Intelligent electric energy meter-oriented residential electricity consumption portrait construction method

Abstract

The invention provides a residential electricity consumption portrait construction method for an intelligent electric energy meter, and relates to the field of electric power systems, and the method comprises the following steps: S1, data collection; s2, constructing a situation coding system; s3, constructing a situation power utilization incidence matrix; s4, constructing a causal reasoning model; s5, generating an anti-factual power utilization scene; s6, determining an optimal energy-saving path; according to the method, the situation-electricity utilization incidence matrix and the causal reasoning model are constructed, and the intelligent electric meter data, the meteorological data, the social economic data and the intelligent household data are integrated, so that the accuracy of the electricity utilization portrait is greatly improved, more accurate electricity utilization insight is provided for residents, and the experience of the residents is improved. The dynamic portrait generation mechanism adaptively adjusts the portrait updating frequency according to the stability of the power consumption behavior, the average energy saving rate of residents is improved, real economic benefits are brought to the residents, and meanwhile, the implementation of the dual-carbon target is facilitated.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of power systems, in particular to a method for constructing a resident electricity portrait based on a smart electricity meter. BACKGROUND

[0002] With the deepening of the construction of smart grids and the increasing complexity of resident electricity behavior, traditional electricity analysis methods have been unable to meet the needs of fine electricity management. Currently, smart electricity meter data analysis mainly focuses on the statistical characteristics of electricity consumption, such as daily average electricity consumption, peak-valley electricity consumption ratio, etc.

[0003] In recent years, research on electricity user portraits has gradually emerged, which can be divided into two categories. One is the architecture type portrait, which builds different technical frameworks for user portraits. The other is the use of user portrait classification models, such as using fuzzy clustering algorithm for customer electricity consumption pattern analysis to realize the electricity behavior pattern portrait of different types of users.

[0004] The existing publication CN118710329A discloses a method and system for constructing a user portrait based on the electricity consumption behavior of a proxy electricity purchasing user. The method includes the following steps: collecting proxy electricity purchasing user related data, constructing a proxy electricity purchasing user data set, integrating an electricity consumption mode industry group and an industry electricity quantity influence factor library to form a portrait framework; based on the proxy electricity purchasing user data set, analyzing the similarity between the capacity utilization rate of the proxy electricity purchasing user and the industry capacity utilization rate, and extracting a typical user according to the similarity; classifying the typical user according to the electricity consumption behavior and electricity consumption mode of the typical user, calculating the characteristic index value of the typical user according to the preset business characteristic index; assigning a user label to the typical user through the user type and the characteristic index value, integrating the user label and the industry label into the portrait framework to generate a multi-dimensional portrait of the typical user; However, the above-mentioned electricity portrait is mainly based on historical data and is statically constructed, which cannot dynamically adapt to changes in resident electricity behavior, resulting in poor timeliness of the portrait and difficulty in providing timely and effective electricity guidance. Only correlation analysis is concerned, such as the correlation between electricity consumption and temperature, but the causal relationship is not deeply mined, which cannot determine which factors really affect electricity consumption behavior, resulting in a lack of scientific basis for energy saving suggestions.

[0005] Therefore, it is necessary to provide a new method for constructing a resident electricity portrait based on a smart electricity meter to solve the above technical problems. SUMMARY

[0006] To solve the above technical problems, the present application provides a method for constructing a resident electricity portrait based on a smart electricity meter.

[0007] The method for constructing a resident electricity portrait based on a smart electricity meter provided by the present application includes the following steps: S1, data collection: collect smart meter data, meteorological data, social and economic data, and smart home data, and perform data cleaning, standardization, and time series alignment on the collected data; S2, constructing a context coding system: based on time, environment, and society, a context coding system is constructed, using clustering algorithm, temperature data is divided into n clusters, each cluster corresponds to an encoding interval, and decision tree algorithm is used to determine the optimal boundary point of population structure; S3, constructing a context electricity correlation matrix: taking time context, environmental context, population structure, and occupation type as grouping conditions, calculating the average electricity consumption of each group, taking the average electricity consumption of the reference context as the reference value, and calculating the context-sensitive electricity rate; S4, causal reasoning model construction: select key features as nodes of causal graph, use NOTEARS algorithm to construct structural causal model, optimize objective function, and calculate average treatment effect to quantify causal effect of key factors; S5, counterfactual electricity scenario generation: determine the counterfactual target, set the target variable to the counterfactual value based on the causal model to generate counterfactual samples, use random forest regression model to predict electricity consumption under counterfactual context to generate counterfactual electricity scenarios, for the same target variable, generate counterfactual scenarios of different intensity, then generate diverse counterfactual samples through implicit counterfactual learning method, avoid bias introduced by explicit modification of text structure or attributes; S6, determining the optimal energy-saving path: calculate the energy-saving potential of each counterfactual scenario, quantify the causal effect of each factor, calculate the comprehensive benefit, determine the optimal path through multi-objective optimization considering implementation difficulty and resident acceptance, according to the comprehensive benefit ranking, select the top n paths with the highest benefit as the recommended paths; S7, dynamic portrait generation: based on context-sensitive electricity rate, causal effect, and counterfactual electricity scenario, construct a dynamic resident electricity portrait.

