Electricity charge calculation optimization method, device and equipment based on user portraits

Abstract

The invention provides an optimization method, device and equipment for electric charge calculation based on a user portrait, and relates to the technical field of electric charge settlement. The method comprises the steps of obtaining a power fingerprint data set of a target user in a target time period, and determining a one-dimensional power utilization sequence of the target user based on the power fingerprint data set; inputting the one-dimensional power utilization sequence into a pre-constructed user portrait classification model to obtain a classification label of the target user; inputting the classification label and the electric energy consumption data set of the target user in the target time period into a pre-constructed electric charge calculation model, and settling the electric charge of the target user; the electric charge calculation model comprises a category module and a metering module, the category module is used for determining the charging strategy of the target user according to the classification label, and the metering module is used for calculating the electric charge according to the electric energy consumption data set of the target user and the charging strategy. According to the invention, fine-grained division of the users can be realized, and the optimal charging strategy is matched for each type of users.

Description

Technical Field

[0001] This invention relates to the field of electricity billing technology, and in particular to an optimized method, apparatus and equipment for electricity bill calculation based on user profiles. Background Technology

[0002] Electricity bill calculation is a core part of electricity management, serving as the fundamental basis for power companies to conduct economic accounting, cost control, and service pricing. A scientific and accurate electricity bill calculation method not only safeguards the legitimate rights and interests of power companies but also effectively promotes the rational allocation and efficient utilization of electricity resources. With the surge in the types and number of market-based users, the contradiction between the requirements for high timeliness and high accuracy in electricity bill calculation has become increasingly prominent.

[0003] Current electricity billing calculations are primarily based on simple linear relationships or fixed price structures. For example, a common hierarchical calculation model represents meters and electricity prices using a tree structure, mapping meters to prices. A formula-parametric calculation model, on the other hand, abstracts the actual electricity consumption of users into user, billing information, metering point, and meter layers, and performs calculations according to formulas for each layer.

[0004] However, current electricity billing methods are primarily applicable to situations where the electricity market was relatively stable and user demand was relatively simple. With the continuous transformation of the electricity market and the increasing diversification of user needs, the limitations of traditional electricity billing methods are becoming increasingly apparent. For example, they fail to accurately reflect real-time supply and demand changes in the electricity market, cannot effectively incentivize users to conserve electricity and avoid peak-hour consumption, and cannot provide power companies with refined operational decision-making support. Summary of the Invention

[0005] This invention provides an optimized method, apparatus, and device for electricity bill calculation based on user profiles, in order to solve the problem that current electricity bill calculation methods are singular and cannot match different billing methods according to users' electricity consumption.

[0006] In a first aspect, embodiments of the present invention provide an optimized method for electricity cost calculation based on user profiles, comprising: Obtain the electricity fingerprint dataset of the target user during the target time period; Determine the one-dimensional electricity consumption sequence of the target user based on the power fingerprint dataset; The one-dimensional electricity consumption sequence is input into a pre-built user profile classification model to obtain the classification label of the target user; The classification labels and the target user's electricity consumption dataset during the target time period are input into a pre-built electricity billing model to settle the target user's electricity bill. The electricity billing model includes a category module and a metering module. The category module is used to determine the billing strategy for the target user based on the classification labels, and the metering module is used to calculate the electricity bill based on the target user's electricity consumption dataset and billing strategy.

[0007] In one possible implementation, the power fingerprint dataset includes a power current dataset, a power load access dataset, and a power consumption dataset.

[0008] In one possible implementation, determining the one-dimensional electricity consumption sequence of a target user based on an electricity fingerprint dataset includes: Calculate the covariance among the current consumption dataset, the load connection dataset, and the energy consumption dataset, construct the covariance matrix, and construct the divergence matrix based on the covariance matrix; A one-dimensional reduced matrix is ​​constructed based on the eigenvalues ​​and eigenvectors of the scatter matrix. The power fingerprint dataset is dimensionality reduced using a one-dimensional dimensionality reduction matrix to obtain the one-dimensional electricity consumption sequence of the target user.

[0009] In one possible implementation, the user profile classification model is a multi-class logistic regression module, and the classification labels include electricity consumption characteristic labels and electricity consumption habit labels.

