Method and system for monitoring elderly people living alone through electricity utilization information

Abstract

The invention relates to the technical field of electricity utilization information application, in particular to a method and system for monitoring elderly people living alone through electricity utilization information, and the method comprises the steps: analyzing the electricity utilization behavior characteristics of a user through a neural network model based on first historical electricity utilization information, and recognizing the elderly people living alone; for the elderly people living alone, acquiring second historical electricity consumption information according to a preset period, and generating a database according to the second historical electricity consumption information; obtaining the state of the elderly user living alone according to the current power consumption data and a database; and responding according to the state of the elderly user living alone and a preset grading response mechanism. According to the method, the neural network is utilized to perform feature extraction on the multi-dimensional power consumption data to identify the elderly people living alone, compared with a traditional threshold method, the identification accuracy is improved, and the corresponding database is established according to the identified historical power consumption information of the elderly people living alone and considering the individual power consumption behavior difference of the elderly people living alone; and the accuracy of a daily monitoring result is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electricity usage information application, and in particular to a method and system for monitoring elderly people living alone through electricity usage information. Background Art

[0002] With the accelerating global aging population, the number of elderly people living alone continues to expand, necessitating urgent home safety monitoring. Traditional monitoring methods, such as wearable devices, face challenges such as low compliance (with forgetfulness rates as high as 37%), privacy concerns surrounding video surveillance, and delayed responses to regular community visits. In recent years, electricity consumption analysis technology based on non-intrusive load monitoring (NILM) has gained increasing attention due to its non-invasive monitoring capabilities, but existing technologies still have significant limitations.

[0003] Some studies have attempted to judge the status by matching electricity consumption patterns during different time periods, but have failed to effectively solve the problem of differences in individual electricity consumption behaviors. In particular, there is a lack of in-depth modeling of the electricity consumption characteristics of the special group of elderly people living alone, which makes it easy to make false alarms. Summary of the Invention

[0004] (1) Purpose of the invention

[0005] The purpose of the present invention is to provide a method and system for monitoring elderly people living alone through electricity usage information, which can improve monitoring accuracy.

[0006] (2) Technical solution

[0007] To solve the above problems, the present invention provides a method for monitoring elderly people living alone through electricity usage information, comprising:

[0008] Based on the first historical electricity consumption information, a neural network model is used to analyze the user's electricity consumption behavior characteristics and identify elderly users living alone;

[0009] For the elderly user living alone, obtaining second historical electricity usage information according to a preset period, and generating a database based on the second historical electricity usage information;

[0010] Obtain the status of elderly users living alone based on current electricity consumption data and the database;

[0011] Respond based on the status of elderly users living alone and the preset hierarchical response mechanism.

[0012] In another aspect of the present invention, preferably, the first historical electricity usage information includes device type, electricity usage period, electricity usage frequency and power data;

[0013] The method of using a neural network model to analyze the user's electricity usage behavior characteristics and identify elderly users living alone includes:

[0014] Extracting time series features from the first historical electricity usage information to generate an input feature vector, the feature vector including: a power usage period distribution curve, a daily usage frequency of a preset high-power device, a nighttime silent duration indicator, and a seasonal power usage pattern difference;

[0015] The feature vector is input into a two-layer LSTM neural network, and the last layer of the two-layer LSTM neural network uses a Sigmoid activation function to output the probability of living alone;

[0016] Based on the probability of living alone, elderly users living alone are identified and a manual review mechanism is triggered.

[0017] In another aspect of the present invention, preferably, for the elderly user living alone, obtaining second historical electricity usage information according to a preset period, and generating a database based on the second historical electricity usage information includes:

[0018] collecting second historical electricity usage information of the elderly user living alone according to a preset period, wherein the second historical electricity usage information includes a distribution curve of electricity usage time periods, a usage frequency of high-power equipment, and an instantaneous power value;

[0019] The second historical electricity usage information is classified and stored according to a preset period and preset structured items to generate a power usage pattern baseline, which constitutes a database.

