Cloud sharing method and system for electricity marketing measurement data

Abstract

The invention discloses an electric power marketing measurement data cloud sharing method and system, and relates to the technical field of electric power data sharing. The power marketing measurement data cloud sharing method comprises the following steps: acquiring power measurement operation time sequence data and power utilization response time sequence data of a plurality of transformer areas in a set area, and uploading the data to a set cloud for storage; based on the electric power metering operation time sequence data of each transformer area in the set area, analyzing a stable efficiency evolution characteristic value of the corresponding transformer area; and based on a pre-trained electricity utilization response evolution mapping model, in combination with the electricity utilization response time sequence data and the stable efficiency evolution characteristic value of each transformer area, analyzing the marketing coordination characteristic value of the corresponding transformer area. Therefore, grouping and strategy sharing are carried out on each transformer area, evolution identification and decision-making coordination of regional marketing data are realized at the cloud level, and the precision of marketing decision-making is obviously enhanced.

Description

Technical Field

[0001] This invention relates to the field of power data sharing technology, specifically to a cloud-based method and system for sharing power marketing metering data. Background Technology

[0002] With the digitalization and intelligentization of electricity marketing, the number of electricity metering devices in the distribution network is increasing rapidly. Electricity metering data and user electricity consumption behavior data at the distribution area level are characterized by high frequency and multi-dimensionality. The existing data structure is scattered and has weak correlation, making it difficult to achieve cross-distribution sharing and dynamic analysis.

[0003] Currently, although power companies have built various metering and marketing analysis systems, most of them are still in the stage of one-way data reporting and static statistics. This can easily lead to information fragmentation in various aspects of marketing management. Especially in scenarios with load fluctuations, power anomalies, or sudden changes in user behavior, existing technologies are unable to identify operational differences between different power stations in a timely manner, resulting in delays in marketing data sharing or decision-making biases.

[0004] Existing technology, such as the patent application with publication number CN119475368A, discloses a cloud-based method for sharing electricity marketing metering data based on blockchain. This method relates to the field of electricity data sharing technology and includes a user terminal, an automatic electricity marketing metering data collection system, a blockchain system, and a data storage system. Multiple user terminals can be configured. The blockchain system includes an encryption algorithm, a blockchain network, blockchain nodes, a privacy data shielding module, and an alarm module. The user terminals are connected to the automatic electricity marketing metering data collection system, which is connected to the blockchain system. The blockchain system is connected to the data storage system. This blockchain-based cloud-based method for sharing electricity marketing metering data means that electricity marketing metering data provided by different user terminals resides on different blockchain nodes. Therefore, even if one blockchain node is damaged or attacked, it will not affect other blockchain nodes, thus ensuring the stability of the entire system.

[0005] Based on the above findings, the limitations of existing technologies include at least the following problems: Existing technologies lack evolutionary analysis of data time-series changes, which weakens the correlation between marketing data, makes it difficult to form a unified feature reflecting the overall operational evolution, and lacks deep structural information for decision-making to reflect the intrinsic relationship between the operating status of power distribution areas and user response behavior. As a result, it is difficult to reflect the dynamic changes in the operating characteristics of each power distribution area and user response behavior, which can easily lead to a lack of coordination at the marketing level, and thus cause problems such as disconnect in marketing decisions, thereby affecting the level of intelligent collaboration in the cloud environment of power marketing. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a cloud-based method and system for sharing electricity marketing metering data, which solves the problem that the lack of time-series evolution analysis in existing technologies leads to weak correlation of marketing data and easily causes imbalances in marketing decisions.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a cloud-based method for sharing electricity marketing metering data, comprising the following steps: acquiring electricity metering operation sequence data and electricity consumption response time sequence data of several transformer substations within a designated area, and uploading them to a designated cloud for storage; analyzing the stability and efficiency evolution characteristic values ​​of each transformer substation based on its electricity metering operation sequence data within the designated area; and analyzing the marketing coordination characteristic values ​​of each transformer substation based on a pre-trained electricity consumption response evolution mapping model, combined with the electricity consumption response time sequence data and stability and efficiency evolution characteristic values ​​of each transformer substation within the designated area; and performing marketing sharing processing on each transformer substation within the designated area based on the marketing coordination characteristic values.

[0008] Furthermore, the power metering operation sequence data includes voltage fluctuation rate, distribution area line loss rate, energy flow coordination index, energy feedback rate, voltage distortion index, and current distribution offset factor at each time point. The specific steps for analyzing the steady-state efficiency evolution characteristic value of each distribution area within the set area are as follows: Based on the power metering operation sequence data of each distribution area within the set area, analyze the operation sequence characteristic set of its corresponding distribution area, including the steady-state efficiency characteristic value and load balance characteristic value at each time point; based on the operation sequence characteristic set of each distribution area within the set area, analyze the steady-state efficiency evolution characteristic value of its corresponding distribution area.

[0009] Furthermore, the specific steps for analyzing the operational sequence characteristic set of each transformer substation within the designated area are as follows: Based on the voltage fluctuation rate, line loss rate, energy flow coordination index, and energy feedback rate of each transformer substation within the designated area at each time point, analyze the steady-state performance characteristic value at the corresponding time point; Based on the voltage distortion index and current distribution offset factor of each transformer substation within the designated area at each time point, analyze the load balance characteristic value at the corresponding time point.

[0010] Furthermore, the electricity response time-series data includes the electricity expenditure rate, electricity load adaptation value, electricity revenue rate value, and electricity structure activity value at each time point. The electricity response evolution mapping model includes an input layer, a time-series evolution layer, a correlation and fusion layer, and an output layer.

[0011] Furthermore, the specific steps for analyzing the marketing coordination characteristic values ​​of each transformer substation within the designated area are as follows: Based on the pre-trained electricity response evolution mapping model and combined with the electricity response time series data of each transformer substation within the designated area, analyze the marketing evolution adaptation characteristic values ​​of the corresponding transformer substation; Based on the marketing evolution adaptation characteristic values ​​and stable efficiency evolution characteristic values ​​of each transformer substation within the designated area, analyze the marketing coordination characteristic values ​​of the corresponding transformer substation.