[0010] In one possible implementation, the classification labels and the target user's electricity consumption dataset for a target time period are input into a pre-built electricity billing model to calculate the target user's electricity bill, including: Input the electricity consumption characteristic tags and electricity consumption habit tags into the category module to obtain the billing strategy for the target user; The billing strategy and the target user's electricity consumption dataset for the target period are input into the metering module to settle the target user's electricity bill.

[0011] In one possible implementation, the category module includes a clustering model; The billing strategies include residential demand response subsidy strategy, residential power factor adjustment strategy, industrial and commercial capacity demand response subsidy strategy, industrial and commercial capacity power factor adjustment strategy, industrial and commercial capacity penalty strategy, industrial and commercial demand response subsidy strategy, large industrial high reliability response subsidy strategy, large industrial agreement penalty strategy, and large industrial agreement power factor adjustment strategy.

[0012] In one possible implementation, the electricity characteristic label includes an appliance composition characteristic label, a load characteristic label, and an electricity consumption characteristic label. Electricity consumption habit labels include tiered electricity pricing percentage labels, total electricity consumption percentage of regional average electricity consumption labels, and electricity cost sensitivity labels.

[0013] In one possible implementation, after obtaining the electricity fingerprint dataset of the target user during the target time period, the following steps are also included: Data cleaning and outlier removal are performed on the power fingerprint dataset of the target user during the target time period.

[0014] Secondly, embodiments of the present invention provide an optimization device for electricity cost calculation based on user profiles, comprising: The data acquisition module is used to acquire the power fingerprint dataset of the target user during the target time period; Construct a sequence module to determine the one-dimensional electricity consumption sequence of a target user based on an electricity fingerprint dataset; The classification tag module is used to input a one-dimensional electricity consumption sequence into a pre-built user profile classification model to obtain the classification tags of the target user; The electricity billing module is used to input the classification labels and the target user's electricity consumption dataset during the target period into the pre-built electricity billing calculation model to settle the target user's electricity bill. The electricity billing calculation model includes a category module and a metering module. The category module is used to determine the billing strategy for the target user based on the classification labels, and the metering module is used to calculate the electricity bill based on the target user's electricity consumption dataset and billing strategy.

[0015] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect or any possible implementation thereof.

[0016] In this embodiment of the invention, to adopt different billing methods for different users, it is necessary to first obtain the electricity fingerprint dataset of the target user during the target time period. After obtaining the electricity fingerprint dataset, it is also necessary to deeply mine the deep correlation between the electricity fingerprint datasets to obtain the one-dimensional electricity consumption sequence of the target user. In order to accurately classify the target user, the one-dimensional electricity consumption sequence needs to be input into the user profile classification model to obtain the classification label of the target user. After determining the classification label of the target user, the classification label and the electricity consumption dataset of the target user during the target time period can be input into the pre-built electricity billing calculation model to complete the electricity billing for the target user. This invention, by mining user electricity consumption behavior from the electricity fingerprint dataset, can accurately capture the differences in electricity consumption patterns of different users. Combined with the user profile classification model, it can achieve fine-grained user segmentation and match the optimal billing strategy for each type of user. Thus, it realizes the upgrade of electricity billing from a single billing standard to multiple standards, providing key technical support for accurate pricing in the context of electricity marketization. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the implementation of the optimized method for electricity cost calculation based on user profiles provided in this embodiment of the invention. Figure 2 This is a schematic diagram of the structure of the optimized device for electricity cost calculation based on user profiles provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0018] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0019] Electricity bill calculation refers to the process of determining the amount of electricity a user should pay based on their electricity consumption, the nature of their electricity use, and the pricing strategy of the power company, using specific mathematical models and algorithms. This process involves several key elements, including electricity price, electricity consumption, and the time of day in which electricity is used. Electricity price is the foundation of electricity bill calculation; it is typically determined by government departments based on factors such as supply and demand in the electricity market, cost structure, and energy policies. Electricity consumption represents the actual amount of electricity consumed by the user and is the direct basis for electricity bill calculation. Different times of day in which electricity is used may lead to differences in electricity prices, thus affecting the total electricity bill.