[0020] In another aspect of the present invention, preferably, obtaining the status of the elderly user living alone based on the current electricity usage data and the database includes:

[0021] Calculate the similarity between the power consumption time distribution curve of the current power consumption data and the power consumption time distribution curve of the power consumption pattern baseline at the corresponding time in the database;

[0022] Calculate the percentage deviation of the high-power device usage frequency in the current power consumption data from the power consumption pattern baseline;

[0023] Calculate the difference multiple between the instantaneous power value in the current power consumption data and the instantaneous power value of the power consumption pattern baseline;

[0024] The status of the elderly user living alone is obtained according to the similarity, deviation percentage and difference multiple.

[0025] In another aspect of the present invention, preferably, the similarity is calculated using the following formula:

[0026]

[0027] Among them, D represents the similarity, y 1i The i-th data point y represents the time period distribution curve of the current electricity consumption data 2i represents the i-th data point of the power consumption period distribution curve of the power consumption pattern baseline at the corresponding time, and N represents the total number of data points.

[0028] In another aspect of the present invention, preferably, the deviation percentage is calculated using the following formula:

[0029]

[0030] Where ΔF represents the deviation percentage, F current Indicates the frequency of high-power equipment usage in the current power consumption data, F baseline Indicates the usage frequency of high-power devices in the power usage pattern baseline;

[0031] The difference multiple is calculated using the following formula:

[0032]

[0033] Where SDM represents the difference multiple, x represents the instantaneous power value in the current power consumption data, and μ represents the instantaneous power value of the power consumption pattern baseline.

[0034] In another aspect of the present invention, preferably, obtaining the status of the elderly user living alone according to the similarity, deviation percentage, and difference multiple includes:

[0035] If any two of the similarity, deviation percentage, and difference multiple exceed a preset first safety threshold, the status of the elderly user living alone is considered primary abnormality;

[0036] If the similarity, deviation percentage and difference multiple exceed the preset first safety threshold, the status of the elderly user living alone is intermediate abnormality;

[0037] If any two of the similarity, deviation percentage, and difference multiple exceed a preset second safety threshold, the status of the elderly user living alone is considered to be a high-level abnormality;

[0038] The preset first safety threshold is smaller than the preset second safety threshold.

[0039] In another aspect of the present invention, preferably, responding according to the status of the elderly user living alone and a preset hierarchical response mechanism includes:

[0040] If the status of the elderly user living alone is a primary anomaly, observe the first time period. If there is no recovery within the first time period, respond as a high-level anomaly.

[0041] If the status of the elderly user living alone is a medium-level abnormality, observe the second time period. If there is no recovery within the second time period, respond as a high-level abnormality;

[0042] If the status of the elderly user living alone is a high-level abnormality, the electricity meter generates an event status word containing the abnormality level code and reports it to the master station.

[0043] In another aspect of the present invention, preferably, the method of reporting to the master station includes: raising the EVENTOUT pin level of the carrier channel to trigger the carrier module to read the event status word and upload it; or the carrier module actively encapsulates the event record as a protocol frame and pushes it to the master station in broadcast mode.

[0044] In another aspect of the present invention, preferably, a system for monitoring elderly people living alone through electricity usage information comprises:

[0045] Identification module: Based on the first historical electricity usage information, a neural network model is used to analyze the user's electricity usage behavior characteristics and identify elderly users living alone;

[0046] A generation module: for the elderly user living alone, obtaining second historical electricity usage information according to a preset period, and generating a database based on the second historical electricity usage information;

[0047] Acquisition module: obtains the status of elderly users living alone based on current electricity consumption data and database;

[0048] Response module: responds based on the status of elderly users living alone and the preset hierarchical response mechanism.

[0049] (3) Beneficial effects

[0050] The above technical solution of the present invention has the following beneficial technical effects:

[0051] The present invention uses a neural network to extract features from multi-dimensional electricity consumption data to identify elderly people living alone. Compared with the traditional threshold method, the recognition accuracy is improved. Based on the historical electricity consumption information of the identified elderly people living alone, the differences in their individual electricity consumption behaviors are taken into consideration to establish a corresponding database, thereby improving the accuracy of daily monitoring results. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is an overall flow chart of an embodiment of the present invention. DETAILED DESCRIPTION

[0053] To make the objectives, technical solutions, and advantages of the present invention more clearly understood, the present invention will be further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings. It should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the present invention. In addition, in the following description, descriptions of well-known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present invention.