[0012] Further, the specific steps for analyzing the marketing evolution adaptation feature values ​​of each transformer substation within the designated area are as follows: input the electricity consumption response time series data of each transformer substation within the designated area into the pre-trained electricity consumption response evolution mapping model, and analyze the electricity consumption evolution response feature set of the corresponding transformer substation, including revenue and expenditure evolution coordination feature values, steady-state feedback feature values, and active coordination response feature values; based on the electricity consumption evolution response feature set of each transformer substation within the designated area, analyze the marketing evolution adaptation feature values ​​of the corresponding transformer substation.

[0013] Furthermore, the specific steps for analyzing the power consumption evolution response feature set of each transformer substation within the designated area are as follows: In the input layer of the power consumption response evolution mapping model, the power consumption response time series data of each transformer substation within the designated area is received and preprocessed; in the time series evolution layer of the power consumption response evolution mapping model, based on the preprocessed power consumption response time series data of each transformer substation within the designated area, the time series response feature vector of the corresponding transformer substation is extracted; in the association and fusion layer of the power consumption response evolution mapping model, based on the time series response feature vector of each transformer substation within the designated area, the power consumption mapping feature vector of the corresponding transformer substation is extracted; in the output layer of the power consumption response evolution mapping model, based on the power consumption mapping feature vector of each transformer substation within the designated area, the power consumption evolution response feature set of the corresponding transformer substation is output.

[0014] Furthermore, the specific formula for calculating the marketing coordination characteristic value of a certain area within the designated region is as follows: ;in, To set the marketing coordination characteristic value for a specific area within a region, To define the stable evolution characteristic value of a certain transformer area within a given region, These are the steady-state evolution adjustment coefficients stored in the database. To set the marketing evolution adaptation characteristic value for a specific area within a given region, The adaptation adjustment coefficients are stored in the database. , The difference adjustment coefficients are stored in the database. These are the coordination coefficients stored in the database.

[0015] Furthermore, the specific steps for marketing sharing processing of each station area within a set area based on marketing coordination characteristic values ​​are as follows: based on the marketing coordination characteristic values ​​of each station area within a set area, the station area is divided into several sharing groups within the set area; and corresponding marketing service strategies are adopted for each sharing group within the set area.

[0016] A cloud-based electricity marketing metering data sharing system includes: a data acquisition and storage module for acquiring electricity metering operation sequence data and electricity consumption response time sequence data of several transformer substations within a designated area, and uploading them to a designated cloud for storage; a stability and efficiency characteristic analysis module for analyzing the stability and efficiency evolution characteristic values ​​of each transformer substation based on the electricity metering operation sequence data of each transformer substation within the designated area; a marketing response mapping module for analyzing the marketing coordination characteristic values ​​of each transformer substation based on a pre-trained electricity consumption response evolution mapping model, combined with the electricity consumption response time sequence data and stability and efficiency evolution characteristic values ​​of each transformer substation within the designated area; and a marketing sharing feedback module for performing marketing sharing processing on each transformer substation within the designated area based on the marketing coordination characteristic values.

[0017] The present invention has the following beneficial effects: (1) The cloud-based method for sharing electricity marketing metering data integrates electricity metering operation sequence data and electricity consumption response sequence data in the cloud, thereby forming marketing evolution adaptation feature values ​​that reflect the overall operation trend of each distribution area. This achieves dynamic matching of time sequence features between different distribution areas, enabling marketing data to have continuous logical correlation in the time dimension. Based on the marketing coordination feature values, each distribution area is grouped and strategies are shared, thereby realizing the evolution identification and decision coordination of regional marketing data at the cloud level, which significantly enhances the accuracy of marketing decisions.

[0018] (2) The cloud-based method for sharing electricity marketing metering data introduces a pre-trained electricity response evolution mapping model to perform multi-layer structured analysis on the electricity response time series data of the distribution area. This allows for the extraction of deep features reflecting the intrinsic relationship between user behavior and operating status from the continuous electricity response time series data, thereby achieving accurate extraction of marketing evolution adaptation feature values. These features are then integrated with the stable efficiency evolution feature values ​​to form marketing coordination feature values, which can then identify dynamic adaptation differences between different distribution areas, realize closed-loop linkage between the data layer and the decision layer, and significantly enhance the intelligent collaborative capability of cloud-based sharing of electricity marketing metering data.

[0019] (3) The cloud-based method for sharing electricity marketing metering data achieves hierarchical sharing and strategy adaptation in the cloud environment by grouping each distribution area according to the marketing coordination characteristic value. It automatically divides the sharing groups according to the marketing coordination characteristic value of different distribution areas and matches differentiated marketing service strategies for each group, so that data sharing changes from static transmission to dynamic optimization process. This enables the cloud to adjust the sharing strategy, ensure execution efficiency, and continuously improve the processing accuracy and response speed of marketing data, thereby promoting the flexibility of electricity marketing management.

[0020] (4) The cloud-based electricity marketing metering data sharing system achieves integrated evaluation of data from various distribution areas through collaborative analysis between modules. The data acquisition and storage module can collect and uniformly store the electricity metering operation sequence data and electricity consumption response time sequence data of multiple distribution areas within a set area in real time. The stability and efficiency feature analysis module analyzes the operation sequence data of the distribution areas and extracts the stability and efficiency evolution feature values. The marketing response mapping module generates marketing coordination feature values ​​based on the pre-trained electricity consumption response evolution mapping model and the stability and efficiency evolution feature values. The marketing sharing feedback module performs cloud grouping and strategy adjustment according to the analysis results, realizing two-way linkage between data sharing and service response, thereby effectively improving the operational efficiency of electricity marketing business.

[0021] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0022] Figure 1 This is a flowchart of a cloud-based method for sharing electricity marketing metering data according to the present invention.