[0020] As introduced in the background section, traditional electricity billing models are mainly based on simple linear relationships or fixed electricity price structures. They are difficult to accurately reflect real-time supply and demand changes in the electricity market, cannot effectively incentivize users to save electricity and use electricity during off-peak hours, and cannot provide power companies with refined operational decision support.

[0021] In order to take advantage of the new changes in the electricity market and meet the new needs of users, it is necessary to develop a new electricity billing method based on the supply and demand dynamics of the electricity market and users' electricity consumption behavior, so as to encourage users to save electricity and use electricity during off-peak hours.

[0022] A user profile is a labeled user model abstracted from information such as a user's social attributes, lifestyle habits, and consumption behavior. User profiles make implicit user characteristics explicit, helping to understand user needs, identify target users, and tap into user control potential, thereby enabling a more accurate understanding of each user's electricity consumption characteristics.

[0023] See Figure 1 The document illustrates a flowchart of the method for optimizing electricity cost calculation based on user profiles, as provided in an embodiment of the present invention, and details are as follows: S110. Obtain the power fingerprint dataset of the target user during the target time period.

[0024] Data collection is a crucial step in electricity bill calculation. The scope of data collection is very broad, including users' electricity consumption data, real-time electricity price data in the electricity market, and operating status data of power equipment.

[0025] In some embodiments, the power fingerprint dataset includes a power current dataset, a power load access dataset, and a power consumption dataset.

[0026] In this embodiment, since these data typically originate from various channels such as smart meters, power trading platforms, and power monitoring systems, they possess different data formats and collection frequencies. Therefore, after acquiring this data, it is necessary to clean it and remove outliers. Invalid and erroneous data are removed, missing data is filled in, and the data undergoes necessary normalization and standardization to eliminate the influence of units and dimensions, providing a solid foundation for subsequent data analysis.

[0027] S120. Determine the one-dimensional electricity consumption sequence of the target user based on the power fingerprint dataset.

[0028] Since multiple different types of data have been acquired, in order to deeply explore the correlation between the data, it is also necessary to determine the one-dimensional electricity consumption sequence of the target user based on the power fingerprint dataset.

[0029] In some embodiments, the first step may be to obtain the access load dataset, current dataset, and energy consumption dataset of all target meters within the target area during the target time period.

[0030] For example, the power fingerprint dataset of smart meters is sampled at regular intervals, including: access load dataset, current dataset and energy consumption dataset. The number of samplings is denoted as n, where n≥1. The access load dataset is denoted as x1, the current dataset as x2, and the energy consumption dataset as x3, forming a high-dimensional dataset X={x1,x2,x3}.

[0031] Then, the covariance between the connected load dataset, the current dataset, and the energy consumption dataset is calculated, a covariance matrix is ​​constructed, and a divergence matrix is ​​constructed based on the covariance matrix.

[0032] Specifically, the covariance between the datasets of each dimension is calculated, and the covariance matrix Cov(x1,x2,x3) is constructed. ; The Cov function mentioned above is the covariance calculation function, which represents the degree of correlation between data in two datasets.

[0033] Calculate the scatter matrix Xᵀ of a high-dimensional dataset X. XXT=(n-1)·Cov(x1,x2,x3).

[0034] Next, a one-dimensional reduced matrix is ​​constructed based on the eigenvalues ​​and eigenvectors of the scatter matrix.

[0035] Specifically, find the eigenvalues ​​and eigenvectors of the scatter matrix XXT, arrange the obtained eigenvectors in descending order of eigenvalues, standardize the first three eigenvectors, combine them into an eigenvector matrix R, and finally take the first row of the eigenvector matrix R to form a one-dimensional reduced matrix P.

[0036] Finally, the power fingerprint dataset is dimensionality reduced based on a one-dimensional dimensionality reduction matrix to obtain the one-dimensional electricity consumption sequence of the target meter.

[0037] Specifically, a one-dimensional reduction matrix P is used to reduce the dimensionality of dataset X, resulting in a dimensionality-reduced dataset Y, where Y = P·X. Dataset Y is a sequence of one row and n columns, representing the one-dimensional electricity consumption sequence of the target user, which is also the user's comprehensive electricity consumption fingerprint.

