Electricity charge abnormal error processing method and system based on intelligent verification

Abstract

The invention discloses an electric charge abnormity error processing method and system based on intelligent verification, and relates to the technical field of electric power billing, and the method comprises the steps: obtaining electric charge abnormity first data, carrying out the abnormity detection of electric charge calculation based on a neural network algorithm, and enabling the electric charge abnormity first data to comprise an electricity price feature, a force adjustment standard feature and a definite proportion feature; when the electric charge calculation is abnormal, correcting the electric quantity based on load flow calculation and an energy balance algorithm, retrieving historical electricity price information of a user, performing consistency verification on the historical electricity price information and current electricity price information, judging and updating the electricity price of the user, and recalculating the electric charge; real-time updating three-color labeling display is performed on an electric charge correction calculation result based on a visualization technology, contribution degrees of input parameters to electric charges are obtained based on a local sensitivity method, and importance ranking is performed on the input parameters according to the contribution degrees; according to the invention, through accurate correction of the key parameters of electricity charge calculation, the efficiency and accuracy of electricity charge error processing are effectively improved, the working pressure of basic units is reduced, and the customer satisfaction is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electricity billing, and more specifically, to a method and system for handling abnormal errors in electricity charges based on intelligent verification. Background Art

[0002] With the development of social economy and the advancement of science and technology, the power system has become an indispensable part of the infrastructure of modern society. The relationship between power consumption and production is becoming increasingly complex. Power companies use smart meters and automated metering systems to accurately monitor power consumption and settle fees. However, with the upgrading of power equipment and the continuous development of smart meter technology, there are still certain errors and anomalies in electricity metering, which affects the fairness and transparency between power suppliers and users.

[0003] For example, the invention patent with announcement number CN117786568A announces a basic electricity fee anomaly detection method, device, equipment and storage medium, including: retrieving electricity consumption data of at least one target user that meets preset conditions from the marketing system database; processing the electricity consumption data according to the preset anomaly detection conditions to determine whether the electricity consumption data is abnormal; if so, giving an abnormal prompt. The present invention effectively reduces the error rate of basic electricity fee review and improves the efficiency and accuracy of basic electricity fee verification.

[0004] For example, the method for determining the abnormal type of an electric meter and its device, electronic device and storage medium announced in the invention patent with announcement number: CN118940199B include: determining the electric meter to be monitored based on the meter replacement service work order, and obtaining the attribute data of the electric meter to be monitored; extracting the metering data of the electric meter to be monitored from the metering database, and preprocessing the metering data to obtain the preprocessed metering data; performing abnormal judgment on the preprocessed metering data based on the pre-constructed abnormal judgment rules to obtain the abnormal judgment result of the electric meter to be monitored; in the case that the electric meter to be monitored has an abnormality, generating abnormal alarm information based on the abnormal judgment result, and sending the abnormal alarm information to the operation and maintenance end. The present invention solves the technical problem of low troubleshooting efficiency in the related art, that is, locating the abnormal electric meter through the electricity charge error of the electric meter, and manually troubleshooting the abnormal electric meter to determine the abnormal type so as to troubleshoot the electric meter abnormality.

[0005] The above disclosed technical solutions have at least the following technical problems: In the power industry, errors in power user profile parameters may lead to inaccurate electricity bill calculations, resulting in electricity bill losses for power grid companies and users, and even customer complaints. Existing electricity bill calculation and correction methods often rely on manual review and manual correction, which is cumbersome and prone to omissions or errors. In addition, user satisfaction is not high. In order to improve the efficiency and accuracy of electricity bill error correction, an intelligent and efficient processing method is urgently needed.

[0006] In view of the above problems, the present invention proposes a solution. Summary of the invention

[0007] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present invention provide a method and system for handling abnormal errors in electricity charges based on intelligent verification, which realizes the accurate correction of key parameters in electricity charge calculation, reproduces the entire process of previous electricity charge calculation and outputs the correct formula for the electricity charge calculation process. It effectively improves the efficiency and accuracy of electricity charge error handling, reduces the work pressure of grassroots units, and improves customer satisfaction, with significant economic and social benefits.

