evaluates the effectiveness of the advice.

[1191] The generative AI learns from the feedback data and reflects it in future advice generation.

[1192] Specific examples

[1193] Examples of reducing electricity usage

[1194] The terminal obtains 24 hours of electricity usage data from the HEMS every three hours and sends it to the server.

[1195] The server receives the sent data and stores it in the database for "Home ID_1234."

[1196] The server extracts the electricity usage data for the last 30 days from the database for "household ID_1234" and detects and corrects any invalid data (e.g., extremely high consumption).

[1197] The generation AI analyzes the data for "household ID_1234" and determines that nighttime consumption is higher than average.

[1198] The AI ​​generates advice recommending that "if you are not using home appliances at night, unplug them" and outputs it to the server.

[1199] The server sends the generated advice to the terminal, which notifies the user within the application and displays the details.

[1200] The user inputs into the application that "I unplugged my home appliances overnight," and the device sends feedback to the server.

[1201] The server analyzes the feedback data and confirms that electricity usage has been reduced.

[1202] The generation AI will incorporate this feedback as learning data and reflect it in generating the next piece of advice.

[1203] Example of reducing gas usage

[1204] The terminal periodically obtains gas usage data from the HEMS and sends it to the server.

[1205] The server receives the data and stores it in a database.

[1206] The server preprocesses the data and adjusts for outliers.

[1207] Generative AI analyzes the data and discovers that certain cooking methods increase gas consumption.

[1208] The generative AI generates advice such as "Use a pressure cooker to reduce cooking time" and outputs it to the server.

[1209] The server sends the generated advice to the terminal, which notifies the user within the application and displays the recommendation.

[1210] The user enters "I used a pressure cooker" in the application, and the device sends the feedback to the server.

[1211] The server analyzes the feedback data and confirms that gas usage has been reduced.

[1212] The generation AI will incorporate this feedback as learning data and reflect it in generating the next piece of advice.

[1213] Through the above processing, the present invention can improve the efficiency of energy use and contribute to reducing utility costs for homes and businesses.

[1214] The processing flow will be explained below.

[1215] Program processing steps

[1216] 1. Data Collection

[1217] Step 1:

[1218] The device connects to the HEMS and collects electricity, gas, and water usage data, including energy usage information for homes and businesses.

[1219] Step 2:

[1220] The device sends the collected data to the server at specified intervals (e.g., every 3 hours).

[1221] 2. Data Receipt and Storage

[1222] Step 3:

[1223] The server receives the energy usage data transmitted from the terminal.

[1224] Step 4:

[1225] The server stores the received data in a database and organizes it by household or business. The data is recorded in chronological order.

[1226] 3. Data Preprocessing

[1227] Step 5:

[1228] The server extracts energy usage data for a certain period (e.g., the last 30 days) from the database.

[1229] Step 6:

[1230] The server completes the extracted data, removes noise, and detects abnormal high-consumption data and corrects it as necessary.

[1231] 4. Data Analysis and Pattern Recognition

[1232] Step 7:

[1233] The server inputs the preprocessed data into the generative AI model.

[1234] Step 8:

[1235] Generative AI recognizes patterns and unusual trends in energy consumption, identifying increases or decreases in consumption for specific times of day or dates.

[1236] 5. Advice Generation

[1237] Step 9:

[1238] The server receives the output of the generative AI model and generates specific energy-saving advice.

[1239] Step 10:

[1240] Generative AI provides customized advice based on individual consumption patterns and usage.

[1241] 6. Advice Delivery

[1242] Step 11:

[1243] The server transmits the generated energy saving advice to the terminal.

[1244] Step 12:

[1245] The terminal notifies the user of the received advice and displays it within the application.

[1246] 7. Gathering Feedback

[1247] Step 13:

[1248] Users can provide feedback within the application about the effectiveness of the advice they have implemented, by entering specific actions and results.

[1249] Step 14:

[1250] The terminal transmits the feedback input by the user to the server.

[1251] 8. Feedback Analysis

[1252] Step 15:

[1253] The server analyzes the received feedback data and evaluates the effectiveness of the advice.

[1254] Step 16:

[1255] The generative AI learns from the feedback data and reflects it in the next advice generation, aiming to provide improved advice.

[1256] Specific examples

[1257] Examples of reducing electricity usage

[1258] Step 1:

[1259] The terminal obtains 24 hours of electricity usage data from the HEMS every three hours.

[1260] Step 2:

[1261] The terminal transmits the collected data to the server.

[1262] Step 3:

[1263] The server receives the transmitted data.

[1264] Step 4:

[1265] The server stores the received data in a database.

[1266] Step 5:

[1267] The server extracts the electricity usage data for the last 30 days from the database for "Home ID_1234."

[1268] Step 6:

[1269] The server detects and corrects abnormally high data consumption.

[1270] Step 7:

[1271] The generative AI analyzes the data and identifies higher-than-average consumption at night.

[1272] Step 8:

[1273] The AI ​​generates advice such as "unplug appliances if you are not using them at night."

[1274] Step 9:

[1275] The server transmits the generated advice to the terminal.

[1276] Step 10:

[1277] The device will notify the user within the application and display details.

[1278] Step 11:

[1279] The user inputs into the application that "I unplugged my home appliances overnight."

[1280] Step 12:

[1281] The terminal sends the feedback to the server.

[1282] Step 13:

[1283] The server analyzes the feedback data and verifies that electricity usage has been reduced.

[1284] Step 14:

[1285] The generative AI incorporates the feedback as learning data and reflects it in generating the next piece of advice.

[1286] Example 1

[1287] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1288] Conventional energy management systems are limited to collecting and simply analyzing energy usage data, making it difficult to efficiently provide specific energy-saving advice to users. They also lack a mechanism for collecting feedback and reflecting the results in the analysis. This creates the challenge of being unable to provide timely and appropriate energy-saving advice tailored to the user's energy consumption patterns.

[1289] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1290] In this invention, the server includes means for periodically transmitting energy usage data, means for receiving the transmitted energy usage data, means for storing the received energy usage data, means for extracting and preprocessing the stored energy usage data, means for inputting the preprocessed data into a generative AI model to recognize energy consumption patterns, means for generating energy saving advice based on the recognition results, means for transmitting the generated energy saving advice, means for displaying the transmitted energy saving advice, means for collecting feedback on the energy saving advice, and means for analyzing the collected feedback and updating the generative AI model. This integrates the collection and analysis of energy usage data, advice generation, feedback collection, and model updating, making it possible to provide users with timely and effective energy saving advice.

[1291] "Energy usage data" refers to data showing the amount of electricity, gas, water, etc. used, and is a measurement of the specific energy usage status consumed within homes and businesses.

[1292] "HEMS" is an abbreviation for Home Energy Management System, a system for optimizing and managing energy usage within the home.

[1293] A "generative AI model" refers to an artificial intelligence model that performs statistical analysis and machine learning based on various data, and uses the knowledge gained from this to generate new information.

[1294] A "prompt sentence" is an input sentence for a specific generative AI model, and is a sentence that instructs the model to produce a specific output.

[1295] "Data preprocessing" refers to the process carried out before data analysis, and refers to the process of preparing data by filling in missing values ​​and removing noise, etc.

[1296] "Feedback" is information that records the user's response to and results of implementing advice provided by the system, and reflects this information back into the system.

[1297] The present invention relates to a system that collects and analyzes energy usage data and provides energy conservation advice based on the results. This invention makes it possible to improve the efficiency of energy usage in homes and businesses and promote reductions in utility costs. Specifically, this system operates according to the following steps.

[1298] System Configuration

[1299] The system mainly includes the following components:

[1300] 1. Collection Method

[1301] 2. Transmission Method

[1302] 3. Preservation means

[1303] 4. Pretreatment Methods

[1304] 5. Analysis method

[1305] 6. Advice Generation Methods

[1306] 7. Display means

[1307] 8. Feedback Collection Methods

[1308] Details of each component are shown below.

[1309] Collection Method

[1310] The device collects energy usage data. Specifically, the device connects to a HEMS (Home Energy Management System) and obtains electricity, gas, and water usage data every three hours. For example, the device obtains the data using the HEMS API ("GET / energy_usage"). The device temporarily stores the data in local storage and prepares to send it to the server later.

[1311] Transmission method

[1312] The device sends the temporarily stored energy usage data to the server using MQTT or REST API. For example, the data to be sent is in JSON format and is assigned an identifier such as "Home ID_1234."

[1313] Preservation means

[1314] The server receives the energy usage data sent from the device and stores it in a NoSQL database (for example, MongoDB). The server stores the received data separately for each household and each business, and manages it in chronological order, which allows for efficient data access.

[1315] Pretreatment means

[1316] The server extracts the most recent 30 days' worth of data from the database, imputes missing values, and performs preprocessing to remove noise. Specifically, the data is cleansed using Python's pandas library, and outliers are corrected using algorithms such as K-Nearest Neighbor (KNN).

[1317] Analysis means

[1318] The server inputs the preprocessed data into a generative AI model to recognize energy consumption patterns and abnormal trends. An example of the generative AI model is OpenAI's GPT. An example prompt is "Please identify abnormal patterns and signs of energy saving from the energy usage data for the last 30 days for household ID_1234."

[1319] Advice Generation Method

[1320] Based on the analysis results, the generation AI generates energy-saving advice tailored to individual consumption patterns. For example, it may generate advice recommending that appliances be unplugged if not in use at night. The generated advice is sent back to the server, which then formats the information in JSON format and stores it back in the database.

[1321] Display means

[1322] The server sends the generated energy-saving advice to the device. The device saves the received advice in local storage and notifies the user within the application. For example, a smartphone push notification can be used to notify the user that "new energy-saving advice is available."

[1323] Feedback collection methods

[1324] The user acts on the advice and provides feedback on the results within the application. Specifically, the user inputs something like, "I changed the refrigerator's set temperature." The device receives this feedback and sends it to the server. The sent feedback is stored in a feedback database on the server.

[1325] Feedback Analysis

[1326] The server analyzes the received feedback data and evaluates the effectiveness of the advice. The generative AI model learns from this feedback data and reflects it in generating new advice. For example, it learns how much power consumption was reduced by changing the refrigerator's set temperature, and reflects this in the next advice.

[1327] Specific examples

[1328] A specific example of operation is shown below.

[1329] 1. The device obtains electricity usage data from the HEMS every three hours and sends the data to the server.

[1330] 2. The server receives the data and stores it in the database as "Home ID_1234".

[1331] 3. The server extracts and preprocesses the data for the last 30 days.

[1332] 4. The generation AI performs an analysis based on the prompt: "Please identify any abnormal patterns or signs of energy saving from the last 30 days of energy usage data for household ID_1234."

[1333] 5. The AI ​​generates advice such as "It is recommended to unplug the device at night" and sends it to the server.

[1334] 6. The server sends the advice to the device, and the device notifies the user via push notification that "New energy saving advice is available."

[1335] 7. The user enters into the application that "I unplugged my appliances overnight," and the device sends feedback to the server.

[1336] 8. The server analyzes the feedback, and the AI ​​uses it to generate the next piece of advice.

[1337] As described above, the present invention can improve the efficiency of energy use and contribute to reducing utility costs for homes and businesses.

[1338] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1339] Step 1: Data collection

[1340] The device connects to the HEMS and obtains electricity, gas, and water usage data every three hours. Specifically, it accesses the HEMS API endpoint "GET / energy_usage" to receive the latest energy usage data. The input is sensor data from the HEMS, and the output is usage data saved in the device's temporary storage. For example, data in the format "Electricity usage: 10kWh" is saved.