[0038] S130. Input the one-dimensional electricity consumption sequence into the pre-built user profile classification model to obtain the classification label of the target user.

[0039] In order to more accurately understand users' electricity consumption characteristics, a user profile classification model was constructed.

[0040] In some embodiments, the user profile classification model is a multi-class logistic regression module, and the classification labels include electricity consumption characteristic labels and electricity consumption habit labels.

[0041] In this embodiment, the electricity consumption characteristic label includes an appliance composition characteristic label, a load characteristic label, and a power consumption characteristic label. The electricity consumption habit label includes a tiered electricity price percentage label, a percentage of total electricity consumption in the regional average electricity consumption label, and an electricity price sensitivity label.

[0042] Specifically, the appliance characteristic label indicates the ratio of the total power of adjustable and reducible loads to the base power. Adjustable loads refer to loads that do not stop running but whose power is adjustable, including air conditioners, electric heating (with power ratings), lighting, and electric vehicles. Reducible loads refer to loads that can be stopped and moved to other time periods, including electric water heaters (with memory function), smart kitchen appliances (without memory function), air conditioners (with memory function), washing machines, and electric vehicles. Memory function means that the power in the next time period is affected by the previous time period; some appliances may have their operating time shifted, but the power remains the same.

[0043] The load characteristic label includes the maximum load power, average load power, and maximum load utilization hours for the target period.

[0044] The power consumption characteristic label is the ratio of user power consumption to the average power consumption per user in the area.

[0045] The tiered electricity pricing percentage label refers to the percentage of the third-tier electricity pricing compared to the first and second-tier electricity pricing.

[0046] The label representing the percentage of total electricity consumption in the region's average electricity consumption refers to the percentage of the total electricity consumption of the target user within the target time period to the region's average electricity consumption within the target time period.

[0047] The electricity cost sensitive label refers to the peak-valley electricity consumption ratio.

[0048] In this embodiment, the training sample set of the user profile classification model includes multiple training samples. Each training sample includes a one-dimensional electricity consumption sequence of the training user, as well as the appliance composition characteristic label, load characteristic label, electricity consumption characteristic label, tiered electricity price ratio label, total electricity consumption ratio in the region's total average electricity consumption label, and electricity price sensitivity label corresponding to the training user.

[0049] S140. Input the classification labels and the target user's electricity consumption dataset during the target time period into the pre-built electricity bill calculation model to settle the target user's electricity bill.

[0050] In some embodiments, the electricity billing model includes a category module and a metering module. The category module is used to determine the billing strategy for the target user based on the category label, and the metering module is used to calculate the electricity bill based on the target user's electricity consumption dataset and the billing strategy.

[0051] In this embodiment, electricity consumption characteristic tags and electricity consumption habit tags can be input into the category module to obtain the billing strategy for the target user. Then, the billing strategy and the target user's electricity consumption dataset for the target period are input into the metering module to settle the target user's electricity bill.

[0052] In this embodiment, the category module includes a strategy clustering model, which is used to obtain the billing strategy for the target user based on the input electricity consumption characteristic tags and electricity consumption habit tags.

[0053] Specifically, the billing strategies include residential demand response subsidy strategy, residential power factor adjustment strategy, industrial and commercial capacity demand response subsidy strategy, industrial and commercial capacity power factor adjustment strategy, industrial and commercial capacity penalty strategy, industrial and commercial demand response subsidy strategy, large industrial high reliability response subsidy strategy, large industrial negotiated penalty electricity price and large industrial negotiated power factor adjustment strategy.

[0054] The strategy clustering model will output the billing strategy for the target user based on the input appliance composition characteristic labels, load characteristic labels, power consumption characteristic labels, tiered electricity price ratio labels, total electricity consumption ratio in the region's average electricity consumption label, and electricity price sensitivity labels.

[0055] The residential demand response subsidy strategy and residential power factor adjustment strategy are billing strategies for residential electricity consumption. The industrial and commercial capacity demand response subsidy strategy, industrial and commercial capacity power factor adjustment strategy, industrial and commercial capacity penalty strategy, and industrial and commercial demand response subsidy strategy are billing strategies for industrial and commercial electricity consumption. The large industrial high reliability response subsidy strategy, large industrial agreement penalty strategy, and large industrial agreement power factor adjustment strategy are billing strategies for large industrial users' electricity consumption.