[0008] Further method, the smart meter data includes active power, reactive power, voltage and current; the meteorological data includes temperature, humidity and light intensity; the social and economic data includes family population and occupation type; the smart home data includes device electricity consumption.

[0009] Further method, the context electricity correlation matrix construction includes the following steps: S301, taking time context, environmental context, population structure, and occupation type as grouping conditions, calculating the average electricity consumption of each group; S302, taking the average electricity consumption of the reference context as the reference value, calculating the context-sensitive electricity rate by dividing the difference between the electricity consumption and the reference value by the reference value.

[0010] Further method, the causal graph construction includes the following steps: S401. Select key features as nodes in the cause-effect graph, including time context, environmental context, population structure, occupation type, and smart socket installation status; S402. Use the NOTEARS algorithm to construct a structural causal model and determine causal relationships by optimizing the objective function; S403. Use a meta-learner to quantify causal effects by comparing the difference in power consumption between implementing and not implementing smart sockets, and calculate the average impact.

[0011] A further method, wherein the generation of the counterfactual electricity consumption scenario includes the following steps: S501. Identify counterfactual objectives; S502. Generate counterfactual samples: Copy the original dataset and set the target variable to a counterfactual value; S503. Generate counterfactual electricity consumption scenarios: Use a random forest regression model to predict electricity consumption under counterfactual scenarios; S504. Generate multiple sets of counterfactual scenarios: Generate counterfactual scenarios of different intensities for the same target variable; S505. Generate diverse counterfactual samples through implicit counterfactual learning methods.

[0012] A further method, wherein determining the optimal energy-saving path includes the following steps: S601. Calculate the energy-saving potential of each counterfactual scenario: actual electricity consumption minus counterfactual electricity consumption; S602. Quantify the causal effects of each factor: Calculate the average impact by comparing the difference in electricity consumption between implementing and not implementing the factor. S603. Calculate implementation costs: including equipment costs, installation costs, and usage costs; S604. Calculate the overall benefits: Energy saving potential divided by implementation costs; S605. Determine the optimal path through multi-objective optimization: Select the path with the highest overall benefits based on considerations of energy-saving benefits, implementation difficulty, and residents' acceptance. S606. Generate the optimal energy-saving path: Based on the overall benefits, select the top 3 paths with the highest benefits as recommended paths.

[0013] A further method, the construction of the dynamic residential electricity consumption profile includes the following steps: S701. Calculate the stability of electricity consumption behavior: Assess the stability of electricity consumption behavior by calculating the standard deviation of electricity consumption. S702. Calculate the adaptive weight: the reciprocal of the standard deviation of electricity consumption, plus a constant; S703, Normalized Adaptive Weights: Adjust the adaptive weights to the range of 0 to 1; S704. Predicting future scenarios based on historical data: Using a random forest classifier to predict scenario encoding for future dates; S705. Generate a dynamic electricity consumption profile: Combine adaptive weights, context-sensitive electricity consumption rate, and causal effects to generate a dynamic electricity consumption profile.

[0014] In a further method, the data cleaning uses linear interpolation to fill in missing values ​​and then uses the IQR method to identify and process outliers.

[0015] Compared with related technologies, the method for constructing residential electricity consumption profiles for smart meters provided by this invention has the following beneficial effects: 1. This invention significantly improves the accuracy of electricity consumption profiles by constructing a context-electricity consumption correlation matrix and a causal reasoning model, and by integrating smart meter data, meteorological data, socio-economic data and smart home data, providing residents with more accurate insights into electricity consumption.