[0008] To achieve the above object, the present invention provides the following technical solutions: The method and system for handling abnormal errors in electricity charges based on intelligent verification include the following steps: obtaining abnormal first data of electricity charges, performing abnormal detection on electricity charge calculation based on a neural network algorithm, wherein the abnormal first data of electricity charges include electricity price characteristics, power regulation standard characteristics, and fixed ratio characteristics; when the electricity charge calculation is abnormal, correcting the power consumption based on a power flow calculation method and an energy balance algorithm; performing consistency verification by retrieving the user's historical electricity price information and the current electricity price information, judging and updating the user's electricity price, and recalculating the electricity charge based on the corrected power consumption and electricity price; updating the three-color annotation display of the electricity charge correction calculation result in real time based on visualization technology, and obtaining the input parameters of the electricity charge, obtaining the contribution of the input parameters to the electricity charge based on a local sensitivity method, and ranking the input parameters in importance according to the contribution.

[0009] In a preferred embodiment, the method obtains the first data of abnormal electricity charges, and performs abnormality detection on the electricity charge calculation based on a neural network algorithm, specifically: obtains historical abnormal electricity charge data, and the historical abnormal electricity charge data includes historical electricity price characteristics, historical power adjustment standard characteristics, and historical fixed ratio characteristics; calculates the average of all data in the historical abnormal electricity charge data; based on a clustering algorithm, sets the average of all data as the initial cluster center, and assigns each data point to the initial cluster center according to the cosine similarity between the historical abnormal electricity charge data and the initial cluster center to form a cluster; recalculates the cluster center of each cluster, and continues to assign until the iteration ends; constructs an anomaly detection model based on the neural network algorithm for the processed data; inputs the first data of abnormal electricity charge into the anomaly detection model, obtains the output of several anomaly detection models and calculates the standard deviation based on a statistical method; if the standard deviation exceeds the preset discrete degree value, there is an abnormal electricity charge.

[0010] In a preferred embodiment, when the electricity fee calculation is abnormal, the electricity quantity is corrected based on the power flow calculation method and the energy balance algorithm, specifically: the corrected indication and meter ratio are obtained by actual detection and measurement of the electricity meter; the topological structure of the power grid is obtained, and the topological structure of the power grid is analyzed based on the power flow calculation method to obtain the power data of each node of the power grid, and the power data includes current, input power, output power and power flow direction; the line loss of the transmission line is obtained according to the power loss formula based on the current, and the line loss is quantified to obtain the corrected line power; the input power and output power are obtained and compared. If there is a deviation between the input power and the output power, the refund and subsidy power is corrected based on the power balance equation in combination with the power difference and the line loss power; the corrected indication, meter ratio, line power and refund and subsidy power are recalculated based on the billing power calculation model. The actual power of the user is recalculated.

[0011] In a preferred embodiment, the combined power difference and line loss power are used to correct the withdrawn power based on the power balance equation, specifically: the withdrawn power objective function is defined by minimizing the difference between the actual measured power and the standard power, and based on the law of conservation of energy, the power balance equation is constructed by combining the input power, output power and line loss power; the withdrawn power objective function is solved based on the least squares method, and the withdrawn power is continuously updated by calculating the difference between the actual measured power and the corrected withdrawn power until the objective function reaches the minimum value, and the iteration is stopped, and the optimal solution obtained is the corrected withdrawn power.

[0012] In a preferred embodiment, the method retrieves the user's historical electricity price information and performs consistency verification with the current electricity price information, determines and updates the user's electricity price, and recalculates the electricity fee based on the revised electricity consumption and electricity price. Specifically, the user's electricity price information is obtained, and the user's electricity price information includes electricity consumption category, user classification, voltage level and time-of-use electricity price information; the user's historical electricity price information is retrieved, and the consistency verification is performed with the current user electricity price information to determine whether the user's electricity price information is changed; if the verification results are inconsistent, the user's changed electricity price information is updated, and the user's changed electricity price is corrected according to the current electricity price standard; according to the revised electricity price and electricity consumption, the kilowatt-hour electricity fee, basic electricity fee, power regulation electricity fee and other electricity fees are calculated respectively, and the electricity fee calculation formula is regenerated.