[0056] Among these, the residential demand response subsidy strategy refers to a strategy where residents receive economic rewards for proactively reducing their electricity load during peak hours to help the power grid balance supply and demand. The residential power factor adjustment strategy refers to a strategy where residents are charged extra for excessive reactive power consumption from electrical appliances.

[0057] The industrial and commercial capacity demand response subsidy strategy refers to a strategy where industrial and commercial users are charged according to capacity and receive economic rewards for actively reducing their electricity load during peak hours to help the grid balance supply and demand. The industrial and commercial capacity power factor adjustment strategy refers to a strategy where industrial and commercial users are charged according to capacity and additional charges are levied for excessive reactive power consumption. The industrial and commercial capacity penalty strategy refers to a strategy where industrial and commercial users are charged according to capacity and additional charges are levied for exceeding their electricity quotas or using polluting energy sources. The industrial and commercial demand response subsidy strategy refers to a strategy where industrial and commercial users are charged according to their maximum instantaneous power and receive economic rewards for actively reducing their electricity load during peak hours to help the grid balance supply and demand.

[0058] The high-reliability response subsidy strategy for large industrial users refers to a strategy where large industrial users adopt high-reliability electricity pricing and actively reduce their electricity load during peak hours to help the power grid balance supply and demand, thereby receiving economic rewards. The large-industry agreement penalty strategy refers to a strategy where large industrial users adopt agreement-based electricity pricing, but are charged extra for exceeding their electricity quotas or using polluting energy sources. The large-industry agreement power factor adjustment strategy refers to a strategy where large industrial users adopt agreement-based electricity pricing, but are charged extra for excessive reactive power equipment usage.

[0059] High reliability electricity price refers to the additional charge for users with high requirements for power supply quality, while negotiated electricity price refers to the electricity price agreed upon with the power grid.

[0060] The optimized electricity billing method provided by this invention requires, firstly, acquiring the electricity fingerprint dataset of the target user during the target time period to apply different billing methods to different users. After acquiring the electricity fingerprint dataset, it is necessary to deeply mine the deep correlations between the electricity fingerprint datasets to obtain a one-dimensional electricity consumption sequence for the target user. To accurately classify the target user, the one-dimensional electricity consumption sequence is input into a user profile classification model to obtain the target user's classification label. After determining the target user's classification label, the classification label and the target user's electricity consumption dataset during the target time period are input into a pre-constructed electricity billing calculation model to complete the electricity billing for the target user. This invention, by mining user electricity consumption behavior from the electricity fingerprint dataset, can accurately capture the differences in electricity consumption patterns among different users. Combined with the user profile classification model, it achieves fine-grained user segmentation and can match the optimal billing strategy for each user category. This realizes the upgrade of electricity billing from a single billing standard to multiple standards, providing key technical support for accurate pricing in a market-oriented electricity environment.

[0061] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0062] The following are device embodiments of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0063] Figure 2 A schematic diagram of the optimized electricity bill calculation device based on user profile provided in an embodiment of the present invention is shown. For ease of explanation, only the parts related to the embodiment of the present invention are shown, and are described in detail below: like Figure 2 As shown, the optimized electricity bill calculation device 200 based on user profiles includes: The data acquisition module 210 is used to acquire the power fingerprint dataset of the target user during the target time period; Construct a sequence module 220 to determine the one-dimensional electricity consumption sequence of a target user based on an electricity fingerprint dataset; The classification label module 230 is used to input a one-dimensional electricity consumption sequence into a pre-built user profile classification model to obtain the classification label of the target user; The electricity billing module 240 is used to input the classification labels and the target user's electricity consumption dataset during the target period into the pre-built electricity billing calculation model to settle the target user's electricity bill. The electricity billing calculation model includes a category module and a metering module. The category module is used to determine the billing strategy of the target user based on the classification labels, and the metering module is used to calculate the electricity bill based on the target user's electricity consumption dataset and billing strategy.

[0064] In one possible implementation, the power fingerprint dataset includes a power current dataset, a power load access dataset, and a power consumption dataset.