[0023] Figure 2 This is a flowchart illustrating the specific steps involved in analyzing the marketing coordination characteristic values ​​of each transformer substation within a designated area in a cloud-based method for sharing electricity marketing metering data, as described in this invention.

[0024] Figure 3 This is a schematic diagram of the power consumption evolution response feature set data of a transformer substation sequence within a set area in the cloud-based power marketing metering data sharing method of the present invention.

[0025] Figure 4 This is a block diagram of a cloud-based electricity marketing metering data sharing system according to the present invention. Detailed Implementation

[0026] Please see Figure 1This invention provides a technical solution: a method for sharing electricity marketing metering data in the cloud, comprising the following steps: within a set period (e.g., 24 hours as a period, with a time step of 5-15 minutes, i.e., a point in time), acquiring the electricity metering operation sequence data and electricity consumption response time sequence data of several transformer substations (i.e., local power supply units powered by the same distribution transformer and containing several end users) within a set area, and uploading them to a set cloud for storage; in the set cloud, analyzing the stability and efficiency evolution characteristic value of the corresponding transformer substation based on the electricity metering operation sequence data of each transformer substation within the set area; and based on a pre-trained electricity consumption response evolution mapping model, combining the electricity consumption response time sequence data and stability and efficiency evolution characteristic value of each transformer substation within the set area, analyzing the marketing coordination characteristic value of the corresponding transformer substation; and performing marketing sharing processing on each transformer substation within the set area based on the marketing coordination characteristic value.

[0027] The specific formula for calculating the marketing coordination characteristic value of a certain area within a defined region is as follows: ;in, To set the marketing coordination characteristic value for a specific area within a region, To define the stable evolution characteristic value of a certain transformer area within a given region, These are the steady-state evolution adjustment coefficients stored in the database. To set the marketing evolution adaptation characteristic value for a specific area within a given region, The adaptation adjustment coefficients are stored in the database. , The difference adjustment coefficients are stored in the database. These are the coordinated adjustment coefficients stored in the database, and in this embodiment, the stable evolution adjustment coefficients stored in the database are... Adaptive adjustment coefficient Difference adjustment coefficient Coordination coefficient The values ​​were 0.600, 0.400, 0.450, and 0.300, respectively.

[0028] The specific steps for marketing sharing processing of each transformer substation within a designated area based on marketing coordination feature values ​​are as follows: Based on the marketing coordination feature values ​​of each transformer substation within a designated area, the area is divided into several sharing groups. Specifically, the marketing coordination feature values ​​of each transformer substation within the designated area are sorted and arranged, and the transformer substations are divided into different sharing groups according to a preset quantile ratio (e.g., top 20%, middle 60%, bottom 20%). Corresponding marketing service strategies are adopted for each sharing group within the designated area, including but not limited to the following marketing service strategies: If the sharing group is in the top 20%, a stability-maintaining marketing service strategy is pushed to the transformer substations in the group on the cloud to maintain the current time-of-use electricity price structure and keep the marketing revenue and user response highly stable. If the shared group is in the middle 60%, a balanced marketing service strategy will be implemented to optimize the electricity price guidance pace and user incentive methods for this group. For example, a personalized goal of reducing electricity consumption by 5%-10% from the baseline will be set, and points (non-cash) will be awarded upon achievement. These points can be redeemed for small discount coupons on electricity bills to improve overall marketing response efficiency. If the shared group is in the bottom 20%, an incentive-based marketing service strategy will be implemented. For example, targeted price incentives will be implemented. For instance, if the user actively reduces their load to below 80% of the preset benchmark value during weekday evening peak hours (e.g., 18:00-20:00) based on their existing electricity price, the electricity saved during this period can be settled based on a preset high subsidy to promote improved marketing response and revenue for this group.

[0029] Specifically, the power metering operation sequence data includes voltage fluctuation rate, distribution area line loss rate, energy flow coordination index, energy feedback rate, voltage distortion index, and current distribution offset factor at each time point. The specific steps for analyzing the steady-state efficiency evolution characteristic value of each distribution area within the set area are as follows: Based on the power metering operation sequence data of each distribution area within the set area, analyze the corresponding operation sequence characteristic set of the distribution area, including the steady-state efficiency characteristic value and load balance characteristic value at each time point; Based on the operation sequence characteristic set of each distribution area within the set area, analyze the steady-state efficiency evolution characteristic value of the corresponding distribution area, specifically: perform weighted processing on the steady-state efficiency characteristic value and load balance characteristic value of each distribution area at each time point within the set area, and perform moving average processing based on the weighted processing result at each time point to obtain the steady-state efficiency evolution characteristic value of the corresponding distribution area (used to characterize the overall operational stability of the power transmission process of the distribution area within the set period).

[0030] The voltage fluctuation rate is the voltage change amplitude of the distribution area at the current time point. It can be obtained by the voltage transformer set on the outgoing line side of the distribution transformer in the distribution area, which can obtain the three-phase voltage value at that time point and extract the three-phase voltage average value. It reads the rated voltage value stored in the database and performs comprehensive processing, namely |three-phase voltage average value - rated voltage value| / rated voltage value, and uses the result as the voltage fluctuation rate value.

[0031] The transformer substation line loss rate is the proportion of power supply energy loss in the transformer substation at the current time point. It can be obtained by setting up energy metering terminals (including current transformers and voltage transformers) on the incoming and outgoing sides of the transformer substation. These terminals can measure the three-phase voltage and current values ​​in real time at each time point and obtain the corresponding power values. Taking the input power value as an example, based on the three-phase voltage and current values ​​on the incoming side, the power values ​​of the corresponding phases are analyzed and the average value is taken to obtain the input power value. The input power value and output power value at the current time point are obtained and then comprehensively processed, i.e., (input power value - output power value) / input power value. The result is used as the transformer substation line loss rate value.