[0054] Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0055] In the description of the present invention, it should be noted that the terms "first", "second" and "third" are only used for descriptive purposes and should not be understood as indicating or implying relative importance.

[0056] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0057] The present invention will be described in more detail below with reference to the accompanying drawings. In each of the accompanying drawings, identical elements are represented by similar reference numerals. For the sake of clarity, the various parts in the accompanying drawings are not drawn to scale.

[0058] Example 1

[0059] A method for monitoring elderly people living alone through electricity usage information. Figure 1 FIG. 1 shows an overall flow chart of an embodiment of the present invention, as shown in FIG. Figure 1 Shown, including:

[0060] Based on the first historical electricity usage information, a neural network model is used to analyze the user's electricity usage behavior characteristics and identify elderly users living alone. The specific content of the first historical electricity usage information is not limited here. The first historical electricity usage information is a wide range of electricity usage information. In this embodiment, the first historical electricity usage information includes device type, electricity usage period, electricity usage frequency and power data. First, in the early stage, the identification of elderly users living alone is carried out. The user's electricity usage information is picked up through the load identification extension module of the Internet of Things electricity meter. The picked-up information includes the electricity usage period, electricity usage frequency, etc. of each electrical device (such as a washing machine, air conditioner, refrigerator, microwave oven, etc.). The analysis of the user's electricity usage information shows that, for example, the user will not use electricity very early in the evening, but will start using electricity very early in the morning, which is in line with the elderly's habit of going to bed and getting up early.

[0061] The method of using a neural network model to analyze the characteristics of user electricity consumption behavior and identify elderly users living alone includes: extracting time series features from the first historical electricity consumption information to generate an input feature vector, wherein the feature vector includes: a power consumption period distribution curve, a preset daily usage frequency of high-power equipment, a nighttime silent duration index, and a seasonal power consumption pattern difference. For example, the power consumption period distribution curve can be drawn by dividing each day into the average power ratio of the three periods of morning, noon, and evening; the preset daily usage frequency of high-power equipment, including the number of starts and stops of air conditioners and washing machines and the duration of each start and stop; the nighttime silent duration index, which calculates the duration of continuous no electricity consumption records from 22:00 to 6:00 the next day; and the seasonal power consumption pattern difference, which calculates the variance of the basic power fluctuation of refrigerators by quarter.

[0062] The feature vector is input into a two-layer LSTM neural network. The final layer of the two-layer LSTM neural network uses a Sigmoid activation function to output the probability of living alone. Based on this probability, elderly users living alone are identified and a manual review mechanism is triggered. The probability of living alone is compared with a preset standard. For example, if the probability of living alone exceeds 0.85 for 30 consecutive days, the user is identified as an elderly user living alone and a manual review mechanism is triggered. The use of a two-layer LSTM to process time series data differs from general neural network structures and emphasizes improved adaptability to the temporal continuity of electricity usage behavior. Multiple safeguards are implemented to avoid privacy risks caused by misidentification.

[0063] For the elderly user living alone, second historical electricity usage information is obtained at a preset period, and a database is generated based on this second historical electricity usage information. The preset period can be divided into hours, and the second historical electricity usage information is obtained at an hourly granularity. The second historical electricity usage information includes a power consumption time distribution curve, the frequency of high-power device usage, and the instantaneous power value. The second historical electricity usage information is categorized and stored according to the preset period and preset structured items to generate a power usage pattern baseline, which constitutes the database. The power usage pattern baseline can include a baseline based on seasonal characteristics, such as a threshold for the average daily activation of heating equipment in winter and a safe upper limit for the continuous operation time of air conditioners in summer.

[0064] Based on the current electricity consumption data and the database, the status of elderly users living alone is obtained, including:

[0065] Calculate the similarity between the time period distribution curve of the current electricity consumption data and the time period distribution curve of the electricity consumption pattern baseline at the corresponding time in the database; extract the electricity consumption pattern baseline for the corresponding time period (such as the same period in the past week or month) from the database. The electricity consumption pattern baseline represents the time period distribution pattern of electricity consumption for elderly people living alone in normal living conditions. At the same time, divide the current electricity consumption data into the same time periods to generate the current electricity consumption time period distribution curve. The similarity is calculated using the following formula:

[0066]

[0067] Among them, D represents the similarity, y 1i The i-th data point y represents the time period distribution curve of the current electricity consumption data 2i The i-th data point of the electricity usage time distribution curve represents the baseline electricity usage pattern at the corresponding time, and N represents the total number of data points. The above algorithm can handle nonlinear alignment issues in time series data and effectively measure the similarity between two curves in shape and trend. A higher similarity value indicates that the current electricity usage time distribution is more consistent with the historical pattern, indicating that the elderly person living alone is likely living a normal life. Conversely, a lower similarity value indicates that there may be anomalies, such as changes in sleep schedules or absence from home.