[0013] In a preferred embodiment, the visualization technology is used to update the three-color annotation display of the electricity charge correction calculation results in real time, specifically: red, yellow and blue are used to guide users to use and clearly display the calculation results; blue is used to mark changeable parameters as input values ​​of the electricity charge calculation formula; red is used to mark corrected parameters, which are displayed side by side with the original data to facilitate user comparison and viewing; yellow represents the output result value, which is displayed side by side with the original data to accurately display the changes in related results; according to the electricity charge composition structure, the electricity charge composition structure includes electricity charge, basic electricity charge, power regulation electricity charge and other electricity charges, and the calculation process after the correction parameters is listed in detail.

[0014] In a preferred embodiment, the input parameters of the electricity bill are obtained, and the contribution of the input parameters to the electricity bill is obtained based on the local sensitivity method, and the input parameters are sorted in importance according to the contribution, specifically: the input parameters affecting the electricity bill are obtained, and the input parameters include electricity price, electricity consumption, basic electricity charge and power regulation electricity charge; for all parameters in the input parameters, based on local sensitivity analysis, the inter-group variance and total variance of all parameters are calculated; the inter-group variance and the total variance are compared and analyzed, and the variance contribution of all parameters in the input parameters is obtained according to the comparison results; the input parameters of the electricity bill are sorted in importance from large to small according to the variance contribution of all parameters.

[0015] In a preferred embodiment, the electricity price characteristics are specifically obtained by the following method: based on time series analysis, a time window for electricity price data is set; for each time point, the data in the time window before and after the time point is taken, and the arithmetic mean of these data is calculated as the smoothed value of the current point; the actual electricity price is obtained, and the deviation value between the actual electricity price and the smoothed value is calculated; the standard deviation of the smoothed value is used as a reference value, and if the deviation value exceeds three times the standard deviation of the smoothed value, the deviation value is quantified as an electricity price characteristic.

[0016] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. Using neural networks and clustering algorithms for anomaly detection can accurately identify the root cause of abnormal electricity charges. By cleaning and processing historical data and analyzing real-time electricity charge data, abnormal fluctuations in electricity charges can be discovered in a timely manner, unnecessary expenses can be reduced, and electricity usage strategies can be optimized.

[0017] 2. Accurately correct abnormal detection results in electricity fee calculation by comprehensively applying intelligent computing models, data proofreading, energy balance algorithms and least squares optimization methods. Its advantages include: ensuring the accuracy of electricity data and avoiding electricity fee errors caused by measurement deviations; realizing electricity fee calculations consistent with users' actual electricity consumption behaviors by correcting parameters such as meter rate, voltage, and current; optimizing the refunded electricity by least squares method to ensure the stability of power grid operation and the energy balance of the power system; at the same time, ensuring the accurate application of electricity price standards, avoiding inconsistencies caused by changes in electricity price policies or user classifications, and improving the accuracy and fairness of electricity fee settlement.

[0018] 3. Combining visualization technology and multivariate sensitivity analysis, the revised electricity fee calculation results are updated and displayed in real time, making the electricity fee calculation process more transparent and intuitive, and convenient for business personnel to quickly check and customers to understand. The three-color marking (red, yellow, and blue) clearly marks the input parameters, revised data, and output results, helping users to quickly identify changes and optimize decisions. At the same time, based on the sensitivity analysis, the importance of each input parameter is ranked to help users prioritize the adjustment of the parameters that have the greatest impact on the electricity fee, thereby improving decision-making efficiency, reducing unnecessary adjustments, and ultimately achieving accuracy and optimization in the electricity fee calculation. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A flow chart of a method for handling abnormal electricity charges based on intelligent verification provided in an embodiment of the present application.