[0032] The energy flow coordination index is the degree of coordination of the three-phase power flow in a transformer substation at the current time point. It can be obtained by setting up an energy metering terminal at the outgoing line of the transformer substation to obtain the three-phase power value at that time point, extracting the maximum power value, minimum power value, and average power value respectively, and performing comprehensive processing, i.e., 1-[(maximum power value - minimum power value) / average power value], and using the result as the energy flow coordination index.

[0033] The energy feedback rate is the proportion of distributed energy in the distribution area that feeds back to the grid at the current time. The main incoming line of the distribution area is equipped with a smart energy meter with bidirectional power measurement function. At each time point, it synchronously acquires the power value flowing into the grid and the feedback power value fed back from the user side, and performs ratio processing, that is, feedback power value / power value flowing into the grid, and uses the result as the energy feedback rate.

[0034] The voltage distortion index is the degree of voltage waveform distortion in a transformer substation at the current time point. It can be obtained by setting voltage sensors at the outgoing line terminals of the transformer substation to continuously collect the three-phase voltage signals at each time point (such as the time interval from the previous time point to the current time point), and performing Fast Fourier Transform (FFT) processing on each signal to extract the amplitude of the fundamental component and the amplitude of each harmonic component of the corresponding phase. Then, a comprehensive processing is performed, that is, the square of the ratio of the amplitude of each harmonic component to the amplitude of the fundamental component of each phase is accumulated, the square root of the accumulated result is taken, and the mean value is taken as the voltage distortion index (it should be noted that the voltage distortion index of the first time point can be set to 0).

[0035] The current distribution offset factor is the degree of difference in load distribution among the branches of a transformer substation at the current time. It can be obtained by acquiring the current values ​​of all branches at the current time through the current transformers of each outgoing branch of the transformer substation, extracting the current standard deviation and the current average value of each branch, and performing a ratio processing, i.e., current standard deviation / current average value, and using the result as the current distribution offset factor.

[0036] The specific steps for analyzing the operational sequence feature set of each transformer substation within the designated area are as follows: Based on the voltage fluctuation rate, line loss rate, energy flow coordination index, and energy feedback rate of each transformer substation within the designated area at each time point, analyze its steady-state performance characteristic value at the corresponding time point. Specifically, standardize the voltage fluctuation rate, line loss rate, energy flow coordination index, and energy feedback rate of each transformer substation within the designated area at each time point, and perform weighted processing based on the standardization results to obtain its steady-state performance characteristic value at the corresponding time point (used to characterize the steady-state energy transmission efficiency of the transformer substation at that time point). In the weighted processing, the standardized voltage fluctuation rate and line loss rate are inverted. Taking the voltage fluctuation rate as an example, its form is: 1 / (1+standardized voltage fluctuation rate value). Based on the voltage distortion index and current distribution offset factor of each transformer substation within the set area at each time point, the load balance characteristic value of the corresponding time point is analyzed. Specifically, the voltage distortion index and current distribution offset factor of each transformer substation within the set area at each time point are standardized, and the standardized result is weighted and then inverted. The weighted result is expressed as 1 / (1+the weighted result) to obtain the load balance characteristic value of the corresponding time point (used to characterize the dynamic operating stability of the transformer substation at that time point).

[0037] In this implementation plan, by refining and analyzing the power metering operation sequence data of each distribution area in the cloud, the stability and efficiency evolution characteristic values ​​are made more consistent with the actual operating status of the distribution area. Secondly, feature values ​​representing steady-state operating efficiency and load balance are extracted from the power metering operation sequence data, and then weighted and moving average processing is used to achieve a smooth transition from instantaneous operating characteristics to periodic stable characteristics. This effectively filters out interference from short-term fluctuations or sudden anomalies, making the results more continuous. At the same time, standardization processing puts data from different sources on the same scale, thereby ensuring the objectivity of the analysis results. Finally, an accurate representation of the operating trend of the distribution area can be formed in the cloud, thereby improving the coordination and response speed of the power marketing process.

[0038] Specifically, the electricity response time series data includes the electricity expenditure rate, electricity load adaptation value, electricity revenue rate value, and electricity structure activity value at each time point. The electricity response evolution mapping model includes an input layer, a time series evolution layer, a correlation and fusion layer, and an output layer.

[0039] The electricity expenditure rate is the change in electricity costs for a distribution area at the current time point. It is used to characterize the trend of economic expenditure changes in a distribution area within a set period and to reflect the demand-side economic expenditure response characteristics. The electricity consumption value at the current time point can be obtained by a smart energy meter set at the main metering node on the outgoing line side of the distribution transformer in the distribution area. The time-of-use price identifier (i.e., the electricity price period category corresponding to the current time point, such as the price parameters for peak, flat, valley, or peak periods) stored in the database is read. The difference between the electricity consumption value at the current time point and the electricity consumption value at the previous time point is processed, and the result is multiplied by the time-of-use price identifier to calculate the electricity expenditure rate value at the current time point (it should be noted that the electricity expenditure rate value at the first time point can be set to 0).

[0040] The load matching value is the degree of economic matching between the load structure of a transformer substation at the current time and the current time-of-use pricing strategy. It is used to characterize the rationality and economic coordination of the load operation of the substation in different time periods. The instantaneous power values ​​of various loads (including industrial loads, residential loads, lighting loads, energy storage loads, etc.) at the current time can be obtained by smart energy meters and load identification sensors installed on each outgoing branch of the substation. The time-of-use pricing identifiers stored in the database are read to determine the time period category (such as peak, flat, valley or peak period) corresponding to the current time. Based on the time-of-use pricing identifiers, the target power ratio of the corresponding time period is called from the preset target power weight table of electricity price and load type. The deviation between the actual power ratio and the target power ratio of various loads is calculated, and all deviation results are normalized and reversed. The result is used as the load matching value at the current time.