[0016] 2. This invention is based on a causal model. It sets the target variable as a counterfactual value to generate counterfactual samples. It uses a random forest regression model to predict electricity consumption under counterfactual scenarios to generate counterfactual electricity consumption scenarios. For the same target variable, it generates counterfactual scenarios of different intensities. Then, through an implicit counterfactual learning method, it generates diverse counterfactual samples, avoiding biases introduced by explicitly modifying text structure or attributes. The dynamic profile generation mechanism adaptively adjusts the profile update frequency according to the stability of electricity consumption behavior, improves the average energy saving rate of residents, brings tangible economic benefits to residents, and helps achieve dual carbon targets. Attached Figure Description

[0017] Figure 1 A flowchart illustrating the method for constructing residential electricity consumption profiles for smart meters provided by this invention; Figure 2 A flowchart illustrating the construction of the contextual electricity correlation matrix provided by this invention; Figure 3 A flowchart illustrating the construction of the causal graph provided by this invention; Figure 4 A flowchart for generating a counterfactual electricity consumption scenario provided by the present invention; Figure 5 A flowchart illustrating the determination of the optimal energy-saving path provided by this invention; Figure 6 A flowchart illustrating the construction of a dynamic residential electricity consumption profile provided by this invention. Detailed Implementation

[0018] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0019] Please refer to the following: Figure 1 , Figure 2 ,Figure 3 , Figure 4 , Figure 5 as well as Figure 6 ,in, Figure 1 A flowchart illustrating the method for constructing residential electricity consumption profiles for smart meters provided by this invention; Figure 2 A flowchart illustrating the construction of the contextual electricity correlation matrix provided by this invention; Figure 3 A flowchart illustrating the construction of the causal graph provided by this invention; Figure 4 A flowchart for generating a counterfactual electricity consumption scenario provided by the present invention; Figure 5 A flowchart illustrating the determination of the optimal energy-saving path provided by this invention; Figure 6 A flowchart illustrating the construction of a dynamic residential electricity consumption profile provided by this invention.

[0020] In the specific implementation process, such as Figure 1 As shown, the method for constructing a residential electricity consumption profile for smart meters includes the following steps: S1. Data Acquisition: Collect smart meter data, meteorological data, socio-economic data, and smart home data; perform data cleaning, standardization, and time series alignment on the collected data. Data is collected from smart meters every 15 minutes, including active energy. Reactive power ,Voltage and current ; Collect meteorological data, obtain temperature data hourly. ,humidity and light intensity ; Collect socioeconomic data, including household population. Occupational type ; Collects smart home data and obtains device power consumption every 10 minutes. ; Data cleaning: Fill in missing values ​​using linear interpolation. ,in This is the electricity consumption data for the time point preceding the point in time where the values ​​are missing. The data represents electricity consumption at the point in time following the point where missing values ​​were found. The IQR method is used to identify and handle outliers in the electricity consumption data. satisfy or When, replace it with or ,in and These are the 25th and 75th percentiles of electricity consumption, respectively. ,in, This is the lower threshold for outliers; data below this value are considered outliers. This is the upper limit threshold for outliers; data exceeding this value are considered outliers. This is the lower limit value after outlier processing, used to replace abnormal electricity consumption values ​​below Q1-3IQR. This is the upper limit value after outlier processing, used to replace abnormal electricity consumption values ​​that are higher than Q3+3IQR; Standardize the data , These are raw data values, i.e., data such as electricity consumption that have not undergone standardization. This represents the maximum value in the dataset, i.e., the highest value among all electricity consumption data. It is the minimum value in the dataset, that is, the lowest value among all electricity consumption data; Aligning Time Series ,in, This refers to the electricity consumption data at the previous time point. For the current time point that needs to be aligned, The moment of the previous time point, The current time point, This is the electricity consumption data at the current point in time; S2. Constructing a contextual coding system: Construct a contextual coding system based on three dimensions: time, environment, and society. Use a clustering algorithm to divide the temperature data into four clusters, each cluster corresponding to a coding interval. Then use a decision tree algorithm to determine the optimal boundary point of the population structure. S3. Construct a contextual electricity consumption correlation matrix: Calculate the average electricity consumption for each group based on time context, environmental context, population structure, and occupation type. Calculate the context-sensitive electricity consumption rate using the average electricity consumption of the baseline context as the baseline value. S4. Causal Inference Model Construction: Select key features as nodes in the causal graph, use the NOTEARS algorithm to construct a structural causal model, optimize the objective function, and calculate the average treatment effect to quantify the causal effect of key factors. S5. Counterfactual Electricity Consumption Scenario Generation: Determine the counterfactual target, and based on the causal model, set the target variable as the counterfactual value to generate counterfactual samples. Use the random forest regression model to predict the electricity consumption under the counterfactual scenario to generate counterfactual electricity consumption scenarios. For the same target variable, generate counterfactual scenarios of different intensities. Then, through the implicit counterfactual learning method, generate diverse counterfactual samples to avoid biases introduced by explicitly modifying the text structure or attributes. S6. Determining the optimal energy-saving path: Calculate the energy-saving potential of each counterfactual scenario, quantify the causal effects of each factor, calculate the comprehensive benefits, and determine the optimal path through multi-objective optimization, taking into account both implementation difficulty and residents' acceptance. Based on the comprehensive benefits, select the top 3 paths with the highest benefits as recommended paths. S7. Dynamic profile generation: Based on context-sensitive electricity consumption rate, causal effect and counterfactual electricity consumption scenarios, construct dynamic residential electricity consumption profiles.