[0020] Figure 2 A schematic diagram of the structure of an abnormal electricity fee error processing system based on intelligent verification provided in an embodiment of the present application. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] Embodiment 1, Figure 1 The schematic diagram of the process flow of the abnormal error handling method of electricity charges based on intelligent verification provided in the embodiment of the present application includes the following steps: S1, obtaining the first data of abnormal electricity charges, and performing abnormality detection on the electricity charge calculation based on the neural network algorithm, specifically: Acquire first data on abnormal electricity charges and historical data on abnormal electricity charges, wherein the first data on abnormal electricity charges include electricity price characteristics, power regulation standard characteristics, and fixed ratio characteristics, and the historical data on abnormal electricity charges include historical electricity price characteristics, historical power regulation standard characteristics, and historical fixed ratio characteristics; Data cleaning and data processing are performed on the abnormal data of historical electricity charges, specifically: Calculate the average of all data in the historical electricity fee abnormal data; Based on the clustering algorithm, the average of all data is set as the initial cluster center. According to the cosine similarity between the historical electricity price anomaly data and the initial cluster center, each data point is assigned to the initial cluster center to form a cluster. Recalculate the cluster center of each cluster and continue to assign until the iteration ends; The processed data is divided into a training set and a test set. Based on the neural network algorithm, the model is continuously trained through the training set and the test set is optimized to build an anomaly detection model. Inputting the first data of abnormal electricity charges into the abnormality detection model to obtain the output of the abnormality detection model; Calculate the standard deviation of the outputs of several anomaly detection models based on statistical methods; If the standard deviation exceeds the preset dispersion value, there is an abnormal electricity charge.

[0023] Electricity price characteristics refer to abnormal electricity prices, which lead to inaccurate electricity fee calculation. Analyzing electricity fee anomalies by obtaining electricity price characteristics has the following advantages: Discover problems promptly: Electricity price characteristics can quickly reveal abnormal fluctuations in electricity charges, helping users or businesses to promptly discover the causes of abnormal electricity charges and avoid unnecessary expenses.

[0024] Accurately identify the cause: By comparing the data of normal electricity prices with abnormal electricity prices, the root cause of price fluctuations can be found, such as changes in supply and demand in the electricity market, adjustments in the pricing strategy of power companies, or equipment failures.

[0025] Optimize electricity usage strategies: By understanding the fluctuation patterns of electricity prices, users can rationally plan electricity usage time and load, avoid using electricity during peak electricity price periods, thereby saving electricity bills and even performing demand response management.

[0026] The specific method for obtaining the electricity price characteristics is as follows: Based on time series analysis, set the time window for electricity price data; For each time point, take the data in the time window before and after the time point, calculate the arithmetic mean of these data as the smoothing value of the current point; Obtain the actual electricity price and calculate the deviation between the actual electricity price and the smoothed value; The standard deviation of the smoothed value is used as a reference value. If the deviation value exceeds three times the standard deviation of the smoothed value, the deviation value is quantified as an electricity price feature.

[0027] The specific calculation formula of the smoothing value is as follows:

[0028] The specific calculation formula of the electricity price characteristics is as follows:

[0029] In the formula, For time point The smoothed value of is the window size, For the The electricity price at a point in time, where =1,2,3,...,R, R is an integer, is the electricity price characteristic, For time point actual electricity price.

[0030] The power regulation standard feature refers to the error generated by the power regulation (load regulation) process. The power regulation standard feature is generally the error between the actual load and the theoretical load. By obtaining the power regulation standard feature to analyze the abnormal electricity charges, there are several advantages: Accurately identify the source of abnormality: The power regulation standard feature can help detect irregular fluctuations or abnormal loads in power usage, thereby helping to locate the root cause of abnormal electricity charges. This can reduce erroneous analysis caused by interference from other factors.

[0031] Early warning: By real-time monitoring of power regulation standard features, early warning can be issued before problems occur, helping businesses or households to promptly detect possible equipment failures or improper electricity use in electricity use, and avoid excessive electricity bills.