[0041] The electricity revenue rate is the rate of change of electricity sales revenue in a set distribution area per unit time at the current time point. It is used to characterize the economic response strength of the electricity sales revenue in the distribution area within a set period, reflecting the economic revenue response characteristics of the supply side. The electricity sales power value at the current time point can be obtained by a smart energy meter installed on the outgoing line side of the distribution transformer in the distribution area (the smart energy meter can collect the three-phase voltage and current values ​​at the current time point, calculate the three-phase active power through power calculation, and take the average value as the electricity sales power value at that time point). At the same time, the time-of-use price identifier stored in the database is read to determine the electricity price period category (such as peak, flat, valley, or peak period) corresponding to the current time point, and the electricity price parameters of the corresponding period are extracted. The electricity sales power value at the current time point and the previous time point are differentially processed to obtain the change in electricity sales power at the current time step. This change is multiplied by the corresponding electricity price parameter and then divided by the time interval between two adjacent time points to obtain the electricity revenue rate value at the current time point (it should be noted that the electricity revenue rate value at the first time point can be set to 0).

[0042] The electricity consumption structure activity value represents the level of user electricity consumption activity in a designated transformer area at the current time point. It is used to characterize the user-side electricity participation level in the transformer area within a set period. It can be obtained by acquiring the power consumption value of each user through smart meters installed on each user side of the transformer area, counting the number of users with power consumption values ​​greater than zero, and comparing them with the total number of users in the transformer area stored in the database to obtain the user participation ratio. The power consumption values ​​of each user are summed to obtain the total power consumption value. The power consumption value of each user is then compared with the total power consumption value, and the results are weighted to obtain the electricity consumption activity value. Finally, the user participation ratio is weighted to obtain the electricity consumption structure activity value.

[0043] The specific steps for analyzing the marketing coordination characteristic values ​​of each transformer substation within the designated area are as follows: Based on the pre-trained electricity response evolution mapping model and combined with the electricity response time series data of each transformer substation within the designated area, analyze the marketing evolution adaptation characteristic values ​​of the corresponding transformer substation; Based on the marketing evolution adaptation characteristic values ​​and stable efficiency evolution characteristic values ​​of each transformer substation within the designated area, analyze the marketing coordination characteristic values ​​of the corresponding transformer substation (used to characterize the overall coordination degree between the changes in marketing effectiveness and the evolution of operating status of the transformer substation within the designated period).

[0044] like Figure 2As shown, the specific steps for analyzing the marketing evolution adaptation feature value of each transformer substation within the set area are as follows: Input the electricity consumption response time series data of each transformer substation within the set area into the pre-trained electricity consumption response evolution mapping model, and analyze the electricity consumption evolution response feature set of the corresponding transformer substation, including revenue and expenditure evolution coordination feature value, steady-state feedback feature value, and active coordination response feature value; Based on the electricity consumption evolution response feature set of each transformer substation within the set area, analyze the marketing evolution adaptation feature value of the corresponding transformer substation (used to characterize the adaptive change ability of the transformer substation's electricity marketing effect within the set period, specifically reflecting the linkage relationship between the transformer substation's electricity expenditure, electricity sales revenue, and user-side electricity participation behavior; when this feature value is high, it indicates that the transformer substation's marketing behavior has formed a stable adaptation pattern between revenue changes and user responses, that is, the trends of expenditure, revenue, and load activity changes are consistent, and it has good marketing coordination).

[0045] The specific formula for calculating the marketing evolution adaptation characteristic value of a certain station area within the set area is as follows: ;in, To set the marketing evolution adaptation characteristic value for a specific area within a given region, To define the collaborative characteristic value of the revenue and expenditure evolution of a certain transformer substation within a given region, These are the income and expenditure evolution adjustment coefficients stored in the database. To define the steady-state feedback characteristic value of a certain transformer area within the set region, These are the steady-state feedback adjustment coefficients stored in the database. To define the active collaborative response characteristic value of a specific transformer station within a given area, These are the active collaborative adjustment coefficients stored in the database. The asynchronous adjustment coefficients stored in the database are used to adjust the degree of influence of the difference between the revenue and expenditure evolution synergy characteristic value and the steady-state feedback characteristic value on the marketing evolution adaptation characteristic value. Furthermore, in this implementation example, the revenue and expenditure evolution adjustment coefficients stored in the database... Steady-state feedback adjustment coefficient Active Coordination Coefficient Asynchronous adjustment coefficient The values ​​were 0.410, 0.340, 0.250, and 0.550, respectively.

[0046] The following is a specific implementation example for calculating the marketing evolution adaptation characteristic value of a certain station area within a set area. Existing data includes the revenue and expenditure evolution synergy characteristic values, steady-state feedback characteristic values, and active synergy response characteristic values ​​of three stations within the set area, as detailed in Table 1 and... Figure 3 As shown: Table 1. Example of power consumption evolution response feature set data for transformer substations within a specified area. Co-evolutionary eigenvalues ​​of income and expenditure Steady-state feedback eigenvalues Active Cooperative Response Feature Values Taiwan District 1 0.734 0.812 0.683 Taiwan Area 2 0.643 0.567 0.781 Taiwan District 3 0.821 0.627 0.758 Income and expenditure evolution adjustment coefficients stored in the database The value is: 0.410; Steady-state feedback adjustment coefficients stored in the database The value is: 0.340; Active co-regulation coefficients stored in the database The value is: 0.250; Asynchronous adjustment coefficients stored in the database The value is: 0.550; Substituting the data from Table 1 and the aforementioned coefficients into the specific formula for calculating the marketing evolution adaptation characteristic value of a certain area within the designated region, we obtain: The marketing evolution adaptation characteristic value of the first station in the set area is (0.410×0.734+0.340×0.812+0.250×0.683)×exp(-0.550×|0.734-0.812|)≈0.716; The marketing evolution adaptation characteristic value of the second station area within the set region is calculated as follows: (0.410×0.643+0.340×0.567+0.250×0.781)×exp(-0.550×|0.643-0.567|)≈0.625; The marketing evolution adaptation characteristic value of the third station in the set area = (0.410×0.821+0.340×0.627+0.250×0.758)×exp(-0.550×|0.821-0.627|)≈0.664.