[0068] Calculate the percentage deviation of the high-power device usage frequency in the current electricity usage data from the electricity usage pattern baseline. Identify high-power devices in the home, such as air conditioners, water heaters, and rice cookers, using smart meters or attached sensors. Count the usage frequency of each high-power device in the current electricity usage data and compare it with the corresponding data in the electricity usage pattern baseline. The deviation percentage is calculated using the following formula:

[0069]

[0070] Where ΔF represents the deviation percentage, F current Indicates the frequency of high-power equipment usage in the current power consumption data, F baseline Indicates the frequency of high-power device usage from the baseline power usage pattern. A large deviation from the baseline percentage may indicate a sudden change in the lifestyle of an elderly person living alone, such as a decrease in daily activities due to declining health, or abnormal appliance usage, such as forgetting to turn off an appliance.

[0071] Calculate the difference multiple between the instantaneous power value in the current power consumption data and the instantaneous power value of the power consumption pattern baseline; monitor the instantaneous power value of the elderly living alone in real time, and record the power data at key time points (such as peak power consumption hours). Compare the current instantaneous power value with the instantaneous power value at the corresponding time point in the power consumption pattern baseline, calculate the difference, and calculate the difference multiple using the following formula:

[0072]

[0073] Where SDM represents the difference multiple, x represents the instantaneous power value in the current power consumption data, and μ represents the instantaneous power value of the baseline power consumption pattern. A significant increase in the difference multiple may indicate the connection of new high-power equipment or an abnormal increase in energy consumption due to a failure of existing equipment. This may signal a change in the living conditions of elderly people living alone.

[0074] According to the similarity, deviation percentage and difference multiple, the status of the elderly user living alone is obtained, including:

[0075] If any two of the similarity, deviation percentage, and difference multiple exceed a preset first safety threshold, the elderly user's status is considered primary abnormal, indicating that their electricity usage behavior has deviated to a certain extent, but not yet reached an emergency level. This status may be caused by a slight change in lifestyle habits or a short-term abnormal event.

[0076] If the similarity, deviation percentage and difference multiple exceed the preset first safety threshold, the status of the elderly user living alone is intermediate abnormality; it indicates that the abnormal status is relatively obvious, which may involve major changes in living habits or persistent abnormal behavior, and requires attention and further investigation.

[0077] If any two of the similarity, deviation percentage and difference multiple exceed the preset second safety threshold, the status of the elderly user living alone is a high-level abnormality; when any two indicators exceed the second safety threshold, it indicates that the abnormal state is extremely serious and may be accompanied by an emergency (such as sudden illness, accidental fall, etc.), and the emergency response mechanism needs to be activated immediately.

[0078] The preset first safety threshold is lower than the preset second safety threshold. The first safety threshold serves as a primary warning line, capturing relatively mild abnormal fluctuations. This low threshold is designed to ensure sensitivity to subtle changes and avoid missing potential risks. The second safety threshold serves as the critical point for advanced warnings, reflecting more significant abnormal conditions. The second safety threshold is higher than the first safety threshold, ensuring higher confidence in the determination of advanced anomalies and reducing false alarms.

[0079] Responses are made based on the status of elderly people living alone and the preset hierarchical response mechanism, including:

[0080] If the status of the elderly user living alone is a primary anomaly, observe the first time period. If there is no recovery within the first time period, respond as a high-level anomaly.

[0081] If the status of the elderly user living alone is a medium-level abnormality, observe the second time period. If there is no recovery within the second time period, respond as a high-level abnormality;

[0082] If an elderly person living alone experiences a high-level abnormality, the meter generates an event status word containing the abnormality level code and reports it to the master station. For primary and intermediate abnormalities, an observation period is set to assess the persistence and severity of the abnormality, avoiding misjudgments caused by short-term fluctuations. For high-level abnormalities, an emergency response mechanism is immediately activated to ensure the safety of the elderly person.