[0065] In one possible implementation, a sequence module 220 is constructed to calculate the covariance between the current dataset, the load access dataset, and the energy consumption dataset, construct a covariance matrix, and construct a divergence matrix based on the covariance matrix. A one-dimensional reduced matrix is ​​constructed based on the eigenvalues ​​and eigenvectors of the scatter matrix. The power fingerprint dataset is dimensionality reduced using a one-dimensional dimensionality reduction matrix to obtain the one-dimensional electricity consumption sequence of the target user.

[0066] In one possible implementation, the user profile classification model is a multi-class logistic regression module, and the classification labels include electricity consumption characteristic labels and electricity consumption habit labels.

[0067] In one possible implementation, the electricity billing module 240 is used to input electricity consumption characteristic tags and electricity consumption habit tags into the category module to obtain the billing strategy for the target user. The billing strategy and the target user's electricity consumption dataset for the target period are input into the metering module to settle the target user's electricity bill.

[0068] In one possible implementation, the category module is a clustering model; The billing strategies include residential demand response subsidy strategy, residential power factor adjustment strategy, industrial and commercial capacity demand response subsidy strategy, industrial and commercial capacity power factor adjustment strategy, industrial and commercial capacity penalty strategy, industrial and commercial demand response subsidy strategy, large industrial high reliability response subsidy strategy, large industrial agreement penalty strategy, and large industrial agreement power factor adjustment strategy.

[0069] In one possible implementation, the electricity characteristic label includes an appliance composition characteristic label, a load characteristic label, and an electricity consumption characteristic label. Electricity consumption habit labels include tiered electricity pricing percentage labels, total electricity consumption percentage of regional average electricity consumption labels, and electricity cost sensitivity labels.

[0070] In one possible implementation, the data acquisition module 210 is used to perform data cleaning and outlier removal on the power fingerprint dataset of the target user during the target time period.

[0071] The electricity billing optimization device provided by this invention requires acquiring the electricity fingerprint dataset of the target user during the target time period in order to adopt different billing methods for different users. After acquiring the electricity fingerprint dataset, it is necessary to deeply mine the deep correlations between the electricity fingerprint datasets to obtain the one-dimensional electricity consumption sequence of the target user. In order to accurately classify the target user, the one-dimensional electricity consumption sequence needs to be input into the user profile classification model to obtain the classification label of the target user. After determining the classification label of the target user, the classification label and the electricity consumption dataset of the target user during the target time period can be input into the pre-constructed electricity billing calculation model to complete the electricity billing for the target user. This invention can accurately capture the differences in electricity consumption patterns of different users by mining user electricity consumption behavior from the electricity fingerprint dataset. Combined with the user profile classification model, it can achieve fine-grained user segmentation and match the optimal billing strategy for each type of user. This realizes the upgrade of electricity billing from a single billing standard to multiple standards, providing key technical support for accurate pricing in the context of electricity marketization.

[0072] Figure 3 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. For example... Figure 3 As shown, the electronic device 3 in this embodiment includes a processor 30 and a memory 31. The memory 31 stores a computer program 32. When the processor 30 executes the computer program 32, it implements the steps in the various method embodiments described above. Alternatively, when the processor 30 executes the computer program 32, it implements the functions of each module / unit in the various device embodiments described above.

[0073] For example, computer program 32 may be divided into one or more modules / units, which are stored in memory 31 and executed by processor 30 to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of computer program 32 in electronic device 3.

[0074] Electronic device 3 may include, but is not limited to, processor 30 and memory 31. Those skilled in the art will understand that... Figure 3This is merely an example of electronic device 3 and does not constitute a limitation on electronic device 3. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device 3 may also include input / output devices, network access devices, buses, etc.

[0075] For the sake of simplicity and clarity, only the above-described functional modules / units are used as examples. In practical applications, the functions described above can be assigned to different functional modules / units as needed. These modules / units can be implemented in hardware, software, or a combination of both.