[0047] In this implementation plan, a pre-trained electricity consumption response evolution mapping model is introduced to perform in-depth analysis of electricity consumption response time series data to extract the electricity consumption evolution response characteristics of the corresponding transformer areas. This allows for accurate representation of the electricity consumption behavior trends of transformer areas within a set period in the cloud, thereby reflecting the effectiveness of marketing strategies. Secondly, marketing evolution adaptation feature values ​​are generated based on the electricity consumption evolution response characteristics and fused with stable efficiency evolution feature values ​​to generate marketing coordination feature values. This enables an overall matching measurement between the operating status and marketing effectiveness of each transformer area, forming a key decision-making basis that can be used for group sharing and strategy adjustment. Finally, based on the marketing coordination feature values, differences in marketing coordination among different transformer areas can be dynamically identified to improve the synergy of electricity marketing in the cloud environment.

[0048] Specifically, the specific steps for analyzing the power consumption evolution response feature set of each transformer substation within the designated area are as follows: In the input layer of the power consumption response evolution mapping model, the power consumption response time series data of each transformer substation within the designated area is received and preprocessed. Specifically, the power consumption response time series data of each transformer substation (including the power consumption expenditure rate value, power load adaptation value, power consumption revenue rate value, and power consumption structure activity value at each time point) are time-aligned according to the timestamp to ensure that the samples corresponding to different parameters at the same time point are consistent. The value range of each power consumption response time series data is normalized to eliminate the numerical differences between different units. The normalization can adopt the minimum-maximum standardization method to map all input parameters to the [0, 1] interval. In the temporal evolution layer of the electricity response evolution mapping model, based on the preprocessed electricity response time-series data of each transformer substation within a defined region, the temporal response feature vector of the corresponding substation is extracted. Specifically, this layer is an LSTM. The LSTM is used to learn the dynamic dependencies between time steps in the input time-series samples to capture the long-term trend and short-term fluctuation characteristics of the transformer substation's electricity response data. In the LSTM, through a recurrent structure of input gate, forget gate, and output gate, the input features of each time step are updated and information is filtered: the input gate is used to control the degree of influence of the current time step features on the memory unit; the forget gate is used to adjust the retention ratio of historical information to prevent the loss of old information in long sequences; the output gate is used to determine the feature output intensity of the time step. The LSTM updates its hidden state and cell state at each time step, thereby retaining the contextual information of previous time steps in the time dimension, and generating new hidden states in combination with the current input features to generate temporal response feature vectors that can reflect these time changes, such as: For each time point, the change in the economic expenditure rate of electricity at adjacent time points is extracted, such as |the difference between the electricity expenditure rate at the first time point and the second time point| / the electricity expenditure rate at the second time point. Autocorrelation processing is then performed on the change in the economic expenditure rate of electricity at each group of adjacent time points. This involves extracting the autocorrelation coefficient of the expenditure rate change values ​​at different lag steps based on multiple set lag steps (e.g., 1, 2, 3 time steps), and performing exponential fitting to extract expenditure response characteristics. These characteristics characterize the temporal persistence of changes in the economic expenditure of electricity in the distribution area. A smaller autocorrelation decay rate indicates that the distribution area maintains its original expenditure change pattern after electricity price or load disturbances, exhibiting strong response inertia. Conversely, a larger autocorrelation decay rate indicates that the distribution area can quickly adjust its economic expenditure structure to respond to external changes, exhibiting weaker response inertia. Finally, the standard deviation of the expenditure rate is extracted based on the electricity expenditure rate value at each time point. For each time point, the load adaptation value is processed by subtracting the load adaptation values ​​of adjacent time points (absolute value is taken), and the economic adaptation deviation variance value is extracted based on the result. The mean value is then taken and inverted to extract the adaptation stability feature, which is used to characterize the stability of the load structure economic adaptation change of the transformer area within a set period. For each time point, the electricity revenue rate value is processed by linear fitting. That is, a revenue change sequence is constructed based on the electricity revenue rate values ​​of all time points, and a first linear regression is performed using the least squares method. The slope obtained by fitting is used as the revenue evolution feature to characterize the trend of economic revenue change within a set period. Based on the electricity revenue rate value at each time point, the maximum, minimum, and average electricity revenue rates are extracted and then comprehensively processed into (maximum electricity revenue rate - minimum electricity revenue rate) / average electricity revenue rate to extract revenue fluctuation characteristics, which are used to characterize the relative fluctuation intensity of economic revenue changes in the distribution area within a set period. For the electricity structure activity value at each time point, the electricity structure activity change values ​​of adjacent time points are extracted and the results are processed using root mean square (RMS) to extract activity diffusion characteristics, which are used to characterize the overall fluctuation rate of user group activity in the distribution area. Based on the electricity consumption activity value at each time point, the mean and standard deviation of electricity consumption activity are extracted. An activity range is then defined using these values: the sum of the mean and standard deviation of electricity consumption activity is the upper limit, and the minus the lower limit is the lower limit. The total number of time points below the lower limit (low activity duration), the total number of time points within the activity range (medium activity duration), and the total number of time points above the upper limit (high activity duration) are then counted. The total cycle time is calculated, and the ratios of low-activity duration, medium-activity duration, and high-activity duration to the total cycle time are processed. These ratios are then weighted to extract activity distribution features, which characterize the overall distribution of user electricity consumption behavior within a given period. A higher value for this feature indicates a larger proportion of high-activity intervals and concentrated and active user participation within the distribution area. Furthermore, expenditure response features, expenditure rate standard deviation, adaptation stability features, revenue evolution features, revenue fluctuation features, activity diffusion features, and activity distribution features are concatenated into a time-series response feature vector. In the correlation and fusion layer of the electricity consumption response evolution mapping model, based on the time-series response feature vector of each transformer area within the set area, the corresponding electricity consumption mapping feature vector of the transformer area is extracted. Specifically, the expenditure response feature and revenue evolution feature in the time-series response feature vector are weighted to extract the revenue and expenditure evolution coordination feature, which is used to characterize the time consistency between expenditure changes and revenue changes in the set transformer area within the set period. When it is high, it indicates that the revenue and expenditure changes are synchronized and the response time is matched; when it is low, it indicates that there is a time lag between expenditure and revenue. The expenditure rate standard deviation, adaptive stationarity feature, and revenue volatility feature in the time-series response feature vector are weighted to extract steady-state feedback features. In this weighting process, the expenditure rate standard deviation and revenue volatility feature are inverted to characterize the feedback suppression capability of the set transformer area for operating fluctuations within a set period. When the value is high, it indicates that the fluctuations are promptly offset and the self-stabilization is strong; when the value is low, it indicates that the fluctuations continue to accumulate and the feedback adjustment is lagging. The active diffusion feature and active distribution feature in the time-series response feature vector are weighted to extract the active collaborative response feature. In this weighting process, the active diffusion feature is inverted to characterize the synchronicity of the electricity consumption activity of the user group in the fixed area within a set period. When it is high, it indicates that the group response direction is consistent; when it is low, it indicates that the group response is dispersed. The characteristics of revenue and expenditure evolution coordination, steady-state feedback, and active coordination response are concatenated into a power consumption mapping feature vector. In the output layer of the electricity consumption response evolution mapping model, based on the electricity consumption mapping feature vector of each transformer area within the set area, the corresponding electricity consumption evolution response feature set of the transformer area is output. Specifically, the revenue and expenditure evolution coordination feature, steady-state feedback feature, and active coordination response feature in the electricity consumption mapping feature vector are activated by the Sigmoid function to obtain revenue and expenditure evolution coordination feature value, steady-state feedback feature value, and active coordination response feature value between 0 and 1.