[0021] It should be noted that the context coding system is constructed based on three dimensions: time, environment, and society. The time dimension refers to calculating the day of the week for each date. ,like (Monday to Friday) ;like (Saturday and Sunday) ; Environmental dimension: based on temperature , Computational environment context coding: ; Social dimension: based on family size Calculate the population structure code: According to occupation type Calculate the occupation type code: ; The optimal coding interval was determined by cluster analysis: the temperature data was divided into 4 clusters using the K-means clustering algorithm, with each cluster corresponding to a coding interval; the optimal boundary point of the population structure was determined using the decision tree algorithm.

[0022] refer to Figure 2 As shown, the construction of the contextual electricity consumption correlation matrix includes the following steps: S301, Based on time context Context Population structure Occupational type Calculate the average electricity consumption for each group, based on the grouping criteria: ,in For the time context, For environmental context, For population structure, As for occupation type, This represents the number of data points in this scenario combination. This refers to the average electricity consumption, which is the average of all electricity consumption data under a specific combination of scenarios. This represents the electricity consumption value for the m-th data point. S302, Average electricity consumption under the baseline scenario Using the baseline value, the scenario-sensitive electricity consumption rate is calculated by dividing the difference between the electricity consumption and the baseline value by the baseline value. .

[0023] refer to Figure 3 As shown, the construction of a causal graph includes the following steps: S401. Select key features as nodes in the cause-effect graph, including time context, environmental context, population structure, occupation type, and smart socket installation status; S402. Use the NOTEARS algorithm to construct a structural causal model and determine causal relationships by optimizing the objective function; S403. Use a meta-learner to quantify causal effects by comparing the difference in power consumption between implementing and not implementing smart sockets, and calculate the average impact.

[0024] refer to Figure 4 As shown, the generation of counterfactual electricity usage scenarios includes the following steps: S501. Identify counterfactual objectives; S502. Generate counterfactual samples: Copy the original dataset and set the target variable to a counterfactual value; S503. Generate counterfactual electricity consumption scenarios: Use a random forest regression model to predict electricity consumption under counterfactual scenarios; S504. Generate multiple sets of counterfactual scenarios: Generate counterfactual scenarios of different intensities for the same target variable; S505. Generate diverse counterfactual samples through implicit counterfactual learning methods.

[0025] refer to Figure 5 As shown, determining the optimal energy-saving path includes the following steps: S601. Calculate the energy-saving potential of each counterfactual scenario: actual electricity consumption minus counterfactual electricity consumption; S602. Quantify the causal effects of each factor: Calculate the average impact by comparing the difference in electricity consumption between implementing and not implementing the factor. S603. Calculate implementation costs: including equipment costs, installation costs, and usage costs; S604. Calculate the overall benefits: Energy saving potential divided by implementation costs; S605. Determine the optimal path through multi-objective optimization: Select the path with the highest overall benefits based on considerations of energy-saving benefits, implementation difficulty, and residents' acceptance. S606. Generate the optimal energy-saving path: Based on the overall benefits, select the top 3 paths with the highest benefits as recommended paths.