[0032] Help formulate energy-saving measures: The power regulation standard feature can provide data support for changes in power load, thereby analyzing which equipment or areas have unreasonable energy consumption, which helps to further optimize the power consumption structure and formulate energy-saving and emission reduction plans.

[0033] Improve electricity bill accuracy: Through in-depth analysis of power regulation standard characteristics, we can better understand the trends and patterns of power consumption, ensure the accuracy of electricity bill calculation, and avoid electricity bill anomalies caused by meter failure or measurement errors.

[0034] The specific method for obtaining the force adjustment standard feature is as follows: Obtain historical load data and initial load data of the current node, and calculate the average of the historical load data based on statistical methods; The actual load data after force adjustment is obtained, and the actual load data is analyzed with the average of the historical load data to obtain the force adjustment standard characteristics.

[0035] The specific calculation formula of the force adjustment standard characteristic is as follows:

[0036] In the formula, To adjust the standard features, is the actual load data, is the initial load data.

[0037] The fixed ratio feature refers to the abnormal ratio of the change in power consumption compared with historical data within a specific time range. This anomaly usually reflects the abnormal fluctuation of electricity charges, which may be caused by equipment failure, changes in usage habits, or power theft.

[0038] The specific method for obtaining the fixed ratio feature is as follows: The abnormal power consumption within the evaluation period is obtained, and the ratio feature is obtained by calculating the abnormal power consumption within the evaluation period with the preset safety power threshold. The preset safety power threshold is a normal power consumption threshold quantified by statistics of a large amount of historical power consumption data of the user and calculating its average.

[0039] The specific calculation formula of the fixed ratio feature is as follows:

[0040] In the formula, is a fixed ratio feature, is the actual power of the current node, The historical electricity.

[0041] The specific calculation formula of the anomaly detection model is as follows:

[0042] In the formula, is the abnormal detection value, is the electricity price characteristic, To adjust the standard features, is a fixed ratio feature, , , is the weight coefficient.

[0043] The outputs of several anomaly detection models are used to calculate their standard deviations based on statistical methods. The specific calculation formula is as follows:

[0044] In the formula, is the standard deviation of the anomaly detection model output, For the Anomaly detection values, For the The average of the anomaly detection values, where =1,2,3,...,R, R is an integer, and N is the total number of abnormal detection values.

[0045] It should be noted that the anomaly detection model can be understood as taking the abnormal impact of electricity price characteristics, power regulation standard characteristics and fixed ratio characteristics as input items, and outputting the detection value of the degree of abnormality causing the electricity fee calculation based on the neural network algorithm.

[0046] S2, when the electricity fee calculation is abnormal, the electricity quantity is corrected based on the power flow calculation method and energy balance algorithm, specifically: By actually testing and measuring the electric meter, the corrected reading and meter multiplication factor are obtained; Obtain the topological structure of the power grid, and analyze the topological structure of the power grid based on the power flow calculation method to obtain power data of each node in the power grid, wherein the power data includes current, input power, output power and power flow direction; The line loss of the transmission line is obtained according to the power loss formula based on the current, and the line loss is quantified to obtain the corrected line power; Obtain the input power and output power and compare them. If there is a deviation between the input power and the output power, the power withdrawal and subsidy will be corrected based on the power balance equation in combination with the power difference and line loss. The corrected indication, meter multiple, line power and refunded power are used to recalculate the user's actual power consumption based on the billing power calculation model.

[0047] The combined power difference and line loss power are used to correct the refund power based on the power balance equation, specifically: The objective function of the electricity withdrawal is defined by minimizing the difference between the actual measured electricity and the standard electricity, and the electricity balance equation is constructed based on the law of conservation of energy by combining the input electricity, output electricity and line loss electricity. The objective function of the power withdrawal and subsidy is solved based on the least squares method. The power withdrawal and subsidy amount is continuously updated by calculating the difference between the actual measured power and the corrected power withdrawal and subsidy amount until the objective function reaches the minimum value. The iteration is stopped and the optimal solution is the corrected power withdrawal and subsidy amount.