[0049] The pre-training steps for the electricity response evolution mapping model are as follows: The labeled dataset was obtained from historical electricity consumption monitoring records and marketing behavior data of multiple typical transformer substations. It was formed by electricity marketing experts based on historical metering data, user behavior feedback, and changes in electricity pricing strategies. Each sample in the labeled dataset includes the electricity expenditure rate, electricity load adaptation value, electricity revenue rate value, and electricity structure activity value of a set transformer substation over multiple consecutive time steps, and is accompanied by corresponding ground truth labels for electricity consumption evolution response features. All samples have been fully labeled and reviewed. The raw data was preprocessed, including time alignment, missing value imputation, and min-max standardization, to ensure the consistency of samples from each transformer substation in terms of time dimension and scale. The preprocessed dataset was then divided into training set, validation set, and test set, with the training set accounting for 80%, and the validation set and test set each accounting for 10%, while maintaining the time order to preserve time-dependent features.

[0050] During the training phase, the electricity response time-series data of each sample is input into the input layer of the model. After preprocessing, it is input into the LSTM time-series evolution layer for feature learning. The LSTM controls the flow of information through the gating mechanism of input gate, forget gate and output gate, and learns the long-term dependencies and short-term fluctuation characteristics in the time series, thereby capturing the response patterns of the transformer area under the influence of electricity price adjustments, load fluctuations and changes in user activity. The model uses the backpropagation algorithm (BPTT) to optimize the weight parameters to minimize the prediction error (such as mean square error MSE), and uses the Adam optimizer for iterative training, gradually adjusting hyperparameters such as learning rate and number of hidden layer units to improve stability and convergence speed.

[0051] During training, the model's performance at different stages is monitored using a validation set to prevent overfitting and ensure generalization ability. After training, the model is evaluated using a test set to verify its feature extraction accuracy and temporal adaptability on unseen samples. The optimized model parameters are saved and deployed in the cloud for automatic feature extraction and evolutionary characterization analysis of the power consumption response time series data of the transformer substation in actual operation, enabling online identification and continuous learning of dynamic characteristics of power consumption behavior.

[0052] In this implementation scheme, a multi-level electricity consumption response evolution mapping model is constructed to enable orderly analysis of time-series electricity consumption data. Secondly, by performing time alignment and normalization processing at the input layer, data from different transformer areas and with different parameters can be compared and modeled at the same time scale, ensuring the consistency and reliability of subsequent analysis results. After introducing LSTM into the time-series evolution layer, the dynamic dependencies between different time points can be identified, capturing the persistence and abrupt change characteristics of transformer area electricity consumption behavior. Combined with multi-feature weighted processing in the correlation and fusion layer, features reflecting the consistency of expenditure and revenue, the stability of operational fluctuations, and the synchronicity of user activity are obtained respectively. After Sigmoid normalization at the output layer, the feature values ​​are expressed intuitively. Finally, through hierarchical extraction and fusion, the depth of understanding of electricity consumption behavior is improved, thereby effectively supporting the stability of cloud-based decision-making.

[0053] Please see Figure 4 This invention provides a technical solution: a cloud-based power marketing metering data sharing system, comprising: a data acquisition and storage module for acquiring power metering operation sequence data and power consumption response time sequence data of several transformer substations within a designated area, and uploading them to a designated cloud for storage; a stability and efficiency characteristic analysis module for analyzing the stability and efficiency evolution characteristic values ​​of each transformer substation based on the power metering operation sequence data of each transformer substation within the designated area; a marketing response mapping module for analyzing the marketing coordination characteristic values ​​of each transformer substation based on a pre-trained power consumption response evolution mapping model, combined with the power consumption response time sequence data and stability and efficiency evolution characteristic values ​​of each transformer substation within the designated area; and a marketing sharing feedback module for performing marketing sharing processing on each transformer substation within the designated area based on the marketing coordination characteristic values.