[0083] The method of reporting to the master station includes: raising the EVENTOUT pin level of the carrier channel to trigger the carrier module to read the event status word and upload it; or the carrier module actively encapsulates the event record into a protocol frame and pushes it to the master station in broadcast mode. When a high-level abnormality occurs, the electricity meter records the corresponding event according to the set event record, and frames it according to the current smart IoT electricity meter event reporting status word communication protocol frame format, and reports the abnormality to the master station based on the level of the event. By raising the EVENTOUT pin of the carrier channel, the carrier module reads the event status word or event record and reports it to the master station; or the carrier module actively initiates reporting information to the master station, and the master station can notify the user's property or relatives by email, text message or phone call.

[0084] The present invention uses a neural network to extract features from multi-dimensional electricity consumption data to identify elderly people living alone. Compared with the traditional threshold method, the recognition accuracy is improved. Based on the historical electricity consumption information of the identified elderly people living alone, the differences in their individual electricity consumption behaviors are taken into consideration to establish a corresponding database, thereby improving the accuracy of daily monitoring results.

[0085] Example 2

[0086] A system for monitoring elderly people living alone through electricity usage information, comprising:

[0087] Identification module: Based on the first historical electricity usage information, a neural network model is used to analyze the user's electricity usage behavior characteristics and identify elderly users living alone;

[0088] A generation module: for the elderly user living alone, obtaining second historical electricity usage information according to a preset period, and generating a database based on the second historical electricity usage information;

[0089] Acquisition module: obtains the status of elderly users living alone based on current electricity consumption data and the database; response module: responds based on the status of elderly users living alone and the preset hierarchical response mechanism.

[0090] It should be understood that the above-described specific embodiments of the present invention are merely illustrative or illustrative of the principles of the present invention and do not constitute limitations of the present invention. Therefore, any modifications, equivalent substitutions, improvements, etc. made without departing from the spirit and scope of the present invention should be included within the scope of protection of the present invention. In addition, the appended claims are intended to cover all variations and modifications that fall within the scope and metes and bounds of the appended claims, or equivalents thereof.

[0091] The present invention has been described above with reference to the embodiments thereof. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of the present invention. The scope of the present invention is defined by the appended claims and their equivalents. Those skilled in the art may make various substitutions and modifications without departing from the scope of the present invention, and such substitutions and modifications are intended to fall within the scope of the present invention.

[0092] Although the embodiments of the present invention have been described in detail, it should be understood that the various changes, substitutions, and alterations could be made hereto without departing from the spirit and scope of the invention.

[0093] Obviously, the above embodiments are merely examples for clarity of explanation and are not intended to limit the implementation methods. Those skilled in the art will readily appreciate that other variations or modifications based on the above descriptions are possible. It is not necessary and impossible to enumerate all implementation methods here. Obvious variations or modifications arising therefrom remain within the scope of protection of the present invention.

Claims

1. A method for monitoring elderly people living alone through electricity usage information, characterized in that: include: Based on the first historical electricity consumption information, a neural network model is used to analyze the user's electricity consumption behavior characteristics and identify elderly users living alone; For the elderly user living alone, obtaining second historical electricity usage information according to a preset period, and generating a database based on the second historical electricity usage information; Obtain the status of elderly users living alone based on current electricity consumption data and the database; Respond based on the status of elderly users living alone and the preset hierarchical response mechanism.

2. The method for monitoring elderly people living alone through electricity usage information according to claim 1, characterized in that: The first historical electricity usage information includes device type, electricity usage period, electricity usage frequency and power data; The method of using a neural network model to analyze the user's electricity usage behavior characteristics and identify elderly users living alone includes: Extracting time series features from the first historical electricity usage information to generate an input feature vector, the feature vector including: a power usage period distribution curve, a preset daily usage frequency of high-power equipment, a nighttime silent duration indicator, and a seasonal power usage pattern difference; The feature vector is input into a two-layer LSTM neural network, and the last layer of the two-layer LSTM neural network uses a Sigmoid activation function to output the probability of living alone; Based on the probability of living alone, elderly users living alone are identified and a manual review mechanism is triggered.