[0048] The specific calculation formula of the target function of the amount of electricity to be withdrawn and subsidized is as follows:

[0049] The specific calculation formula of the power balance equation is as follows:

[0050] The specific calculation formula of the billing power calculation model is as follows:

[0051] In the formula, is the corrected refund amount, For the The actual power of each measuring point, is the standard power, is the number of measurement points, where =1,2,3,...,R, R is an integer, is the input power, is the output power, is the line loss power, is the actual power, To correct the indication, To correct the meter rate, To correct the line power, To correct the amount of electricity refunded.

[0052] It should be noted that the power flow calculation method is a very important part of power system analysis. It uses mathematical models to calculate the current, voltage, power and other parameters in the power grid to analyze the operating status of the power system, which helps us estimate the power loss of each node and transmission line. The goal of the power balance equation is to ensure that the power input and output of the power grid are consistent while taking into account the losses. If there is a difference between the measured power and the theoretical power, it can be corrected by adjusting the withdrawal power. The least squares method is a commonly used optimization method to minimize the difference between actual measured data and predicted data. It can be used to adjust the withdrawal data of power to make it more consistent with the energy balance relationship of the power grid.

[0053] S3, by retrieving the user's historical electricity price information and performing consistency verification with the current electricity price information, determine and update the user's electricity price, and recalculate the electricity fee based on the revised electricity volume and electricity price, specifically: Obtaining user electricity price information, the user electricity price information including electricity consumption category, user classification, voltage level and time-of-use electricity price information; Retrieve the user's historical electricity price information, perform consistency verification with the current user electricity price information, and determine whether the user's electricity price information needs to be changed; If the verification results are inconsistent, the electricity price information selected by the user will be updated, and the electricity price selected by the user will be corrected according to the current electricity price standard; According to the revised electricity price and electricity consumption, the electricity fee, basic electricity fee, power regulation electricity fee and other electricity fees are calculated respectively, and the electricity fee calculation formula is regenerated.

[0054] The electricity fee calculation formula is as follows:

[0055] In the formula, The revised electricity rate is is the actual power, is the revised electricity price, is the maximum demand, is the power factor adjustment coefficient, For other electricity charges.

[0056] It should be noted that For electricity charges, For basic electricity charges, In order to adjust the electricity price, the electricity consumption category determines the demand characteristics of electricity users, such as residential electricity, commercial electricity or industrial electricity, and the electricity prices for different categories of users may be different. For example, industrial users may enjoy preferential electricity prices, while residential users may adopt tiered electricity prices. The electricity consumption category is an important basis for the selection of electricity prices. User classification: According to the nature of the user, the power company has different electricity price standards for different categories of users. Large customers usually customize electricity prices according to the contract, while ordinary users are charged according to the standard electricity price. Voltage level: The voltage level represents the scale and demand of the user's electricity consumption. Generally speaking, users with larger electricity consumption will use higher voltage levels (such as high voltage electricity). Different electricity prices will apply to users of different voltage levels. For example, the use of high voltage electricity is often accompanied by a lower unit electricity price because the power company can reduce the cost of transformation and transmission. Time-of-use electricity price information is charged according to different time periods (such as peak time, normal time, and valley time) to balance the load of the power grid and promote users to use electricity during low peak hours. Whether the user implements time-of-use electricity price will affect the calculation method of its electricity price. The user's electricity consumption behavior and electricity demand may change over time. For example, a user may have been a residential user, but later became a commercial user, and the power demand and price will change. Therefore, by retrieving and verifying the historical power price information, it can be ensured that the user's power price matches its current power consumption.

[0057] In order to ensure that the electricity price is correct, the revised electricity price needs to be verified: Check the user's electricity bill and historical records: Compare the revised electricity price calculation results with the user's historical electricity bill to ensure consistency. If there is a large difference, the process of electricity price revision needs to be further checked. User feedback and confirmation: If conditions permit, a new electricity price plan can be provided to the user, and their feedback can be solicited and confirmed. If the user has any objections, timely adjustments will be made.