[0054] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0055] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for cloud-based sharing of electricity marketing metering data, characterized in that, Includes the following steps: Acquire the power metering operation sequence data and power consumption response sequence data of several transformer substations within a designated area, and upload them to the designated cloud for storage; Based on the power metering operation sequence data of each transformer substation within the designated area, analyze the stability and efficiency evolution characteristic values ​​of the corresponding transformer substation. Based on the pre-trained electricity response evolution mapping model, and combined with the electricity response time series data and stable efficiency evolution characteristic values ​​of each transformer substation within the set area, the marketing coordination characteristic values ​​of the corresponding transformer substation are analyzed. Marketing sharing is performed on each station area within a designated region based on marketing coordination feature values.

2. The method for cloud-based sharing of electricity marketing metering data according to claim 1, characterized in that, The power metering operation sequence data includes voltage fluctuation rate, transformer area line loss rate, energy flow coordination index, energy feedback rate, voltage distortion index, and current distribution offset factor at each time point. The specific steps for analyzing the stability and efficiency evolution characteristics of each transformer area within the set area are as follows: Based on the power metering operation sequence data of each transformer substation within the designated area, analyze the operation sequence feature set of the corresponding transformer substation, including the steady-state performance feature value and load balance feature value at each time point; Based on the runtime sequence feature set of each transformer area within the designated area, the stability and efficiency evolution characteristic values ​​of the corresponding transformer area are analyzed.

3. The method for cloud-based sharing of electricity marketing metering data according to claim 2, characterized in that, The specific steps for analyzing the runtime sequence feature set of each station within the defined area are as follows: Based on the voltage fluctuation rate, line loss rate, energy flow coordination index, and energy feedback rate of each transformer area within the designated area at each time point, analyze the steady-state performance characteristics at the corresponding time points. Based on the voltage distortion index and current distribution offset factor of each transformer area within the set area at each time point, the load balance characteristic value of the corresponding time point is analyzed.

4. The method for cloud-based sharing of electricity marketing metering data according to claim 1, characterized in that, The electricity consumption response time-series data includes the electricity expenditure rate, electricity load adaptation value, electricity revenue rate value, and electricity structure activity value at each time point. The electricity consumption response evolution mapping model includes an input layer, a time-series evolution layer, a correlation and fusion layer, and an output layer.

5. The method for cloud-based sharing of electricity marketing metering data according to claim 4, characterized in that, The specific steps for analyzing the marketing coordination characteristic values ​​of each station area within the defined region are as follows: Based on the pre-trained electricity consumption response evolution mapping model, and combined with the electricity consumption response time series data of each transformer area within the set area, the marketing evolution adaptation characteristic value of the corresponding transformer area is analyzed. Based on the marketing evolution adaptation characteristic value and stable efficiency evolution characteristic value of each station area within the set area, the marketing coordination characteristic value of the corresponding station area is analyzed.

6. The method for cloud-based sharing of electricity marketing metering data according to claim 5, characterized in that, The specific steps for analyzing the marketing evolution adaptation characteristic values ​​of each station area within the defined region are as follows: The power consumption response time series data of each transformer substation within the designated area are input into the pre-trained power consumption response evolution mapping model to analyze the power consumption evolution response feature set of the corresponding transformer substation, including revenue and expenditure evolution coordination feature value, steady-state feedback feature value, and active coordination response feature value. Based on the electricity consumption evolution response feature set of each transformer substation within a defined area, the marketing evolution adaptation feature value of the corresponding transformer substation is analyzed.

7. The method for cloud-based sharing of electricity marketing metering data according to claim 6, characterized in that, The specific steps for analyzing the power consumption evolution response characteristic set of each transformer substation within the designated area are as follows: In the input layer of the power consumption response evolution mapping model, the power consumption response time series data of each transformer area within the set area are received and preprocessed. In the time-series evolution layer of the electricity consumption response evolution mapping model, based on the preprocessed electricity consumption response time-series data of each transformer substation within the set area, the time-series response feature vector of the corresponding transformer substation is extracted. In the correlation and fusion layer of the electricity consumption response evolution mapping model, the electricity consumption mapping feature vector of the corresponding transformer area is extracted based on the time-series response feature vector of each transformer area within the set area. In the output layer of the electricity consumption response evolution mapping model, based on the electricity consumption mapping feature vector of each transformer substation within a set area, the corresponding electricity consumption evolution response feature set of the transformer substation is output.

8. The method for cloud-based sharing of electricity marketing metering data according to claim 1, characterized in that, The specific formula for calculating the marketing coordination characteristic value of a certain area within a defined region is as follows: ； in, To set the marketing coordination characteristic value for a specific area within a region, To define the stable evolution characteristic value of a certain transformer area within a given region, These are the steady-state evolution adjustment coefficients stored in the database. To set the marketing evolution adaptation characteristic value for a specific area within a given region, The adaptation adjustment coefficients are stored in the database. , The difference adjustment coefficients are stored in the database. These are the coordination coefficients stored in the database.

9. The method for cloud-based sharing of electricity marketing metering data according to claim 1, characterized in that, The specific steps for performing marketing sharing processing on each transformer area within a designated region based on marketing coordination feature values ​​are as follows: Based on the marketing coordination characteristic values ​​of each station area within the designated area, the area is divided into several shared groups within the designated area; And adopt corresponding marketing service strategies for each shared group within the designated area.

10. A cloud-based system for sharing electricity marketing metering data, employing the cloud-based method for sharing electricity marketing metering data as described in any one of claims 1-9, characterized in that, include: The data acquisition and storage module is used to acquire the power metering operation sequence data and power consumption response time sequence data of several transformer substations within a set area, and upload them to the set cloud for storage. The stability and efficiency characteristic analysis module is used to analyze the stability and efficiency evolution characteristic values ​​of each transformer substation based on the power metering operation sequence data of each substation within a set area. The marketing response mapping module is used to analyze the marketing coordination characteristic values ​​of the corresponding transformer area based on the pre-trained electricity response evolution mapping model and the electricity response time series data and stable efficiency evolution characteristic values ​​of each transformer area within the set area. The Marketing Sharing Feedback Module is used to perform marketing sharing processing on each station area within a set region based on marketing coordination feature values.

Images