[1341] Step 2: Send data

[1342] The energy usage data collected by the device is sent to the server. The data is sent periodically (for example, every three hours) using MQTT or REST API. The input is temporarily stored usage data, and the output is the data to be sent to the server. Specifically, data such as "electricity usage: 10kWh, gas usage: 5m3, water usage: 100L" is structured in JSON format and sent to the server.

[1343] Step 3: Data reception and storage

[1344] The server receives the data sent from the device and stores it in a NoSQL database (for example, MongoDB). The input is the JSON data sent from the device, and the output is the energy usage data stored in the database. Specifically, the received data is organized by household and company, and stored using an identifier such as "Home ID_1234."

[1345] Step 4: Preprocessing the data

[1346] The server extracts the last 30 days' worth of energy usage data from the database. It then performs pre-processing to fill in missing values ​​in the extracted data and remove noise. The input is energy usage data from the database (e.g., the last 30 days' worth of data for "household ID_1234"), and the output is pre-processed, clean data. Specifically, it uses Python's pandas library to fill in missing values ​​and the KNN algorithm to detect and correct outliers.

[1347] Step 5: Data analysis and pattern recognition

[1348] The server inputs the preprocessed data into the generative AI model. The input is the preprocessed data (e.g., energy usage data for the last 30 days for "household ID_1234"), and the output is consumption patterns and abnormal trends recognized by the generative AI model. Specifically, the generative AI is prompted with the following prompt: "Please identify abnormal patterns and signs of energy saving from the energy usage data for the last 30 days for household ID_1234."

[1349] Step 6: Advice Generation

[1350] The generative AI generates energy-saving advice based on the analysis results. The input is the output of the generative AI model (e.g., consumption patterns and abnormal trends), and the output is specific energy-saving advice. For example, the generated advice recommends "unplugging appliances when not in use at night." This advice is sent to the server, which then formats the advice content in JSON format and stores it in a database.

[1351] Step 7: Advice Delivery

[1352] The server sends the generated energy-saving advice to the device. The input is the generated advice (e.g., "Unplug appliances if they are not in use at night"), and the output is the data sent to the device. The device saves the received advice in local storage and notifies the user within the application. Specifically, it uses a push notification on the smartphone to notify the user that "new energy-saving advice is available."

[1353] Step 8: Gather feedback

[1354] The results of the user's execution of advice within the application are fed back. The input is information about the user's actions (e.g., "I changed the refrigerator's set temperature"), and the output is the feedback data. The device receives feedback from the user and sends it to the server. Specifically, it formats the user's input in JSON format and sends a POST request to the server.

[1355] Step 9: Analyze feedback

[1356] The server analyzes the received feedback data and evaluates the effectiveness of the advice. The input is the feedback data (e.g., "As a result of changing the set temperature, power consumption decreased by 5%), and the output is the evaluation result. The generation AI learns from this feedback data and reflects it in generating new advice. For example, it learns that "power consumption at night has decreased" and uses this information to generate the next piece of advice.

[1357] The above are the specific processing steps of this system.

[1358] (Application example 1)

[1359] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1360] In modern brick-and-mortar stores, efficient energy use is important from the perspective of environmental protection and cost reduction. However, in reality, there is a lack of means to grasp energy usage patterns in detail and receive efficient energy-saving advice. As a result, store managers are unable to implement appropriate energy-saving measures, and wasteful energy consumption continues. The objective of this invention is to promote efficient energy use and reduce wasteful energy consumption by analyzing energy usage data in brick-and-mortar stores in detail and providing specific energy-saving advice.

[1361] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1362] In this invention, the server includes means for collecting energy usage data, means for transmitting the energy usage data, means for storing the transmitted energy usage data, means for preprocessing the stored energy usage data, means for analyzing the stored energy usage data, means for generating energy saving advice based on the analysis results, means for generating the generated energy saving advice for specific actions in the store, means for displaying the generated energy saving advice, and means for collecting feedback on the energy saving advice. This makes it possible to understand energy usage patterns in the physical store in detail and to specifically implement efficient energy saving measures.

[1363] "Energy usage data" refers to data that records the amount of energy consumed in a physical store, such as electricity, gas, and water.

[1364] The "transmission means" is a function for transmitting collected energy usage data to a server.

[1365] The "storage means" is a function for storing the transmitted energy usage data in a storage device such as a database.

[1366] The "preprocessing means" is a function for performing preprocessing such as complementing missing values ​​and removing outliers on stored energy usage data.

[1367] The "analysis means" is a function for analyzing the pre-processed energy usage data and identifying patterns and anomalies in energy consumption.

[1368] The "generation means" is a function for generating energy saving advice based on the analysis results.

[1369] The "display means" is a function for presenting the generated energy saving advice to the operator of the physical store.

[1370] The "feedback collection means" is a function for collecting feedback regarding the energy saving advice that the operator has implemented.

[1371] "Generative AI" is an artificial intelligence model that analyzes energy usage data and generates specific energy-saving advice.

[1372] A "prompt sentence" is an instruction sentence that provides specific energy-saving advice to the generation AI.

[1373] "In-store energy saving advice" is advice that recommends specific energy saving actions for specific equipment or activities in a physical store.

[1374] The present invention relates to a system that collects and analyzes energy usage data and provides energy conservation advice based on the results, thereby making it possible to improve the efficiency of energy usage in physical stores and promote reductions in utility costs.

[1375] The system includes a collection means, a transmission means, a storage means, a pre-processing means, an analysis means, a generation means, a display means, and a feedback collection means for handling energy usage data.

[1376] Specific hardware and software configuration:

[1377] The terminal, a smartphone or tablet, connects to the HEMS (Home Energy Management System) in the physical store and collects energy usage data.

[1378] The server receives the collected energy usage data and stores it in a database, where it is organized by household and business and recorded in chronological order.

[1379] The preprocessing means extracts energy usage data for a certain period from the database and performs preprocessing such as filling in missing values ​​and removing outliers, thereby ensuring the reliability of the data.

[1380] The analytics tool feeds the pre-processed data into a generative AI model to identify patterns and unusual trends in energy consumption.

[1381] The AI ​​that generates the energy-saving advice is based on the analysis results and is specific to the user's consumption patterns.

[1382] The terminal as a display means notifies the user of the generated energy saving advice and displays it within the application.

[1383] The feedback collection means collects feedback regarding the effect of the energy saving advice implemented by the user and transmits the feedback to the server.

[1384] Software used and detailed data processing:

[1385] Data collection: The device acquires energy usage data from the HEMS. The data is sent to the server periodically (e.g., every three hours).

[1386] Data storage and preprocessing: The server stores the data in a database and preprocesses it, including imputing missing values ​​and removing outliers.

[1387] Data analysis and pattern recognition: Analyze the pre-processed data and recognize consumption patterns and unusual trends from historical data. Generative AI models take on this role and automate the process.

[1388] Generating energy-saving advice: The AI ​​generates energy-saving advice based on the analysis results. For example, it may generate advice such as, "Since power consumption is high at night, we recommend turning off devices outside of business hours."

[1389] User notification and feedback collection: The generated advice is notified to the user and feedback is collected within the application. The feedback is sent back to the server and reflected in subsequent advice generation.

[1390] Examples and prompts:

[1391] Example: Identifying high consumption at night and making specific recommendations on which appliances in a store should be turned off outside of business hours.

[1392] Example prompt: "Our store consumes a lot of electricity at night. Please have your Generative AI suggest which devices in our store should be turned off after hours."

[1393] In this way, the present invention efficiently manages energy usage in physical stores and provides specific energy-saving advice, enabling operators of physical stores to implement appropriate energy-saving measures.

[1394] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1395] Step 1:

[1396] Data collection

[1397] The terminal obtains electricity, gas, and water usage data every three hours from the physical store's HEMS (Home Energy Management System). The input is real-time energy usage data from the HEMS, and the output is collected energy usage data, which is used for further analysis.

[1398] Step 2:

[1399] Data transmission

[1400] The terminal periodically transmits the collected energy usage data to the server. The input is the collected energy usage data, and the output is the energy usage data transmitted to the server. The transmitted data is stored on the server.

[1401] Step 3:

[1402] Data storage

[1403] The server receives the transmitted energy usage data and stores it in a database. The input is the energy usage data transmitted to the server, and the output is the energy usage data stored in the database. This data is used for later analysis.

[1404] Step 4:

[1405] Data Preprocessing

[1406] The server extracts energy usage data for a certain period of time (e.g., the last 30 days) from the database, and completes missing values ​​and removes outliers. The input is the energy usage data stored in the database, and the output is the corrected and preprocessed energy usage data. Specifically, it completes missing values ​​and removes outliers using the 3 sigma rule.

[1407] Step 5:

[1408] Data Analysis and Pattern Recognition

[1409] The server inputs the preprocessed energy usage data into a generative AI model to identify energy consumption patterns and anomalies. The input is the preprocessed energy usage data, and the output is the recognition results of consumption patterns and anomalies. Specifically, it finds consumption patterns from past data and identifies abnormal trends.

[1410] Step 6:

[1411] Energy saving advice generation

[1412] The server uses a generative AI model based on the analysis results to generate energy-saving advice. The input is the consumption pattern and anomaly recognition results, and the output is the generated energy-saving advice. For example, it generates specific advice such as "Since consumption is high at night, it is recommended that you turn off devices outside of business hours."

[1413] Step 7:

[1414] Advice display

[1415] The device notifies the user of the generated energy-saving advice and displays it within the application. The input is the generated energy-saving advice, and the output is the advice displayed to the user. Specifically, a notification is sent to the user's smartphone or tablet.

[1416] Step 8:

[1417] Feedback collection

[1418] The user inputs feedback about the effectiveness of the energy-saving advice they have implemented through the application. The input is the user's feedback, and the output is feedback data. Specifically, the user inputs an action such as "I unplugged my home appliances at night" into the application.

[1419] Step 9:

[1420] Feedback Analysis

[1421] The server analyzes the received feedback data, evaluates the effectiveness of the advice, and provides feedback to the generation AI model. The input is the feedback data sent by the user, and the output is the effectiveness evaluation result. Specifically, the effectiveness of the energy-saving advice is evaluated and reflected in the generation of the next advice.

[1422] By following these steps, energy usage in physical stores can be managed efficiently and specific energy-saving advice can be provided.

[1423] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1424] System Overview

[1425] The present invention combines a system that collects and analyzes energy usage data and provides energy-saving advice based on the results with an emotion engine that recognizes user emotions. This makes it possible to improve the efficiency of energy use in homes and businesses, promote reductions in utility bills, and provide advice that takes user emotions into consideration. The system mainly includes a collection means for handling energy usage data, a transmission means, a storage means, an analysis means, a generation means, a display means, an emotion engine, and a feedback collection means.

[1426] Program processing explanation

[1427] 1. Data Collection

[1428] The terminals are connected to the HEMS (Home Energy Management System) to collect electricity, gas, and water usage data from each home or business.

[1429] The terminal periodically (e.g., every three hours) obtains usage data from the HEMS and sends it to the server.

[1430] 2. Data Receipt and Storage

[1431] The server receives the energy usage data transmitted from the terminal.

[1432] The server stores the received data in a database, organizing it by household and business and recording it in chronological order.

[1433] 3. Data Preprocessing

[1434] The server extracts energy usage data for a certain period (e.g., the last 30 days) from the database.

[1435] The server performs pre-processing to complement missing values ​​in the extracted data and remove noise. Specifically, it detects and corrects abnormally high-consumption data.

[1436] 4. Data Analysis and Pattern Recognition

[1437] The server inputs the preprocessed data into the generative AI model.

[1438] Generative AI uses past data to recognize patterns of energy consumption and unusual trends.