[0058] S4, based on visualization technology, updates the three-color annotation display of the electricity charge correction calculation results in real time, obtains the input parameters of the electricity charge, obtains the contribution of the input parameters to the electricity charge based on the local sensitivity method, and sorts the input parameters by importance according to the contribution.

[0059] The three-color marking and detailed calculation process display make the electricity fee calculation results and parameter changes clear at a glance, which is convenient for business personnel to check and verify and for customers to understand. By analyzing the changes in multiple input parameters, it is possible to identify which parameters have the greatest impact on the electricity fee calculation and which parameter changes have a smaller impact on the electricity fee results. This can help users optimize their decisions, give priority to adjusting parameters that have a greater impact on electricity fees, and reduce unnecessary adjustments.

[0060] Based on visualization technology, the electricity fee correction calculation results are updated in real time with three-color annotation display, specifically: Use red, yellow, and blue to guide users and clearly display calculation results; The parameters marked in blue can be changed and serve as input values ​​for the electricity fee calculation formula; Corrected parameters are marked in red and displayed side by side with the original data for easy comparison by users; Yellow represents the output result value, which is displayed side by side with the original data to accurately show the changes in the associated results; At the same time, according to the electricity fee composition structure, which includes electricity fee, basic electricity fee, power regulation electricity fee and other electricity fees, the calculation process after the modified parameters is listed in detail and supports verification by business personnel.

[0061] It should be noted that users can clearly see the electricity fee calculation process, the corrected data comparison and the final calculation results in an intuitive and easy-to-use interface, and can also efficiently perform parameter correction and verification. The color marking (red, yellow, and blue) provides a clear visual hierarchy, which helps users quickly locate and understand each part of the content. The original data can be understood as the original parameter data without correction.

[0062] The input parameters of the electricity fee are obtained, and the contribution of the input parameters to the electricity fee is obtained based on the local sensitivity method, and the input parameters are sorted by importance according to the contribution, specifically: Acquire input parameters affecting electricity charges, the input parameters including electricity price, electricity quantity, basic electricity charge and power regulation electricity charge; For all parameters in the input parameters, the between-group variance and total variance of all parameters were calculated based on local sensitivity analysis; Compare and analyze the inter-group variance and the total variance, and obtain the variance contribution of all parameters in the input parameters based on the comparison results; The input parameters of electricity charges are ranked in order of importance according to the variance contribution of all parameters from large to small; The larger the variance contribution, the more important the change in this parameter is in affecting the electricity bill.

[0063] It should be noted that the local sensitivity analysis can be understood as analyzing the impact of changes in each input parameter within a certain range on the electricity fee, that is, only one input parameter is changed each time to analyze its impact on the electricity fee output.

[0064] Embodiment 2, Figure 2 The schematic diagram of the structure of the electricity fee anomaly error processing system based on intelligent verification provided in the embodiment of the present application includes an electricity fee anomaly detection module, an electricity fee calculation correction module, an electricity price and electricity fee calculation correction module, and a visualization and importance analysis module. There are connections between the modules: An electricity charge anomaly detection module, used to obtain first data of electricity charge anomaly, and perform anomaly detection on electricity charge calculation based on a neural network algorithm, wherein the first data of electricity charge anomaly includes electricity price characteristics, power regulation standard characteristics, and fixed ratio characteristics; The power calculation correction module is used to correct the power based on the power flow calculation method and energy balance algorithm when the electricity fee calculation is abnormal; The electricity price and electricity fee calculation and correction module is used to retrieve the user's historical electricity price information and verify the consistency with the current electricity price information, determine and update the user's electricity price, and recalculate the electricity fee based on the corrected electricity quantity and electricity price; The visualization and importance analysis module is used to update the three-color annotation display of the electricity charge correction calculation results in real time based on visualization technology, obtain the input parameters of the electricity charge, obtain the contribution of the input parameters to the electricity charge based on the local sensitivity method, and sort the importance of the input parameters according to the contribution.