[0076] In the above embodiments, the descriptions of each embodiment have their own emphasis. Parts not detailed or described in a particular embodiment can be referred to in the relevant descriptions of other embodiments. Unless otherwise specified or in conflict with logic, the terminology and / or descriptions between different embodiments are consistent and can be referenced interchangeably. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0077] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. An optimized method for electricity cost calculation based on user profiles, characterized in that, include: Obtain the electricity fingerprint dataset of the target user during the target time period; Based on the power fingerprint dataset, a one-dimensional electricity consumption sequence of the target user is determined; The one-dimensional electricity consumption sequence is input into a pre-built user profile classification model to obtain the classification label of the target user; The classification labels and the target user's electricity consumption dataset during the target time period are input into a pre-built electricity billing model to settle the target user's electricity bill. The electricity billing model includes a category module and a metering module. The category module is used to determine the billing strategy for the target user based on the classification labels, and the metering module is used to calculate the electricity bill based on the target user's electricity consumption dataset and the billing strategy.

2. The optimized method for electricity cost calculation based on user profiles according to claim 1, characterized in that, The power fingerprint dataset includes a power current dataset, a power load access dataset, and a power consumption dataset.

3. The optimized method for electricity cost calculation based on user profiles according to claim 2, characterized in that, Determining the one-dimensional electricity consumption sequence of the target user based on the electricity fingerprint dataset includes: Calculate the covariance among the current consumption dataset, the load access dataset, and the energy consumption dataset, construct a covariance matrix, and construct a divergence matrix based on the covariance matrix; Based on the eigenvalues ​​and eigenvectors of the scatter matrix, a one-dimensional reduced-dimensional matrix is ​​constructed. The power fingerprint dataset is dimensionality reduced based on the one-dimensional reduction matrix to obtain the one-dimensional electricity consumption sequence of the target user.

4. The optimized method for electricity cost calculation based on user profiles according to claim 1, characterized in that, The user profile classification model is a multi-class logistic regression module, and the classification labels include electricity consumption characteristic labels and electricity consumption habit labels.

5. The optimized method for electricity cost calculation based on user profiles according to claim 4, characterized in that, The step of inputting the classification labels and the target user's electricity consumption dataset during the target time period into a pre-built electricity bill calculation model to settle the target user's electricity bill includes: Input the electricity consumption characteristic tags and electricity consumption habit tags into the category module to obtain the billing strategy for the target user; The billing strategy and the target user's electricity consumption dataset during the target period are input into the metering module to settle the target user's electricity bill.

6. The optimized method for electricity cost calculation based on user profiles according to claim 5, characterized in that, The category module includes a clustering model; The billing strategies include residential demand response subsidy strategy, residential power factor adjustment strategy, industrial and commercial capacity demand response subsidy strategy, industrial and commercial capacity power factor adjustment strategy, industrial and commercial capacity penalty strategy, industrial and commercial demand response subsidy strategy, large industrial high reliability response subsidy strategy, large industrial agreement penalty strategy, and large industrial agreement power factor adjustment strategy.

7. The optimized method for electricity bill calculation based on user profiles according to any one of claims 4-6, characterized in that, The electricity consumption characteristic label includes an electrical component characteristic label, a load characteristic label, and a power consumption characteristic label; The electricity consumption habit labels include tiered electricity price percentage labels, total electricity consumption percentage of the regional average electricity consumption labels, and electricity price sensitivity labels.

8. The optimized method for electricity bill calculation based on user profiles according to any one of claims 1-6, characterized in that, After obtaining the power fingerprint dataset of the target user during the target time period, the method further includes: Data cleaning and outlier removal are performed on the power fingerprint dataset of the target user during the target time period.

9. An optimization device for electricity cost calculation based on user profiles, characterized in that, include: The data acquisition module is used to acquire the power fingerprint dataset of the target user during the target time period; A sequence module is constructed to determine the one-dimensional electricity consumption sequence of the target user based on the electricity fingerprint dataset; The classification label module is used to input the one-dimensional electricity consumption sequence into a pre-built user profile classification model to obtain the classification label of the target user; The electricity billing module is used to input the classification labels and the target user's electricity consumption dataset during the target period into a pre-built electricity billing calculation model to settle the target user's electricity bill. The electricity billing calculation model includes a category module and a metering module. The category module is used to determine the billing strategy of the target user based on the classification labels, and the metering module is used to calculate the electricity bill based on the target user's electricity consumption dataset and the billing strategy.

10. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method as described in any one of claims 1 to 8.

Images