[1439] 5. Advice Generation

[1440] The server generates specific energy-saving advice based on the output from the generative AI model.

[1441] Generative AI provides advice tailored to individual consumption patterns.

[1442] 6. Emotion recognition and advice adjustment

[1443] The device analyzes the user's feedback, voice, and facial expressions to recognize the user's emotions.

[1444] The emotion engine tailors the generated energy-saving advice based on the recognized emotion: for example, if the user is feeling stressed, it generates a message containing encouraging and kind words.

[1445] 7. Advice Delivery

[1446] The server transmits the adjusted energy saving advice to the terminal.

[1447] The terminal notifies the user of the received advice and displays it within the application.

[1448] 8. Collecting Feedback

[1449] Users can provide feedback within the application about the effectiveness of the advice they have implemented, by entering specific actions and results.

[1450] The terminal transmits the feedback input by the user to the server.

[1451] 9. Feedback Analysis

[1452] The server analyzes the received feedback data and evaluates the effectiveness of the advice.

[1453] The generation AI learns from the feedback data and reflects it in generating the next piece of advice.

[1454] Specific examples

[1455] Examples of reducing electricity usage

[1456] System Operation

[1457] The device retrieves 24 hours of electricity usage data from the HEMS every three hours.

[1458] The terminal transmits the collected data to the server.

[1459] The server receives the data and stores it in a database.

[1460] The server extracts the electricity usage data for the last 30 days from the database for "Home ID_1234."

[1461] The server detects and corrects abnormally high data consumption.

[1462] Generative AI analyzes the data and identifies higher-than-average consumption at night.

[1463] The AI ​​generates advice such as "unplug appliances if you are not using them at night."

[1464] The device analyzes the user's feedback, voice, and facial expressions, and uses an emotion engine to check whether the user is feeling stressed.

[1465] The emotion engine tailors advice as needed, adding encouragement and kind words.

[1466] The server sends the tailored advice to the terminal, which notifies the user within the application and displays the details.

[1467] The user inputs into the application that "I unplugged my home appliances overnight."

[1468] The device sends the feedback to the server.

[1469] The server analyzes the feedback data and confirms that electricity usage has been reduced.

[1470] The generation AI will incorporate this feedback as learning data and reflect it in generating the next piece of advice.

[1471] The system not only improves energy efficiency and reduces utility bills, but also enhances the user experience by providing emotionally sensitive advice, making energy management in homes and businesses more effective and satisfying.

[1472] The processing flow will be explained below.

[1473] Program processing steps

[1474] 1. Data Collection

[1475] Step 1:

[1476] The device connects to the HEMS and collects electricity, gas, and water usage data, which is obtained from sensors and meters installed in each home or business.

[1477] Step 2:

[1478] The device sends the collected usage data to the server at specified intervals (e.g., every 3 hours).

[1479] 2. Data Receipt and Storage

[1480] Step 3:

[1481] The server receives the energy usage data transmitted from the terminal.

[1482] Step 4:

[1483] The server stores the received data in a database, organizing it by household and business and recording it in chronological order.

[1484] 3. Data Preprocessing

[1485] Step 5:

[1486] The server detects missing values ​​in the energy usage data and completes them as necessary.

[1487] Step 6:

[1488] The server scrutinizes the data to remove abnormal values ​​and noise, and detects and corrects abnormally high consumption data.

[1489] 4. Data Analysis and Pattern Recognition

[1490] Step 7:

[1491] The server inputs the preprocessed data into the generative AI model.

[1492] Step 8:

[1493] Generative AI recognizes patterns and unusual trends in energy consumption, for example, identifying sudden increases or decreases in consumption on certain days or during certain times of the day.

[1494] 5. Advice Generation

[1495] Step 9:

[1496] The server generates specific advice for energy conservation based on the analysis results of the generation AI.

[1497] Step 10:

[1498] The generative AI provides customized advice based on consumption patterns, such as "unplug appliances if you're not using them overnight" or "use a pressure cooker to reduce cooking time."

[1499] 6. Emotion recognition and advice adjustment

[1500] Step 11:

[1501] The device analyzes the user's feedback, voice, and facial expressions to recognize the user's emotions. Technologies used include voice recognition and facial recognition.

[1502] Step 12:

[1503] The emotion engine adjusts the energy-saving advice based on the recognized emotion. For example, if the user is feeling stressed, the engine generates advice that includes encouraging and kind words.

[1504] 7. Advice Delivery

[1505] Step 13:

[1506] The server transmits the adjusted energy saving advice to the terminal.

[1507] Step 14:

[1508] The device will notify the user and provide detailed advice within the application, which the user can review at any time.

[1509] 8. Collecting Feedback

[1510] Step 15:

[1511] The user provides feedback on the effectiveness of the advice they received within the application and whether they implemented it. For example, they can input information such as "I changed the refrigerator temperature setting" or "I used a pressure cooker."

[1512] Step 16:

[1513] The terminal transmits the feedback input by the user to the server.

[1514] 9. Feedback Analysis

[1515] Step 17:

[1516] The server analyzes the received feedback data and evaluates the effectiveness of the advice, for example, checking whether electricity usage has been reduced.

[1517] Step 18:

[1518] The AI ​​learns from the feedback data and reflects it in the next advice generation, allowing it to provide more effective and tailored advice to the user.

[1519] Specific examples

[1520] Reducing electricity consumption

[1521] Step 1:

[1522] The device retrieves 24 hours of electricity usage data from the HEMS every three hours.

[1523] Step 2:

[1524] The terminal transmits the collected data to the server.

[1525] Step 3:

[1526] The server receives the transmitted data.

[1527] Step 4:

[1528] The server stores the received data in a database.

[1529] Step 5:

[1530] The server extracts the electricity usage data for the last 30 days from the database for "Home ID_1234."

[1531] Step 6:

[1532] The server detects and corrects abnormally high data consumption.

[1533] Step 7:

[1534] Generative AI analyzes the data and identifies higher-than-average consumption at night.

[1535] Step 8:

[1536] The AI ​​generates advice such as "unplug appliances if you are not using them at night."

[1537] Step 9:

[1538] The server transmits the generated advice to the terminal.

[1539] Step 10:

[1540] The device will notify the user within the application and display details.

[1541] Step 11:

[1542] The device analyzes the user's voice and facial expressions and uses an emotion engine to check whether the user is feeling stressed.

[1543] Step 12:

[1544] The emotion engine tailors advice as needed, adding encouragement and kind words.

[1545] Step 13:

[1546] The server sends the adjusted advice to the terminal.

[1547] Step 14:

[1548] The device will notify the user within the application and display details.

[1549] Step 15:

[1550] The user inputs into the application that "I unplugged my home appliances overnight."

[1551] Step 16:

[1552] The device sends the feedback to the server.

[1553] Step 17:

[1554] The server analyzes the feedback data and confirms that electricity usage has been reduced.

[1555] Step 18:

[1556] The generation AI will incorporate this feedback as learning data and reflect it in generating the next piece of advice.

[1557] Example 2

[1558] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1559] Existing systems that collect and analyze energy usage data and provide energy-saving advice rarely provide advice that takes users' emotions into consideration, resulting in insufficient user satisfaction and energy reduction effects. Another problem is that there is a lack of a mechanism for continuously improving the performance of energy-saving advice using collected feedback, which limits the effectiveness of the advice.

[1560] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1561] In this invention, the server includes means for collecting energy usage data, means for transmitting the energy usage data, means for storing the transmitted energy usage data, means for preprocessing the stored energy usage data, means for analyzing the preprocessed data and recognizing energy consumption patterns, means for generating energy saving advice based on the analysis results, means for adjusting the generated energy saving advice based on a user's emotions, means for displaying the adjusted energy saving advice, means for collecting feedback on the energy saving advice, and means for analyzing the collected feedback and improving the performance of the energy saving advice generator. This makes it possible to provide energy saving advice that takes the user's emotions into consideration, thereby improving user satisfaction and maximizing energy reduction effects.

[1562] "Energy usage data" refers to data that indicates the amount of energy consumed, such as electricity, gas, and water, in homes and businesses.

[1563] "Collection means" refers to the means for capturing energy usage data, and specifically includes connection to devices such as HEMS (Home Energy Management System).

[1564] The "transmission means" is a means for sending the collected energy usage data to the server via a network.

[1565] The "storage means" is a means for storing and managing the energy usage data sent to the server by the transmission means in a database.

[1566] The "preprocessing means" is a means for complementing missing values ​​and removing noise from the stored energy usage data.

[1567] "Analysis means" refers to means for using the pre-processed data to recognize energy consumption patterns and detect abnormal trends.

[1568] "Generation means" refers to the means for creating specific energy-saving advice based on the analysis results, and includes the use of a generative AI model.

[1569] The "adjustment means" is a means for modifying the content of the generated energy saving advice based on the user's feelings.

[1570] The "display means" is a means for providing the user with the adjusted energy saving advice, and includes displays and notifications within the application.

[1571] The "feedback collection means" is a means for transmitting to the server the effects of advice taken by the user and their impressions.

[1572] The "feedback analysis means" is a means used to analyze collected feedback data and improve the performance of subsequent advice generation.

[1573] The present invention is a system that collects and analyzes energy usage data and provides energy-saving advice based on the results, and combines it with an emotion engine that recognizes the user's emotions. This system makes it possible to improve the efficiency of energy usage in homes and businesses, promote reductions in utility costs, and provide advice that takes the user's emotions into consideration. An embodiment of the present invention is described in detail below.

[1574] System Overview

[1575] The system mainly includes a collection means, a transmission means, a storage means, a pre-processing means, an analysis means, a generation means, an adjustment means, a display means, a feedback collection means, and a feedback analysis means for handling energy usage data. These means are executed by terminals installed in homes and businesses and a server that collects and analyzes data.

[1576] Data collection

[1577] The device connects to a HEMS (Home Energy Management System) installed in a home or business to collect electricity, gas, and water usage data. This connection is via Wi-Fi or wired LAN. The device encrypts the collected energy usage data and sends it to a server over the network.

[1578] Data Receipt and Storage

[1579] The server receives the energy usage data sent from the device and stores it in a database in real time, so that the data is properly categorized and recorded in chronological order.

[1580] Data Preprocessing

[1581] The server extracts energy usage data for the past 30 days from the database and performs preprocessing such as filling in missing values ​​and removing noise. Specifically, it fills in missing data using linear interpolation and average value interpolation, detects abnormally high consumption data, and corrects it based on past trends.

[1582] Data Analysis and Pattern Recognition

[1583] The server inputs the preprocessed data into a generative AI model to recognize energy consumption patterns and abnormal trends. The generative AI model learns past consumption patterns based on a large amount of data, enabling highly accurate pattern recognition.

[1584] Example prompt: "Generate energy conservation advice based on the past 30 days of energy consumption data, specifically nighttime consumption patterns."

[1585] Advice Generation

[1586] The generator generates specific energy-saving advice based on the analysis results, such as "unplug appliances when not in use at night."

[1587] Emotion Recognition and Advice Adjustment

[1588] The device collects the user's feedback, voice, and facial expressions through a camera and microphone, and the emotion engine analyzes them. If the user is feeling stressed, the emotion engine generates advice with encouraging and kind words.

[1589] Advice Delivery

[1590] The server then sends the adjusted energy saving advice to the device, which then notifies the user within the application, for example, by using a push notification.

[1591] Feedback collection

[1592] The user inputs feedback about the effectiveness of the advice within the application. Specifically, the application records the action of "unplugging home appliances overnight." The device then sends the user-entered feedback to the server.

[1593] Feedback Analysis

[1594] The server analyzes the received feedback data and checks whether energy usage has actually been reduced. The AI ​​learns from this feedback data and reflects it in the generation of next advice.