[0026] refer to Figure 6As shown, the construction of a dynamic residential electricity consumption profile includes the following steps: S701. Calculate the stability of electricity consumption behavior: Assess the stability of electricity consumption behavior by calculating the standard deviation of electricity consumption. S702. Calculate the adaptive weight: the reciprocal of the standard deviation of electricity consumption, plus a constant; S703, Normalized Adaptive Weights: Adjust the adaptive weights to the range of 0 to 1; S704. Predicting future scenarios based on historical data: Using a random forest classifier to predict scenario encoding for future dates; S705. Generate a dynamic electricity consumption profile: Combine adaptive weights, context-sensitive electricity consumption rate, and causal effects to generate a dynamic electricity consumption profile.

[0027] According to embodiments of the present invention, a computing device that can be used to implement the above method includes a processor and a memory; The processor can be a multi-core processor or include multiple processors. In some embodiments, the processor may include a general-purpose main processor and one or more special coprocessors, such as a graphics processing unit (GPU), a digital signal processor (DSP), etc. In some embodiments, the processor may be implemented using custom circuitry, such as an application-specific integrated circuit (ASIC) or a field-programmable gate array (FPGA).

[0028] Memory can include various types of storage units, such as system memory, read-only memory (ROM), and permanent storage devices. ROM can store static data or instructions required by the processor or other modules of the computer. Permanent storage devices can be read-write storage devices. Permanent storage devices can be non-volatile storage devices that retain stored instructions and data even when the computer is powered off. In some embodiments, permanent storage devices use mass storage devices (e.g., magnetic or optical disks, flash memory) as permanent storage devices. In other embodiments, permanent storage devices can be removable storage devices (e.g., floppy disks, optical drives). System memory can be a read-write storage device or a volatile read-write storage device, such as dynamic random access memory. System memory can store some or all of the instructions and data required by the processor during operation. Furthermore, memory can include any combination of computer-readable storage media, including various types of semiconductor memory chips (DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), and disks and / or optical disks can also be used. In some implementations, the memory may include removable storage devices that are readable and / or writable, such as laser discs (CDs), read-only digital versatile optical discs (e.g., DVD-ROMs, dual-layer DVD-ROMs), read-only Blu-ray discs, ultra-high density optical discs, flash memory cards (e.g., SD cards, mini SD cards, Micro-SD cards, etc.), magnetic floppy disks, etc. Computer-readable storage media do not include carrier waves and transient electronic signals transmitted wirelessly or via wired connections.

[0029] It should be understood that, unless otherwise expressly stated herein, there is no strict order restriction on the execution of the above steps, and these steps may be executed in other orders. Moreover, at least some steps in the processes involved in the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but may be executed at different times. The execution order of these steps or stages is not necessarily sequential, but may be performed alternately or in turn with other steps or at least some of the steps or stages in other steps.

[0030] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or basic characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects. The scope of the invention is defined by the appended claims rather than the foregoing description. Therefore, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention, and no reference numerals in the claims should be construed as limiting the scope of the claims.

[0031] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment includes only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for constructing residential electricity consumption profiles for smart meters, characterized in that, Includes the following steps: S1. Data Acquisition: Collect smart meter data, meteorological data, socio-economic data, and smart home data; perform data cleaning, standardization, and time series alignment on the collected data. S2. Constructing a contextual coding system: Construct a contextual coding system based on three dimensions: time, environment, and society. Use a clustering algorithm to divide the temperature data into n clusters, with each cluster corresponding to a coding interval. Then, use a decision tree algorithm to determine the optimal boundary point of the population structure. S3. Construct a contextual electricity consumption correlation matrix: Calculate the average electricity consumption for each group based on time context, environmental context, population structure, and occupation type. Calculate the context-sensitive electricity consumption rate using the average electricity consumption of the baseline context as the baseline value. S4. Causal Inference Model Construction: Select key features as nodes in the causal graph, use the NOTEARS algorithm to construct a structural causal model, optimize the objective function, and calculate the average treatment effect to quantify the causal effect of key factors. S5. Counterfactual Electricity Consumption Scenario Generation: Determine the counterfactual target, and based on the causal model, set the target variable as the counterfactual value to generate counterfactual samples. Use the random forest regression model to predict the electricity consumption under the counterfactual scenario to generate counterfactual electricity consumption scenarios. For the same target variable, generate counterfactual scenarios of different intensities. Then, through the implicit counterfactual learning method, generate diverse counterfactual samples to avoid biases introduced by explicitly modifying the text structure or attributes. S6. Determining the optimal energy-saving path: Calculate the energy-saving potential of each counterfactual scenario, quantify the causal effects of each factor, calculate the comprehensive benefits, determine the optimal path through multi-objective optimization, while considering the implementation difficulty and residents' acceptance, and select the top n paths with the highest benefits as recommended paths according to the comprehensive benefits ranking. S7. Dynamic profile generation: Based on context-sensitive electricity consumption rate, causal effect and counterfactual electricity consumption scenarios, construct dynamic residential electricity consumption profiles.