3. The method for monitoring elderly people living alone through electricity usage information according to claim 1, characterized in that: For the elderly user living alone, obtaining second historical electricity usage information according to a preset period, and generating a database based on the second historical electricity usage information includes: collecting second historical electricity usage information of the elderly user living alone according to a preset period, wherein the second historical electricity usage information includes a distribution curve of electricity usage time periods, a usage frequency of high-power equipment, and an instantaneous power value; The second historical electricity usage information is classified and stored according to a preset period and preset structured items to generate a power usage pattern baseline, which constitutes a database.

4. The method for monitoring elderly people living alone through electricity usage information according to claim 3, characterized in that: Based on the current electricity consumption data and the database, the status of elderly users living alone is obtained, including: Calculate the similarity between the power consumption time distribution curve of the current power consumption data and the power consumption time distribution curve of the power consumption pattern baseline at the corresponding time in the database; Calculate the percentage deviation of the high-power device usage frequency in the current power consumption data from the power consumption pattern baseline; Calculate the difference multiple between the instantaneous power value in the current power consumption data and the instantaneous power value of the power consumption pattern baseline; The status of the elderly user living alone is obtained according to the similarity, deviation percentage and difference multiple.

5. The method for monitoring elderly people living alone through electricity usage information according to claim 3, characterized in that: The similarity is calculated using the following formula: Among them, D represents the similarity, y 1i The i-th data point y represents the time period distribution curve of the current electricity consumption data 2i represents the i-th data point of the power consumption period distribution curve of the power consumption pattern baseline at the corresponding time, and N represents the total number of data points.

6. The method for monitoring elderly people living alone through electricity usage information according to claim 3, characterized in that: The deviation percentage is calculated using the following formula: Where ΔF represents the deviation percentage, F current Indicates the frequency of high-power equipment usage in the current power consumption data, F baseline Indicates the usage frequency of high-power devices in the power usage pattern baseline; The difference multiple is calculated using the following formula: Where SDM represents the difference multiple, x represents the instantaneous power value in the current power consumption data, and μ represents the instantaneous power value of the power consumption pattern baseline.

7. The method for monitoring elderly people living alone through electricity usage information according to claim 6, characterized in that: According to the similarity, deviation percentage, and difference multiple, obtaining the status of the elderly user living alone includes: If any two of the similarity, deviation percentage, and difference multiple exceed a preset first safety threshold, the status of the elderly user living alone is considered primary abnormality; If the similarity, deviation percentage and difference multiple exceed the preset first safety threshold, the status of the elderly user living alone is intermediate abnormality; If any two of the similarity, deviation percentage, and difference multiple exceed a preset second safety threshold, the status of the elderly user living alone is considered to be a high-level abnormality; The preset first safety threshold is smaller than the preset second safety threshold.

8. The method for monitoring elderly people living alone through electricity usage information according to claim 7, characterized in that: Based on the status of the elderly user living alone and the preset hierarchical response mechanism, the response includes: If the status of the elderly user living alone is a primary anomaly, observe the first time period. If there is no recovery within the first time period, respond as a high-level anomaly. If the status of the elderly user living alone is a medium-level abnormality, observe the second time period. If there is no recovery within the second time period, respond as a high-level abnormality; If the status of the elderly user living alone is a high-level abnormality, the electricity meter generates an event status word containing the abnormality level code and reports it to the master station.

9. The method for monitoring elderly people living alone through electricity usage information according to claim 8, characterized in that: The method of reporting to the master station includes: pulling up the EVENTOUT pin level of the carrier channel to trigger the carrier module to read the event status word and upload it; or the carrier module actively encapsulates the event record into a protocol frame and pushes it to the master station in broadcast mode.

10. A system for monitoring elderly people living alone through electricity usage information, characterized in that: include: Identification module: Based on the first historical electricity usage information, a neural network model is used to analyze the user's electricity usage behavior characteristics and identify elderly users living alone; A generation module: for the elderly user living alone, obtaining second historical electricity usage information according to a preset period, and generating a database based on the second historical electricity usage information; Acquisition module: obtains the status of elderly users living alone based on current electricity consumption data and database; Response module: responds based on the status of elderly users living alone and the preset hierarchical response mechanism.

Images