[1595] This not only improves energy efficiency and reduces utility bills, but also improves the user experience by providing emotionally sensitive advice, making energy management in homes and businesses more effective and satisfying.

[1596] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1597] The flow of this system's program processing

[1598] Step 1: Data collection

[1599] The terminal connects to a HEMS (Home Energy Management System).

[1600] Input: Energy usage data from a HEMS installed in the user's home or business

[1601] Output: Collected energy usage data

[1602] Specific operation: Acquires and collects electricity, gas, and water usage data from the HEMS. This data includes timestamps and consumption amounts.

[1603] Step 2: Send data

[1604] The terminal transmits the collected energy usage data to a server.

[1605] Input: Energy usage data obtained from HEMS

[1606] Output: Energy usage data sent to the server

[1607] What it does: Encrypts data and sends it over the internet to a server via Wi-Fi or wired LAN.

[1608] Step 3: Data reception and storage

[1609] The server receives the energy usage data sent from the terminal and stores it in a database.

[1610] Input: Energy usage data sent from the device

[1611] Output: Energy usage data stored in a database

[1612] Specific operation: The received data is stored in a database in real time and categorized by household or business.

[1613] Step 4: Preprocessing the data

[1614] The server extracts energy usage data for the last 30 days from the database, fills in missing values, and removes noise.

[1615] Input: Energy usage data extracted from a database

[1616] Output: Preprocessed energy usage data

[1617] Specific operation: For the extracted data, missing values ​​are filled in using linear interpolation or mean value interpolation, and abnormal high-consumption data is detected and corrected.

[1618] Step 5: Data analysis and pattern recognition

[1619] The server feeds the pre-processed data into a generative AI model to recognize energy consumption patterns and unusual trends.

[1620] Input: Preprocessed energy usage data

[1621] Output: Recognized energy consumption patterns and unusual trends

[1622] Specific operation: Data is input into the generative AI model to analyze consumption patterns and detect anomalies.

[1623] Example prompt: "Generate energy conservation advice based on the past 30 days of energy consumption data, specifically nighttime consumption patterns."

[1624] Step 6: Advice Generation

[1625] The server generates specific energy-saving advice using the output of the generative AI model.

[1626] Input: Analysis results from a generative AI model

[1627] Output: Specific energy saving advice

[1628] Specific actions: Based on the analysis results, specific advice for energy conservation is generated, such as "unplug appliances at night."

[1629] Step 7: Emotion recognition and advice adjustment

[1630] The device collects the user's feedback, voice, and facial expressions, which are then analyzed by an emotion engine.

[1631] Input: User feedback, voice, facial expressions

[1632] Output: Energy saving advice tailored based on user's emotions

[1633] Specific behavior: The system uses a camera and microphone to recognize the user's emotions and adjusts the content of the energy-saving advice it generates based on the user's emotions. If the user is feeling stressed, it will include encouraging or kind words.

[1634] Step 8: Advice Delivery

[1635] The server sends the adjusted energy saving advice to the terminal, which notifies the user within the application.

[1636] Input: Tailored energy saving advice

[1637] Output: Advice given to the user

[1638] Specific operation: The adjusted advice is sent to the device, and the device notifies the user through the application.

[1639] Step 9: Gather feedback

[1640] The user enters feedback within the application about the effectiveness of the advice they have implemented.

[1641] Input: User feedback on advice taken

[1642] Output: Feedback sent to the device

[1643] Specific actions: Using the application, you can input specific actions and their results. For example, you can input "I unplugged the appliances overnight."

[1644] Step 10: Feedback analysis

[1645] The server analyzes the received feedback data and reflects it in the generation of the next advice.

[1646] Input: Feedback data sent from the device

[1647] Output: More accurate next energy saving advice

[1648] What it does: Analyzes the feedback, checks whether electricity usage has been reduced, and trains the generating AI based on the feedback data, which improves the accuracy of the next advice.

[1649] (Application example 2)

[1650] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1651] Conventional energy management systems were able to collect energy usage data from homes and businesses and provide energy-saving advice, but they did not take user emotions into account when providing advice or collecting and analyzing feedback. This resulted in issues such as a lack of improvement in the user experience and limited effectiveness of advice. Furthermore, in large facilities such as factories, energy consumption patterns are complex, so conventional energy management systems were unable to achieve sufficient energy-saving effects.

[1652] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1653] In this invention, the server includes means for collecting energy usage data, means for transmitting the energy usage data, means for storing the transmitted energy usage data, means for analyzing the stored energy usage data, means for generating energy saving advice based on the analysis results, means for recognizing a user's emotion when generating the energy saving advice and adjusting the content of the advice in accordance with the recognized emotion, and means for collecting feedback on the energy saving advice. This achieves more efficient energy use and reduced utility costs, and also provides advice that takes the user's emotion into consideration, improving the user experience and enabling more satisfying energy management.

[1654] "Energy usage data" refers to data on the amount of energy used, such as electricity, gas, and water, by households, businesses, factories, etc.

[1655] "Means of collection" refers to devices or systems for obtaining energy usage data, specifically sensors and meters.

[1656] The "transmitting means" refers to the communication device or protocol for transferring the collected energy usage data to the server.

[1657] "Means for storage" refers to a database or storage device for retaining and managing the transmitted energy usage data for a long period of time.

[1658] "Means for analysis" refers to software or algorithms used to analyze stored energy usage data and detect consumption patterns and outliers.

[1659] "Means for generating energy-saving advice" refers to a generative AI model or program that generates specific advice recommending energy-efficient ways of using energy based on the analyzed data.

[1660] The "display means" refers to a display, monitor, or application that notifies the user of the generated energy saving advice and visually presents it to the user.

[1661] "Means for recognizing emotions" refers to emotion recognition engines or software that analyze and identify emotions from the user's voice, facial expressions, etc.

[1662] The "means for collecting feedback" refers to an interface or system for collecting information about the results and effects of the energy conservation advice that the user has implemented.

[1663] "In-factory energy usage data" refers to data on the amount of energy consumed by the factory's production lines and each device and equipment.

[1664] A "generative AI model" is an artificial intelligence model that generates new energy-saving advice based on past energy usage data.

[1665] A "prompt sentence" is an input sentence that instructs the generative AI model on what kind of analysis or advice to generate.

[1666] The system that realizes this application example is a combination of specific hardware and software for energy management and emotion recognition within a factory. The main components are sensors and meters for collecting energy usage data, a server for managing and analyzing the data, a robot terminal that provides an interface with the user, and an emotion engine for recognizing emotions. A detailed example of the system is described below.

[1667] Hardware and Software

[1668] 1. Sensors and meters

[1669] Sensors: Installed on each production line in the factory, they collect real-time data on electricity, gas, water, etc. For example, smart meters and environmental monitoring sensors are used.

[1670] Meter: A meter that measures the energy consumption of each device or equipment, such as a smart electricity meter.

[1671] 2. Server

[1672] Database: A database for centrally storing collected energy usage data. For example, MySQL can be used.

[1673] Analysis software: Analyzes the stored data to detect energy consumption patterns and outliers, for example using data analysis libraries such as Pandas and Scikit-learn.

[1674] Generative AI model: A generative AI model for generating energy-saving advice based on data analysis results. For example, the latest generative AI models such as GPT-4 can be used.

[1675] 3. Robot terminal

[1676] Display: A display to inform and visually present energy saving advice to the user.

[1677] Voice input device: A microphone to transmit the user's voice to the emotion recognition engine.

[1678] Emotion recognition engine: Analyzes the user's emotions and adjusts the advice content. For example, OpenFace or Emotion API is used.

[1679] Data processing details

[1680] The server uses sensors and meters to collect energy usage data from each production line in the factory. The collected data is sent to the server and stored in a database. The stored data is then pre-processed using analytical software to detect energy consumption patterns and outliers.

[1681] The analyzed data is input into a generative AI model, which generates specific energy-saving advice. Furthermore, during this generation process, an emotion recognition engine is used to analyze the user's emotions and tailor the advice content based on the recognized emotions. For example, if the user is feeling stressed, encouraging or kind words may be added. This tailored advice is then displayed on the robot terminal's display, informing the user.

[1682] Specific examples

[1683] For example, if analysis reveals that a factory's production line consumes a lot of energy at night, the generative AI model generates advice recommending "replacing nighttime lighting with LEDs." Furthermore, if the emotion recognition engine determines that the operator is stressed, the advice message will include a gentle note: "You don't have to make all the changes at once." This tailored advice is then displayed on the robot terminal's display and provided to the operator.

[1684] Prompt Sentence Examples

[1685] "Please suggest energy-saving measures based on the energy usage data from the last 30 days. Please identify any abnormal consumption patterns and provide specific advice accordingly. If the user is feeling stressed, please offer encouragement and kind words."

[1686] By using this system, factories can achieve more efficient energy use, reducing utility costs and improving the user experience. Furthermore, by analyzing feedback data and reflecting it in the next advice generation, the system can continuously learn and improve the accuracy of its advice.

[1687] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1688] Step 1:

[1689] The terminal collects energy usage data from sensors and meters within the factory. The input is real-time data from the sensors, and the output is the collected energy data. This data includes the electricity, gas, and water usage of each production line and piece of equipment. Specifically, the terminal reads data from the sensors at regular intervals (e.g., every three hours) and temporarily stores the data.

[1690] Step 2:

[1691] The terminal sends the collected energy usage data to the server. The input is the energy data collected in step 1 above, and the output is the data sent to the server. Specifically, the terminal periodically packets the collected data and sends it to the server via the network.

[1692] Step 3:

[1693] The server receives energy data sent from the device and stores it in a database. The input is the sent energy data, and the output is the data organized and stored in the database. Specifically, the server verifies the data it receives, organizes it in chronological order, and stores it.

[1694] Step 4:

[1695] The server preprocesses the stored energy data. The input is the energy data stored in the database, and the output is the preprocessed data. This preprocessing includes filling in missing values ​​and removing noise. Specifically, the server cleans the data using a data processing library such as Pandas.

[1696] Step 5:

[1697] The server inputs the preprocessed data into a generative AI model for analysis. The input is the preprocessed energy data, and the output is the analysis results. Specifically, the server uses a machine learning library such as Scikit-learn to pass the data to the model and identify energy consumption patterns and anomalies.

[1698] Step 6:

[1699] The server creates energy-saving advice generated by the generative AI model. The input is the analysis results, and the output is specific energy-saving advice. Specifically, the server uses the generative AI model to generate specific measures for saving electricity and water. For example, advice such as "replace nighttime lighting with LEDs" is generated.

[1700] Step 7:

[1701] The device analyzes the user's voice and facial expressions and uses an emotion recognition engine to recognize the user's emotions. The input is the user's voice and facial expression data, and the output is the recognized emotion. Specifically, the device analyzes the voice and images using the Emotion API, etc., to identify the emotion.

[1702] Step 8:

[1703] The server adjusts the advice content based on the recognized emotion. The input is energy-saving advice and the recognized emotion data, and the output is the adjusted advice. Specifically, the server adds encouraging or kind words to the advice message. For example, if the user is feeling stressed, the server adds the message, "You don't need to make all the changes at once."

[1704] Step 9:

[1705] The device notifies the user of the adjusted energy-saving advice and displays details. The input is the adjusted advice, and the output is the state notified to the user. Specifically, the device displays the advice on the display and notifies the user in a format that is easy for the user to understand.

[1706] Step 10:

[1707] The user provides feedback on the results of the energy-saving advice they have implemented to the terminal. The input is the user's feedback information, and the output is feedback data. Specifically, the user inputs the results of implementing the advice through the application.