2. The method for constructing residential electricity consumption profiles for smart meters according to claim 1, characterized in that, The smart meter data includes active energy, reactive energy, voltage, and current; the meteorological data includes temperature, humidity, and light intensity; the socioeconomic data includes household population and occupation type; and the smart home data includes device power consumption.

3. The method for constructing residential electricity consumption profiles for smart meters according to claim 1, characterized in that, The construction of the contextual electricity consumption correlation matrix includes the following steps: S301. Calculate the average electricity consumption for each group based on time context, environmental context, population structure, and occupation type. S302. Using the average electricity consumption under the baseline scenario as the baseline value, calculate the scenario-sensitive electricity consumption rate by dividing the difference between the electricity consumption and the baseline value by the baseline value.

4. The method for constructing residential electricity consumption profiles for smart meters according to claim 1, characterized in that, The causal mapping construction includes the following steps: S401. Select key features as nodes in the cause-effect graph, including time context, environmental context, population structure, occupation type, and smart socket installation status; S402. Use the NOTEARS algorithm to construct a structural causal model and determine causal relationships by optimizing the objective function; S403. Use a meta-learner to quantify causal effects by comparing the difference in power consumption between implementing and not implementing smart sockets, and calculate the average impact.

5. The method for constructing residential electricity consumption profiles for smart meters according to claim 1, characterized in that, The generation of the counterfactual electricity consumption scenario includes the following steps: S501. Identify counterfactual objectives; S502. Generate counterfactual samples: Copy the original dataset and set the target variable to a counterfactual value; S503. Generate counterfactual electricity consumption scenarios: Use a random forest regression model to predict electricity consumption under counterfactual scenarios; S504. Generate multiple sets of counterfactual scenarios: Generate counterfactual scenarios of different intensities for the same target variable; S505. Generate diverse counterfactual samples through implicit counterfactual learning methods.

6. The method for constructing residential electricity consumption profiles for smart meters according to claim 1, characterized in that, The determination of the optimal energy-saving path includes the following steps: S601. Calculate the energy-saving potential of each counterfactual scenario: actual electricity consumption minus counterfactual electricity consumption; S602. Quantify the causal effects of each factor: Calculate the average impact by comparing the difference in electricity consumption between implementing and not implementing the factor. S603. Calculate implementation costs: including equipment costs, installation costs, and usage costs; S604. Calculate the overall benefits: Energy saving potential divided by implementation costs; S605. Determine the optimal path through multi-objective optimization: Select the path with the highest overall benefits based on considerations of energy-saving benefits, implementation difficulty, and residents' acceptance. S606. Generate the optimal energy-saving path: Based on the overall benefits, select the top 3 paths with the highest benefits as recommended paths.

7. The method for constructing residential electricity consumption profiles for smart meters according to claim 1, characterized in that, The construction of the dynamic residential electricity consumption profile includes the following steps: S701. Calculate the stability of electricity consumption behavior: Assess the stability of electricity consumption behavior by calculating the standard deviation of electricity consumption. S702. Calculate the adaptive weight: the reciprocal of the standard deviation of electricity consumption, plus a constant; S703, Normalized Adaptive Weights: Adjust the adaptive weights to the range of 0 to 1; S704. Predicting future scenarios based on historical data: Using a random forest classifier to predict scenario encoding for future dates; S705. Generate a dynamic electricity consumption profile: Combine adaptive weights, context-sensitive electricity consumption rate, and causal effects to generate a dynamic electricity consumption profile.

8. The method for constructing residential electricity consumption profiles for smart meters according to claim 1, characterized in that, The data cleaning process uses linear interpolation to fill in missing values ​​and then uses the IQR method to identify and process outliers.

Images