[1708] Step 11:

[1709] The terminal sends feedback from the user to the server. The input is the feedback data, and the output is the data sent to the server. Specifically, the terminal packetizes the feedback information and sends it to the server via the network.

[1710] Step 12:

[1711] The server analyzes the feedback data and evaluates the effectiveness of the advice. The input is the sent feedback data, and the output is the evaluation result. Specifically, the server analyzes the feedback using a data analysis tool and evaluates the effectiveness of the advice. The result is reflected in the next advice generation.

[1712] In this way, a series of steps will result in a system that will improve energy efficiency within the factory and also enhance the user experience.

[1713] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1714] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1715] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1716] [Fourth embodiment]

[1717] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1718] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1719] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1720] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1721] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1722] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1723] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1724] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1725] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1726] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1727] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1728] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1729] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1730] System Overview

[1731] The present invention relates to a system that collects and analyzes energy usage data and provides energy-saving advice based on the results. This enables homes and businesses to use energy more efficiently and promotes reductions in utility costs. The system mainly includes a collection means, transmission means, storage means, analysis means, generation means, display means, and feedback collection means for handling energy usage data.

[1732] Program processing explanation

[1733] 1. Data Collection

[1734] The terminals are connected to the HEMS (Home Energy Management System) to collect electricity, gas, and water usage data from each home or business.

[1735] The terminal periodically (e.g., every three hours) obtains usage data from the HEMS and sends it to the server.

[1736] 2. Data Receipt and Storage

[1737] The server receives the energy usage data transmitted from the terminal.

[1738] The server stores the received data in a database, organizing it by household and business and recording it in chronological order.

[1739] 3. Data Preprocessing

[1740] The server extracts energy usage data for a certain period of time (e.g., the last 30 days) from the database.

[1741] The server performs pre-processing to complement missing values ​​in the extracted data and remove noise. Specifically, it detects and corrects abnormally high-consumption data.

[1742] 4. Data Analysis and Pattern Recognition

[1743] The server inputs the preprocessed data into the generative AI model.

[1744] Generative AI uses past data to recognize patterns of energy consumption and unusual trends.

[1745] 5. Advice Generation

[1746] The server generates specific energy-saving advice based on the output from the generative AI model.

[1747] Generative AI provides advice tailored to individual consumption patterns.

[1748] 6. Advice Delivery

[1749] The server transmits the generated energy saving advice to the terminal.

[1750] The terminal notifies the user of the received advice and displays it within the application.

[1751] 7. Gathering Feedback

[1752] The user receives feedback within the application about the effectiveness of the advice they have implemented, for example, by inputting a specific action such as "I changed the refrigerator temperature setting."

[1753] The terminal transmits the feedback input by the user to the server.

[1754] 8. Feedback Analysis

[1755] The server analyzes the received feedback data and evaluates the effectiveness of the advice.

[1756] The generative AI learns from the feedback data and reflects it in future advice generation.

[1757] Specific examples

[1758] Examples of reducing electricity usage

[1759] The terminal obtains 24 hours of electricity usage data from the HEMS every three hours and sends it to the server.

[1760] The server receives the sent data and stores it in the database for "Home ID_1234."

[1761] The server extracts the electricity usage data for the last 30 days from the database for "household ID_1234" and detects and corrects any invalid data (e.g., extremely high consumption).

[1762] The generation AI analyzes the data for "household ID_1234" and determines that nighttime consumption is higher than average.

[1763] The AI ​​generates advice recommending that "if you are not using home appliances at night, unplug them" and outputs it to the server.

[1764] The server sends the generated advice to the terminal, which notifies the user within the application and displays the details.

[1765] The user inputs into the application that "I unplugged my home appliances overnight," and the device sends feedback to the server.

[1766] The server analyzes the feedback data and confirms that electricity usage has been reduced.

[1767] The generation AI will incorporate this feedback as learning data and reflect it in generating the next piece of advice.

[1768] Example of reducing gas usage

[1769] The terminal periodically obtains gas usage data from the HEMS and sends it to the server.

[1770] The server receives the data and stores it in a database.

[1771] The server preprocesses the data and adjusts for outliers.

[1772] Generative AI analyzes the data and discovers that certain cooking methods increase gas consumption.

[1773] The generative AI generates advice such as "Use a pressure cooker to reduce cooking time" and outputs it to the server.

[1774] The server sends the generated advice to the terminal, which notifies the user within the application and displays the recommendation.

[1775] The user enters "I used a pressure cooker" in the application, and the device sends the feedback to the server.

[1776] The server analyzes the feedback data and confirms that gas usage has been reduced.

[1777] The generation AI will incorporate this feedback as learning data and reflect it in generating the next piece of advice.

[1778] Through the above processing, the present invention can improve the efficiency of energy use and contribute to reducing utility costs for homes and businesses.

[1779] The processing flow will be explained below.

[1780] Program processing steps

[1781] 1. Data Collection

[1782] Step 1:

[1783] The device connects to the HEMS and collects electricity, gas, and water usage data, including energy usage information for homes and businesses.

[1784] Step 2:

[1785] The device sends the collected data to the server at specified intervals (e.g., every 3 hours).

[1786] 2. Data Receipt and Storage

[1787] Step 3:

[1788] The server receives the energy usage data transmitted from the terminal.

[1789] Step 4:

[1790] The server stores the received data in a database and organizes it by household or business. The data is recorded in chronological order.

[1791] 3. Data Preprocessing

[1792] Step 5:

[1793] The server extracts energy usage data for a certain period (e.g., the last 30 days) from the database.

[1794] Step 6:

[1795] The server completes the extracted data, removes noise, and detects abnormal high-consumption data and corrects it as necessary.

[1796] 4. Data Analysis and Pattern Recognition

[1797] Step 7:

[1798] The server inputs the preprocessed data into the generative AI model.

[1799] Step 8:

[1800] Generative AI recognizes patterns and unusual trends in energy consumption, identifying increases or decreases in consumption for specific times of day or dates.

[1801] 5. Advice Generation

[1802] Step 9:

[1803] The server receives the output of the generative AI model and generates specific energy-saving advice.

[1804] Step 10:

[1805] Generative AI provides customized advice based on individual consumption patterns and usage.

[1806] 6. Advice Delivery

[1807] Step 11:

[1808] The server transmits the generated energy saving advice to the terminal.

[1809] Step 12:

[1810] The terminal notifies the user of the received advice and displays it within the application.

[1811] 7. Gathering Feedback

[1812] Step 13:

[1813] Users can provide feedback within the application about the effectiveness of the advice they have implemented, by entering specific actions and results.

[1814] Step 14:

[1815] The terminal transmits the feedback input by the user to the server.

[1816] 8. Feedback Analysis

[1817] Step 15:

[1818] The server analyzes the received feedback data and evaluates the effectiveness of the advice.

[1819] Step 16:

[1820] The generative AI learns from the feedback data and reflects it in the next advice generation, aiming to provide improved advice.

[1821] Specific examples

[1822] Examples of reducing electricity usage

[1823] Step 1:

[1824] The terminal obtains 24 hours of electricity usage data from the HEMS every three hours.

[1825] Step 2:

[1826] The terminal transmits the collected data to the server.

[1827] Step 3:

[1828] The server receives the transmitted data.

[1829] Step 4:

[1830] The server stores the received data in a database.

[1831] Step 5:

[1832] The server extracts the electricity usage data for the last 30 days from the database for "Home ID_1234."

[1833] Step 6:

[1834] The server detects and corrects abnormally high data consumption.

[1835] Step 7:

[1836] The generative AI analyzes the data and identifies higher-than-average consumption at night.

[1837] Step 8:

[1838] The AI ​​generates advice such as "unplug appliances if you are not using them at night."

[1839] Step 9:

[1840] The server transmits the generated advice to the terminal.

[1841] Step 10:

[1842] The device will notify the user within the application and display details.

[1843] Step 11:

[1844] The user inputs into the application that "I unplugged my home appliances overnight."

[1845] Step 12:

[1846] The terminal sends the feedback to the server.

[1847] Step 13:

[1848] The server analyzes the feedback data and verifies that electricity usage has been reduced.

[1849] Step 14:

[1850] The generative AI incorporates the feedback as learning data and reflects it in generating the next piece of advice.

[1851] Example 1

[1852] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1853] Conventional energy management systems are limited to collecting and simply analyzing energy usage data, making it difficult to efficiently provide specific energy-saving advice to users. They also lack a mechanism for collecting feedback and reflecting the results in the analysis. This creates the challenge of being unable to provide timely and appropriate energy-saving advice tailored to the user's energy consumption patterns.

[1854] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1855] In this invention, the server includes means for periodically transmitting energy usage data, means for receiving the transmitted energy usage data, means for storing the received energy usage data, means for extracting and preprocessing the stored energy usage data, means for inputting the preprocessed data into a generative AI model to recognize energy consumption patterns, means for generating energy saving advice based on the recognition results, means for transmitting the generated energy saving advice, means for displaying the transmitted energy saving advice, means for collecting feedback on the energy saving advice, and means for analyzing the collected feedback and updating the generative AI model. This integrates the collection and analysis of energy usage data, advice generation, feedback collection, and model updating, making it possible to provide users with timely and effective energy saving advice.

[1856] "Energy usage data" refers to data showing the amount of electricity, gas, water, etc. used, and is a measurement of the specific energy usage status consumed within homes and businesses.

[1857] "HEMS" is an abbreviation for Home Energy Management System, a system for optimizing and managing energy usage within the home.

[1858] A "generative AI model" refers to an artificial intelligence model that performs statistical analysis and machine learning based on various data, and uses the knowledge gained from this to generate new information.

[1859] A "prompt sentence" is an input sentence for a specific generative AI model, and is a sentence that instructs the model to produce a specific output.

[1860] "Data preprocessing" refers to the process carried out before data analysis, and refers to the process of preparing data by filling in missing values ​​and removing noise, etc.

[1861] "Feedback" is information that records the user's response to and results of implementing advice provided by the system, and reflects this information back into the system.

[1862] The present invention relates to a system that collects and analyzes energy usage data and provides energy conservation advice based on the results. This invention makes it possible to improve the efficiency of energy usage in homes and businesses and promote reductions in utility costs. Specifically, this system operates according to the following steps.

[1863] System Configuration

[1864] The system mainly includes the following components:

[1865] 1. Collection Method

[1866] 2. Transmission Method

[1867] 3. Preservation means

[1868] 4. Pretreatment Methods

[1869] 5. Analysis method

[1870] 6. Advice Generation Methods

[1871] 7. Display means

[1872] 8. Feedback Collection Methods

[1873] Details of each component are shown below.

[1874] Collection Method

[1875] The device collects energy usage data. Specifically, the device connects to a HEMS (Home Energy Management System) and obtains electricity, gas, and water usage data every three hours. For example, the device obtains the data using the HEMS API ("GET / energy_usage"). The device temporarily stores the data in local storage and prepares to send it to the server later.

[1876] Transmission method

[1877] The device sends the temporarily stored energy usage data to the server using MQTT or REST API. For example, the data to be sent is in JSON format and is assigned an identifier such as "Home ID_1234."

[1878] Preservation means

[1879] The server receives the energy usage data sent from the device and stores it in a NoSQL database (for example, MongoDB). The server stores the received data separately for each household and each business, and manages it in chronological order, which allows for efficient data access.

[1880] Pretreatment means

[1881] The server extracts the most recent 30 days' worth of data from the database, imputes missing values, and performs preprocessing to remove noise. Specifically, the data is cleansed using Python's pandas library, and outliers are corrected using algorithms such as K-Nearest Neighbor (KNN).

[1882] Analysis means

[1883] The server inputs the preprocessed data into a generative AI model to recognize energy consumption patterns and abnormal trends. An example of the generative AI model is OpenAI's GPT. An example prompt is "Please identify abnormal patterns and signs of energy saving from the energy usage data for the last 30 days for household ID_1234."

[1884] Advice Generation Method

[1885] Based on the analysis results, the generation AI generates energy-saving advice tailored to individual consumption patterns. For example, it may generate advice recommending that appliances be unplugged if not in use at night. The generated advice is sent back to the server, which then formats the information in JSON format and stores it back in the database.

[1886] Display means

[1887] The server sends the generated energy-saving advice to the device. The device saves the received advice in local storage and notifies the user within the application. For example, a smartphone push notification can be used to notify the user that "new energy-saving advice is available."

[1888] Feedback collection methods

[1889] The user acts on the advice and provides feedback on the results within the application. Specifically, the user inputs something like, "I changed the refrigerator's set temperature." The device receives this feedback and sends it to the server. The sent feedback is stored in a feedback database on the server.

[1890] Feedback Analysis

[1891] The server analyzes the received feedback data and evaluates the effectiveness of the advice. The generative AI model learns from this feedback data and reflects it in generating new advice. For example, it learns how much power consumption was reduced by changing the refrigerator's set temperature, and reflects this in the next advice.

[1892] Specific examples

[1893] A specific example of operation is shown below.

[1894] 1. The device obtains electricity usage data from the HEMS every three hours and sends the data to the server.

[1895] 2. The server receives the data and stores it in the database as "Home ID_1234".

[1896] 3. The server extracts and preprocesses the data for the last 30 days.

[1897] 4. The generation AI performs an analysis based on the prompt: "Please identify any abnormal patterns or signs of energy saving from the last 30 days of energy usage data for household ID_1234."

[1898] 5. The AI ​​generates advice such as "It is recommended to unplug the device at night" and sends it to the server.

[1899] 6. The server sends the advice to the device, and the device notifies the user via push notification that "New energy saving advice is available."

[1900] 7. The user enters into the application that "I unplugged my appliances overnight," and the device sends feedback to the server.

[1901] 8. The server analyzes the feedback, and the AI ​​uses it to generate the next piece of advice.

[1902] As described above, the present invention can improve the efficiency of energy use and contribute to reducing utility costs for homes and businesses.

[1903] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1904] Step 1: Data collection

[1905] The device connects to the HEMS and obtains electricity, gas, and water usage data every three hours. Specifically, it accesses the HEMS API endpoint "GET / energy_usage" to receive the latest energy usage data. The input is sensor data from the HEMS, and the output is usage data saved in the device's temporary storage. For example, data in the format "Electricity usage: 10kWh" is saved.

[1906] Step 2: Send data

[1907] The energy usage data collected by the device is sent to the server. The data is sent periodically (for example, every three hours) using MQTT or REST API. The input is temporarily stored usage data, and the output is the data to be sent to the server. Specifically, data such as "electricity usage: 10kWh, gas usage: 5m3, water usage: 100L" is structured in JSON format and sent to the server.

[1908] Step 3: Data reception and storage

[1909] The server receives the data sent from the device and stores it in a NoSQL database (for example, MongoDB). The input is the JSON data sent from the device, and the output is the energy usage data stored in the database. Specifically, the received data is organized by household and company, and stored using an identifier such as "Home ID_1234."

[1910] Step 4: Preprocessing the data

[1911] The server extracts the last 30 days' worth of energy usage data from the database. It then performs pre-processing to fill in missing values ​​in the extracted data and remove noise. The input is energy usage data from the database (e.g., the last 30 days' worth of data for "household ID_1234"), and the output is pre-processed, clean data. Specifically, it uses Python's pandas library to fill in missing values ​​and the KNN algorithm to detect and correct outliers.

[1912] Step 5: Data analysis and pattern recognition

[1913] The server inputs the preprocessed data into the generative AI model. The input is the preprocessed data (e.g., energy usage data for the last 30 days for "household ID_1234"), and the output is consumption patterns and abnormal trends recognized by the generative AI model. Specifically, the generative AI is prompted with the following prompt: "Please identify abnormal patterns and signs of energy saving from the energy usage data for the last 30 days for household ID_1234."

[1914] Step 6: Advice Generation

[1915] The generative AI generates energy-saving advice based on the analysis results. The input is the output of the generative AI model (e.g., consumption patterns and abnormal trends), and the output is specific energy-saving advice. For example, the generated advice recommends "unplugging appliances when not in use at night." This advice is sent to the server, which then formats the advice content in JSON format and stores it in a database.

[1916] Step 7: Advice Delivery

[1917] The server sends the generated energy-saving advice to the device. The input is the generated advice (e.g., "Unplug appliances if they are not in use at night"), and the output is the data sent to the device. The device saves the received advice in local storage and notifies the user within the application. Specifically, it uses a push notification on the smartphone to notify the user that "new energy-saving advice is available."

[1918] Step 8: Gather feedback

[1919] The results of the user's execution of advice within the application are fed back. The input is information about the user's actions (e.g., "I changed the refrigerator's set temperature"), and the output is the feedback data. The device receives feedback from the user and sends it to the server. Specifically, it formats the user's input in JSON format and sends a POST request to the server.

[1920] Step 9: Analyze feedback

[1921] The server analyzes the received feedback data and evaluates the effectiveness of the advice. The input is the feedback data (e.g., "As a result of changing the set temperature, power consumption decreased by 5%), and the output is the evaluation result. The generation AI learns from this feedback data and reflects it in generating new advice. For example, it learns that "power consumption at night has decreased" and uses this information to generate the next piece of advice.

[1922] The above are the specific processing steps of this system.

[1923] (Application example 1)

[1924] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1925] In modern brick-and-mortar stores, efficient energy use is important from the perspective of environmental protection and cost reduction. However, in reality, there is a lack of means to grasp energy usage patterns in detail and receive efficient energy-saving advice. As a result, store managers are unable to implement appropriate energy-saving measures, and wasteful energy consumption continues. The objective of this invention is to promote efficient energy use and reduce wasteful energy consumption by analyzing energy usage data in brick-and-mortar stores in detail and providing specific energy-saving advice.

[1926] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1927] In this invention, the server includes means for collecting energy usage data, means for transmitting the energy usage data, means for storing the transmitted energy usage data, means for preprocessing the stored energy usage data, means for analyzing the stored energy usage data, means for generating energy saving advice based on the analysis results, means for generating the generated energy saving advice for specific actions in the store, means for displaying the generated energy saving advice, and means for collecting feedback on the energy saving advice. This makes it possible to understand energy usage patterns in the physical store in detail and to specifically implement efficient energy saving measures.

[1928] "Energy usage data" refers to data that records the amount of energy consumed in a physical store, such as electricity, gas, and water.

[1929] The "transmission means" is a function for transmitting collected energy usage data to a server.

[1930] The "storage means" is a function for storing the transmitted energy usage data in a storage device such as a database.

[1931] The "preprocessing means" is a function for performing preprocessing such as complementing missing values ​​and removing outliers on stored energy usage data.

[1932] The "analysis means" is a function for analyzing the pre-processed energy usage data and identifying patterns and anomalies in energy consumption.

[1933] The "generation means" is a function for generating energy saving advice based on the analysis results.

[1934] The "display means" is a function for presenting the generated energy saving advice to the operator of the physical store.

[1935] The "feedback collection means" is a function for collecting feedback regarding the energy saving advice that the operator has implemented.

[1936] "Generative AI" is an artificial intelligence model that analyzes energy usage data and generates specific energy-saving advice.

[1937] A "prompt sentence" is an instruction sentence that provides specific energy-saving advice to the generation AI.

[1938] "In-store energy saving advice" is advice that recommends specific energy saving actions for specific equipment or activities in a physical store.

[1939] The present invention relates to a system that collects and analyzes energy usage data and provides energy conservation advice based on the results, thereby making it possible to improve the efficiency of energy usage in physical stores and promote reductions in utility costs.

[1940] The system includes a collection means, a transmission means, a storage means, a pre-processing means, an analysis means, a generation means, a display means, and a feedback collection means for handling energy usage data.

[1941] Specific hardware and software configuration:

[1942] The terminal, a smartphone or tablet, connects to the HEMS (Home Energy Management System) in the physical store and collects energy usage data.

[1943] The server receives the collected energy usage data and stores it in a database, where it is organized by household and business and recorded in chronological order.

[1944] The preprocessing means extracts energy usage data for a certain period from the database and performs preprocessing such as filling in missing values ​​and removing outliers, thereby ensuring the reliability of the data.

[1945] The analytics tool feeds the pre-processed data into a generative AI model to identify patterns and unusual trends in energy consumption.

[1946] The AI ​​that generates the energy-saving advice is based on the analysis results and is specific to the user's consumption patterns.

[1947] The terminal as a display means notifies the user of the generated energy saving advice and displays it within the application.

[1948] The feedback collection means collects feedback regarding the effect of the energy saving advice implemented by the user and transmits the feedback to the server.

[1949] Software used and detailed data processing:

[1950] Data collection: The device acquires energy usage data from the HEMS. The data is sent to the server periodically (e.g., every three hours).

[1951] Data storage and preprocessing: The server stores the data in a database and preprocesses it, including imputing missing values ​​and removing outliers.

[1952] Data analysis and pattern recognition: Analyze the pre-processed data and recognize consumption patterns and unusual trends from historical data. Generative AI models take on this role and automate the process.

[1953] Generating energy-saving advice: The AI ​​generates energy-saving advice based on the analysis results. For example, it may generate advice such as, "Since power consumption is high at night, we recommend turning off devices outside of business hours."

[1954] User notification and feedback collection: The generated advice is notified to the user and feedback is collected within the application. The feedback is sent back to the server and reflected in subsequent advice generation.

[1955] Examples and prompts:

[1956] Example: Identifying high consumption at night and making specific recommendations on which appliances in a store should be turned off outside of business hours.

[1957] Example prompt: "Our store consumes a lot of electricity at night. Please have your Generative AI suggest which devices in our store should be turned off after hours."

[1958] In this way, the present invention efficiently manages energy usage in physical stores and provides specific energy-saving advice, enabling operators of physical stores to implement appropriate energy-saving measures.

[1959] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1960] Step 1:

[1961] Data collection

[1962] The terminal obtains electricity, gas, and water usage data every three hours from the physical store's HEMS (Home Energy Management System). The input is real-time energy usage data from the HEMS, and the output is collected energy usage data, which is used for further analysis.

[1963] Step 2:

[1964] Data transmission

[1965] The terminal periodically transmits the collected energy usage data to the server. The input is the collected energy usage data, and the output is the energy usage data transmitted to the server. The transmitted data is stored on the server.

[1966] Step 3:

[1967] Data storage

[1968] The server receives the transmitted energy usage data and stores it in a database. The input is the energy usage data transmitted to the server, and the output is the energy usage data stored in the database. This data is used for later analysis.

[1969] Step 4:

[1970] Data Preprocessing

[1971] The server extracts energy usage data for a certain period of time (e.g., the last 30 days) from the database, and completes missing values ​​and removes outliers. The input is the energy usage data stored in the database, and the output is the corrected and preprocessed energy usage data. Specifically, it completes missing values ​​and removes outliers using the 3 sigma rule.

[1972] Step 5:

[1973] Data Analysis and Pattern Recognition

[1974] The server inputs the preprocessed energy usage data into a generative AI model to identify energy consumption patterns and anomalies. The input is the preprocessed energy usage data, and the output is the recognition results of consumption patterns and anomalies. Specifically, it finds consumption patterns from past data and identifies abnormal trends.

[1975] Step 6:

[1976] Energy saving advice generation

[1977] The server uses a generative AI model based on the analysis results to generate energy-saving advice. The input is the consumption pattern and anomaly recognition results, and the output is the generated energy-saving advice. For example, it generates specific advice such as "Since consumption is high at night, it is recommended that you turn off devices outside of business hours."

[1978] Step 7:

[1979] Advice display

[1980] The device notifies the user of the generated energy-saving advice and displays it within the application. The input is the generated energy-saving advice, and the output is the advice displayed to the user. Specifically, a notification is sent to the user's smartphone or tablet.

[1981] Step 8:

[1982] Feedback collection

[1983] The user inputs feedback about the effectiveness of the energy-saving advice they have implemented through the application. The input is the user's feedback, and the output is feedback data. Specifically, the user inputs an action such as "I unplugged my home appliances at night" into the application.

[1984] Step 9:

[1985] Feedback Analysis

[1986] The server analyzes the received feedback data, evaluates the effectiveness of the advice, and provides feedback to the generation AI model. The input is the feedback data sent by the user, and the output is the effectiveness evaluation result. Specifically, the effectiveness of the energy-saving advice is evaluated and reflected in the generation of the next advice.

[1987] By following these steps, energy usage in physical stores can be managed efficiently and specific energy-saving advice can be provided.

[1988] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1989] System Overview

[1990] The present invention combines a system that collects and analyzes energy usage data and provides energy-saving advice based on the results with an emotion engine that recognizes user emotions. This makes it possible to improve the efficiency of energy use in homes and businesses, promote reductions in utility bills, and provide advice that takes user emotions into consideration. The system mainly includes a collection means for handling energy usage data, a transmission means, a storage means, an analysis means, a generation means, a display means, an emotion engine, and a feedback collection means.

[1991] Program processing explanation

[1992] 1. Data Collection

[1993] The terminals are connected to the HEMS (Home Energy Management System) to collect electricity, gas, and water usage data from each home or business.

[1994] The terminal periodically (e.g., every three hours) obtains usage data from the HEMS and sends it to the server.

[1995] 2. Data Receipt and Storage

[1996] The server receives the energy usage data transmitted from the terminal.

[1997] The server stores the received data in a database, organizing it by household and business and recording it in chronological order.

[1998] 3. Data Preprocessing

[1999] The server extracts energy usage data for a certain period (e.g., the last 30 days) from the database.

[2000] The server performs pre-processing to complement missing values ​​in the extracted data and remove noise. Specifically, it detects and corrects abnormally high-consumption data.

[2001] 4. Data Analysis and Pattern Recognition

[2002] The server inputs the preprocessed data into the generative AI model.

[2003] Generative AI uses past data to recognize patterns of energy consumption and unusual trends.

[2004] 5. Advice Generation

[2005] The server generates specific energy-saving advice based on the output from the generative AI model.

[2006] Generative AI provides advice tailored to individual consumption patterns.

[2007] 6. Emotion recognition and advice adjustment

[2008] The device analyzes the user's feedback, voice, and facial expressions to recognize the user's emotions.

[2009] The emotion engine tailors the generated energy-saving advice based on the recognized emotion: for example, if the user is feeling stressed, it generates a message containing encouraging and kind words.

[2010] 7. Advice Delivery

[2011] The server transmits the adjusted energy saving advice to the terminal.

[2012] The terminal notifies the user of the received advice and displays it within the application.

[2013] 8. Collecting Feedback

[2014] Users can provide feedback within the application about the effectiveness of the advice they have implemented, by entering specific actions and results.

[2015] The terminal transmits the feedback input by the user to the server.

[2016] 9. Feedback Analysis

[2017] The server analyzes the received feedback data and evaluates the effectiveness of the advice.

[2018] The generation AI learns from the feedback data and reflects it in generating the next piece of advice.

[2019] Specific examples

[2020] Examples of reducing electricity usage

[2021] System Operation

[2022] The device retrieves 24 hours of electricity usage data from the HEMS every three hours.

[2023] The terminal transmits the collected data to the server.

[2024] The server receives the data and stores it in a database.

[2025] The server extracts the electricity usage data for the last 30 days from the database for "Home ID_1234."

[2026] The server detects and corrects abnormally high data consumption.

[2027] Generative AI analyzes the data and identifies higher-than-average consumption at night.

[2028] The AI ​​generates advice such as "unplug appliances if you are not using them at night."

[2029] The device analyzes the user's feedback, voice, and facial expressions, and uses an emotion engine to check whether the user is feeling stressed.

[2030] The emotion engine tailors advice as needed, adding encouragement and kind words.

[2031] The server sends the tailored advice to the terminal, which notifies the user within the application and displays the details.

[2032] The user inputs into the application that "I unplugged my home appliances overnight."

[2033] The device sends the feedback to the server.

[2034] The server analyzes the feedback data and confirms that electricity usage has been reduced.

[2035] The generation AI will incorporate this feedback as learning data and reflect it in generating the next piece of advice.

[2036] The system not only improves energy efficiency and reduces utility bills, but also enhances the user experience by providing emotionally sensitive advice, making energy management in homes and businesses more effective and satisfying.

[2037] The processing flow will be explained below.

[2038] Program processing steps

[2039] 1. Data Collection

[2040] Step 1:

[2041] The device connects to the HEMS and collects electricity, gas, and water usage data, which is obtained from sensors and meters installed in each home or business.

[2042] Step 2:

[2043] The device sends the collected usage data to the server at specified intervals (e.g., every 3 hours).

[2044] 2. Data Receipt and Storage

[2045] Step 3:

[2046] The server receives the energy usage data transmitted from the terminal.

[2047] Step 4:

[2048] The server stores the received data in a database, organizing it by household and business and recording it in chronological order.

[2049] 3. Data Preprocessing

[2050] Step 5:

[2051] The server detects missing values ​​in the energy usage data and completes them as necessary.

[2052] Step 6:

[2053] The server scrutinizes the data to remove abnormal values ​​and noise, and detects and corrects abnormally high consumption data.

[2054] 4. Data Analysis and Pattern Recognition

[2055] Step 7:

[2056] The server inputs the preprocessed data into the generative AI model.

[2057] Step 8:

[2058] Generative AI recognizes patterns and unusual trends in energy consumption, for example, identifying sudden increases or decreases in consumption on certain days or during certain times of the day.

[2059] 5. Advice Generation

[2060] Step 9:

[2061] The server generates specific advice for energy conservation based on the analysis results of the generation AI.

[2062] Step 10:

[2063] The generative AI provides customized advice based on consumption patterns, such as "unplug appliances if you're not using them overnight" or "use a pressure cooker to reduce cooking time."

[2064] 6. Emotion recognition and advice adjustment

[2065] Step 11:

[2066] The device analyzes the user's feedback, voice, and facial expressions to recognize the user's emotions. Technologies used include voice recognition and facial recognition.

[2067] Step 12:

[2068] The emotion engine adjusts the energy-saving advice based on the recognized emotion. For example, if the user is feeling stressed, the engine generates advice that includes encouraging and kind words.

[2069] 7. Advice Delivery

[2070] Step 13:

[2071] The server transmits the adjusted energy saving advice to the terminal.

[2072] Step 14:

[2073] The device will notify the user and provide detailed advice within the application, which the user can review at any time.

[2074] 8. Collecting Feedback

[2075] Step 15:

[2076] The user provides feedback on the effectiveness of the advice they received within the application and whether they implemented it. For example, they can input information such as "I changed the refrigerator temperature setting" or "I used a pressure cooker."

[2077] Step 16:

[2078] The terminal transmits the feedback input by the user to the server.

[2079] 9. Feedback Analysis

[2080] Step 17:

[2081] The server analyzes the received feedback data and evaluates the effectiveness of the advice, for example, checking whether electricity usage has been reduced.

[2082] Step 18:

[2083] The AI ​​learns from the feedback data and reflects it in the next advice generation, allowing it to provide more effective and tailored advice to the user.

[2084] Specific examples

[2085] Reducing electricity consumption

[2086] Step 1:

[2087] The device retrieves 24 hours of electricity usage data from the HEMS every three hours.

[2088] Step 2:

[2089] The terminal transmits the collected data to the server.

[2090] Step 3:

[2091] The server receives the transmitted data.

[2092] Step 4:

[2093] The server stores the received data in a database.

[2094] Step 5:

[2095] The server extracts the electricity usage data for the last 30 days from the database for "Home ID_1234."

[2096] Step 6:

[2097] The server detects and corrects abnormally high data consumption.

[2098] Step 7:

[2099] Generative AI analyzes the data and identifies higher-than-average consumption at night.

[2100] Step 8:

[2101] The AI ​​generates advice such as "unplug appliances if you are not using them at night."

[2102] Step 9:

[2103] The server transmits the generated advice to the terminal.

[2104] Step 10:

[2105] The device will notify the user within the application and display details.

[2106] Step 11:

[2107] The device analyzes the user's voice and facial expressions and uses an emotion engine to check whether the user is feeling stressed.

[2108] Step 12:

[2109] The emotion engine tailors advice as needed, adding encouragement and kind words.

[2110] Step 13:

[2111] The server sends the adjusted advice to the terminal.

[2112] Step 14:

[2113] The device will notify the user within the application and display details.

[2114] Step 15:

[2115] The user inputs into the application that "I unplugged my home appliances overnight."

[2116] Step 16:

[2117] The device sends the feedback to the server.

[2118] Step 17:

[2119] The server analyzes the feedback data and confirms that electricity usage has been reduced.

[2120] Step 18:

[2121] The generation AI will incorporate this feedback as learning data and reflect it in generating the next piece of advice.

[2122] Example 2

[2123] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[2124] Existing systems that collect and analyze energy usage data and provide energy-saving advice rarely provide advice that takes users' emotions into consideration, resulting in insufficient user satisfaction and energy reduction effects. Another problem is that there is a lack of a mechanism for continuously improving the performance of energy-saving advice using collected feedback, which limits the effectiveness of the advice.

[2125] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[2126] In this invention, the server includes means for collecting energy usage data, means for transmitting the energy usage data, means for storing the transmitted energy usage data, means for preprocessing the stored energy usage data, means for analyzing the preprocessed data and recognizing energy consumption patterns, means for generating energy saving advice based on the analysis results, means for adjusting the generated energy saving advice based on a user's emotions, means for displaying the adjusted energy saving advice, means for collecting feedback on the energy saving advice, and means for analyzing the collected feedback and improving the performance of the energy saving advice generator. This makes it possible to provide energy saving advice that takes the user's emotions into consideration, thereby improving user satisfaction and maximizing energy reduction effects.

[2127] "Energy usage data" refers to data that indicates the amount of energy consumed, such as electricity, gas, and water, in homes and businesses.

[2128] "Collection means" refers to the means for capturing energy usage data, and specifically includes connection to devices such as HEMS (Home Energy Management System).

[2129] The "transmission means" is a means for sending the collected energy usage data to the server via a network.

[2130] The "storage means" is a means for storing and managing the energy usage data sent to the server by the transmission means in a database.

[2131] The "preprocessing means" is a means for complementing missing values ​​and removing noise from the stored energy usage data.

[2132] "Analysis means" refers to means for using the pre-processed data to recognize energy consumption patterns and detect abnormal trends.

[2133] "Generation means" refers to the means for creating specific energy-saving advice based on the analysis results, and includes the use of a generative AI model.

[2134] The "adjustment means" is a means for modifying the content of the generated energy saving advice based on the user's feelings.

[2135] The "display means" is a means for providing the user with the adjusted energy saving advice, and includes displays and notifications within the application.

[2136] The "feedback collection means" is a means for transmitting to the server the effects of advice taken by the user and their impressions.

[2137] The "feedback analysis means" is a means used to analyze collected feedback data and improve the performance of subsequent advice generation.

[2138] The present invention is a system that collects and analyzes energy usage data and provides energy-saving advice based on the results, and combines it with an emotion engine that recognizes the user's emotions. This system makes it possible to improve the efficiency of energy usage in homes and businesses, promote reductions in utility costs, and provide advice that takes the user's emotions into consideration. An embodiment of the present invention is described in detail below.

[2139] System Overview

[2140] The system mainly includes a collection means, a transmission means, a storage means, a pre-processing means, an analysis means, a generation means, an adjustment means, a display means, a feedback collection means, and a feedback analysis means for handling energy usage data. These means are executed by terminals installed in homes and businesses and a server that collects and analyzes data.

[2141] Data collection

[2142] The device connects to a HEMS (Home Energy Management System) installed in a home or business to collect electricity, gas, and water usage data. This connection is via Wi-Fi or wired LAN. The device encrypts the collected energy usage data and sends it to a server over the network.

[2143] Data Receipt and Storage

[2144] The server receives the energy usage data sent from the device and stores it in a database in real time, so that the data is properly categorized and recorded in chronological order.

[2145] Data Preprocessing

[2146] The server extracts energy usage data for the past 30 days from the database and performs preprocessing such as filling in missing values ​​and removing noise. Specifically, it fills in missing data using linear interpolation and average value interpolation, detects abnormally high consumption data, and corrects it based on past trends.

[2147] Data Analysis and Pattern Recognition

[2148] The server inputs the preprocessed data into a generative AI model to recognize energy consumption patterns and abnormal trends. The generative AI model learns past consumption patterns based on a large amount of data, enabling highly accurate pattern recognition.

[2149] Example prompt: "Generate energy conservation advice based on the past 30 days of energy consumption data, specifically nighttime consumption patterns."

[2150] Advice Generation

[2151] The generator generates specific energy-saving advice based on the analysis results, such as "unplug appliances when not in use at night."

[2152] Emotion Recognition and Advice Adjustment

[2153] The device collects the user's feedback, voice, and facial expressions through a camera and microphone, and the emotion engine analyzes them. If the user is feeling stressed, the emotion engine generates advice with encouraging and kind words.

[2154] Advice Delivery

[2155] The server then sends the adjusted energy saving advice to the device, which then notifies the user within the application, for example, by using a push notification.

[2156] Feedback collection

[2157] The user inputs feedback about the effectiveness of the advice within the application. Specifically, the application records the action of "unplugging home appliances overnight." The device then sends the user-entered feedback to the server.

[2158] Feedback Analysis

[2159] The server analyzes the received feedback data and checks whether energy usage has actually been reduced. The AI ​​learns from this feedback data and reflects it in the generation of next advice.

[2160] This not only improves energy efficiency and reduces utility bills, but also improves the user experience by providing emotionally sensitive advice, making energy management in homes and businesses more effective and satisfying.

[2161] The flow of the identification process in the second embodiment will be described with reference to FIG.

[2162] The flow of this system's program processing

[2163] Step 1: Data collection

[2164] The terminal connects to a HEMS (Home Energy Management System).

[2165] Input: Energy usage data from a HEMS installed in the user's home or business

[2166] Output: Collected energy usage data

[2167] Specific operation: Acquires and collects electricity, gas, and water usage data from the HEMS. This data includes timestamps and consumption amounts.

[2168] Step 2: Send data

[2169] The terminal transmits the collected energy usage data to a server.

[2170] Input: Energy usage data obtained from HEMS

[2171] Output: Energy usage data sent to the server

[2172] What it does: Encrypts data and sends it over the internet to a server via Wi-Fi or wired LAN.

[2173] Step 3: Data reception and storage

[2174] The server receives the energy usage data sent from the terminal and stores it in a database.

[2175] Input: Energy usage data sent from the device

[2176] Output: Energy usage data stored in a database

[2177] Specific operation: The received data is stored in a database in real time and categorized by household or business.

[2178] Step 4: Preprocessing the data

[2179] The server extracts energy usage data for the last 30 days from the database, fills in missing values, and removes noise.

[2180] Input: Energy usage data extracted from a database

[2181] Output: Preprocessed energy usage data

[2182] Specific operation: For the extracted data, missing values ​​are filled in using linear interpolation or mean value interpolation, and abnormal high-consumption data is detected and corrected.

[2183] Step 5: Data analysis and pattern recognition

[2184] The server feeds the pre-processed data into a generative AI model to recognize energy consumption patterns and unusual trends.

[2185] Input: Preprocessed energy usage data

[2186] Output: Recognized energy consumption patterns and unusual trends

[2187] Specific operation: Data is input into the generative AI model to analyze consumption patterns and detect anomalies.

[2188] Example prompt: "Generate energy conservation advice based on the past 30 days of energy consumption data, specifically nighttime consumption patterns."

[2189] Step 6: Advice Generation

[2190] The server generates specific energy-saving advice using the output of the generative AI model.

[2191] Input: Analysis results from a generative AI model

[2192] Output: Specific energy saving advice

[2193] Specific actions: Based on the analysis results, specific advice for energy conservation is generated, such as "unplug appliances at night."

[2194] Step 7: Emotion recognition and advice adjustment

[2195] The device collects the user's feedback, voice, and facial expressions, which are then analyzed by an emotion engine.

[2196] Input: User feedback, voice, facial expressions

[2197] Output: Energy saving advice tailored based on user's emotions

[2198] Specific behavior: The system uses a camera and microphone to recognize the user's emotions and adjusts the content of the energy-saving advice it generates based on the user's emotions. If the user is feeling stressed, it will include encouraging or kind words.

[2199] Step 8: Advice Delivery

[2200] The server sends the adjusted energy saving advice to the terminal, which notifies the user within the application.

[2201] Input: Tailored energy saving advice

[2202] Output: Advice given to the user

[2203] Specific operation: The adjusted advice is sent to the device, and the device notifies the user through the application.

[2204] Step 9: Gather feedback

[2205] The user enters feedback within the application about the effectiveness of the advice they have implemented.

[2206] Input: User feedback on advice taken

[2207] Output: Feedback sent to the device

[2208] Specific actions: Using the application, you can input specific actions and their results. For example, you can input "I unplugged the appliances overnight."

[2209] Step 10: Feedback analysis

[2210] The server analyzes the received feedback data and reflects it in the generation of the next advice.

[2211] Input: Feedback data sent from the device

[2212] Output: More accurate next energy saving advice

[2213] What it does: Analyzes the feedback, checks whether electricity usage has been reduced, and trains the generating AI based on the feedback data, which improves the accuracy of the next advice.

[2214] (Application example 2)

[2215] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[2216] Conventional energy management systems were able to collect energy usage data from homes and businesses and provide energy-saving advice, but they did not take user emotions into account when providing advice or collecting and analyzing feedback. This resulted in issues such as a lack of improvement in the user experience and limited effectiveness of advice. Furthermore, in large facilities such as factories, energy consumption patterns are complex, so conventional energy management systems were unable to achieve sufficient energy-saving effects.

[2217] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[2218] In this invention, the server includes means for collecting energy usage data, means for transmitting the energy usage data, means for storing the transmitted energy usage data, means for analyzing the stored energy usage data, means for generating energy saving advice based on the analysis results, means for recognizing a user's emotion when generating the energy saving advice and adjusting the content of the advice in accordance with the recognized emotion, and means for collecting feedback on the energy saving advice. This achieves more efficient energy use and reduced utility costs, and also provides advice that takes the user's emotion into consideration, improving the user experience and enabling more satisfying energy management.

[2219] "Energy usage data" refers to data on the amount of energy used, such as electricity, gas, and water, by households, businesses, factories, etc.

[2220] "Means of collection" refers to devices or systems for obtaining energy usage data, specifically sensors and meters.

[2221] The "transmitting means" refers to the communication device or protocol for transferring the collected energy usage data to the server.

[2222] "Means for storage" refers to a database or storage device for retaining and managing the transmitted energy usage data for a long period of time.

[2223] "Means for analysis" refers to software or algorithms used to analyze stored energy usage data and detect consumption patterns and outliers.

[2224] "Means for generating energy-saving advice" refers to a generative AI model or program that generates specific advice recommending energy-efficient ways of using energy based on the analyzed data.

[2225] The "display means" refers to a display, monitor, or application that notifies the user of the generated energy saving advice and visually presents it to the user.

[2226] "Means for recognizing emotions" refers to emotion recognition engines or software that analyze and identify emotions from the user's voice, facial expressions, etc.

[2227] The "means for collecting feedback" refers to an interface or system for collecting information about the results and effects of the energy conservation advice that the user has implemented.

[2228] "In-factory energy usage data" refers to data on the amount of energy consumed by the factory's production lines and each device and equipment.

[2229] A "generative AI model" is an artificial intelligence model that generates new energy-saving advice based on past energy usage data.

[2230] A "prompt sentence" is an input sentence that instructs the generative AI model on what kind of analysis or advice to generate.

[2231] The system that realizes this application example is a combination of specific hardware and software for energy management and emotion recognition within a factory. The main components are sensors and meters for collecting energy usage data, a server for managing and analyzing the data, a robot terminal that provides an interface with the user, and an emotion engine for recognizing emotions. A detailed example of the system is described below.

[2232] Hardware and Software

[2233] 1. Sensors and meters

[2234] Sensors: Installed on each production line in the factory, they collect real-time data on electricity, gas, water, etc. For example, smart meters and environmental monitoring sensors are used.

[2235] Meter: A meter that measures the energy consumption of each device or equipment, such as a smart electricity meter.

[2236] 2. Server

[2237] Database: A database for centrally storing collected energy usage data. For example, MySQL can be used.

[2238] Analysis software: Analyzes the stored data to detect energy consumption patterns and outliers, for example using data analysis libraries such as Pandas and Scikit-learn.

[2239] Generative AI model: A generative AI model for generating energy-saving advice based on data analysis results. For example, the latest generative AI models such as GPT-4 can be used.

[2240] 3. Robot terminal

[2241] Display: A display to inform and visually present energy saving advice to the user.

[2242] Voice input device: A microphone to transmit the user's voice to the emotion recognition engine.

[2243] Emotion recognition engine: Analyzes the user's emotions and adjusts the advice content. For example, OpenFace or Emotion API is used.

[2244] Data processing details

[2245] The server uses sensors and meters to collect energy usage data from each production line in the factory. The collected data is sent to the server and stored in a database. The stored data is then pre-processed using analytical software to detect energy consumption patterns and outliers.

[2246] The analyzed data is input into a generative AI model, which generates specific energy-saving advice. Furthermore, during this generation process, an emotion recognition engine is used to analyze the user's emotions and tailor the advice content based on the recognized emotions. For example, if the user is feeling stressed, encouraging or kind words may be added. This tailored advice is then displayed on the robot terminal's display, informing the user.

[2247] Specific examples

[2248] For example, if analysis reveals that a factory's production line consumes a lot of energy at night, the generative AI model...

Claims

1. a means for collecting energy usage data; means for transmitting the energy usage data; means for storing the transmitted energy usage data; means for analyzing the stored energy usage data; means for generating energy saving advice based on the analysis results; a means for displaying the generated energy saving advice; means for collecting feedback on the energy saving advice; A system including:

2. The system according to claim 1 , wherein the energy usage data is electricity, gas, and water usage data.

3. The system according to claim 1 , wherein the energy saving advice is generated using a generation AI.

4. The system of claim 1 , wherein the energy usage data is stored and analyzed on a server.

5. The system according to claim 1 , wherein the energy saving advice is notified to the user via a terminal.

Images