[0065] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0066] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0067] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0068] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0069] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0070] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. The method for handling abnormal electricity charges based on intelligent verification is characterized in that: The steps include: Acquire first data of abnormal electricity charges, and perform abnormal detection on electricity charge calculation based on a neural network algorithm, wherein the first data of abnormal electricity charges includes electricity price characteristics, power regulation standard characteristics, and fixed ratio characteristics; When the electricity fee calculation is abnormal, the electricity quantity is corrected based on the power flow calculation method and energy balance algorithm; By retrieving the user's historical electricity price information and performing consistency verification with the current electricity price information, the user's electricity price is determined and updated, and the electricity fee is recalculated based on the revised electricity volume and electricity price; Based on visualization technology, the three-color annotation display of the electricity charge correction calculation results is updated in real time, and the input parameters of the electricity charge are obtained. Based on the local sensitivity method, the contribution of the input parameters to the electricity charge is obtained, and the input parameters are ranked in importance according to the contribution.

2. The method for handling abnormal electricity charges based on intelligent verification according to claim 1, characterized in that: The obtaining of the first data of abnormal electricity charges and performing abnormality detection on the electricity charge calculation based on the neural network algorithm is specifically as follows: Acquire historical electricity fee anomaly data, wherein the historical electricity fee anomaly data includes historical electricity price characteristics, historical power regulation standard characteristics, and historical fixed ratio characteristics; All data in the historical electricity fee abnormal data are averaged separately; Based on the clustering algorithm, the average of all data is set as the initial cluster center. According to the cosine similarity between the historical electricity price anomaly data and the initial cluster center, each data point is assigned to the initial cluster center to form a cluster. Recalculate the cluster center of each cluster and continue to assign until the iteration ends; The processed data is used to build an anomaly detection model based on a neural network algorithm; Inputting the first data of abnormal electricity charges into an anomaly detection model, obtaining outputs of several anomaly detection models and calculating standard deviations based on statistical methods; If the standard deviation exceeds the preset dispersion value, there is an abnormal electricity charge.

3. The method for handling abnormal electricity charges based on intelligent verification according to claim 1, characterized in that: When the electricity fee calculation is abnormal, the electricity quantity is corrected based on the power flow calculation method and the energy balance algorithm, specifically: By actually testing and measuring the electric meter, the corrected reading and meter multiplication factor are obtained; Obtain the topological structure of the power grid, and analyze the topological structure of the power grid based on the power flow calculation method to obtain power data of each node in the power grid, wherein the power data includes current, input power, output power and power flow direction; The line loss of the transmission line is obtained according to the power loss formula based on the current, and the line loss is quantified to obtain the corrected line power; Obtain the input power and output power and compare them. If there is a deviation between the input power and the output power, the power withdrawal and subsidy will be corrected based on the power balance equation in combination with the power difference and line loss. The corrected indication, meter multiple, line power and refunded power are used to recalculate the user's actual power consumption based on the billing power calculation model.

4. The method for handling abnormal electricity charges based on intelligent verification according to claim 3 is characterized in that: The combined power difference and line loss power are used to correct the refund power based on the power balance equation, specifically: The objective function of the electricity withdrawal is defined by minimizing the difference between the actual measured electricity and the standard electricity, and the electricity balance equation is constructed based on the law of conservation of energy by combining the input electricity, output electricity and line loss electricity. The objective function of the power withdrawal and subsidy is solved based on the least squares method. The power withdrawal and subsidy amount is continuously updated by calculating the difference between the actual measured power and the corrected power withdrawal and subsidy amount until the objective function reaches the minimum value. The iteration is stopped and the optimal solution is the corrected power withdrawal and subsidy amount.

5. The method for handling abnormal electricity charges based on intelligent verification according to claim 1, characterized in that: The method of retrieving the user's historical electricity price information and performing consistency verification with the current electricity price information, determining and updating the user's electricity price, and recalculating the electricity fee based on the revised electricity quantity and electricity price is as follows: Obtaining user electricity price information, the user electricity price information including electricity consumption category, user classification, voltage level and time-of-use electricity price information; Retrieve the user's historical electricity price information, perform consistency verification with the current user electricity price information, and determine whether the user's electricity price information needs to be changed; If the verification results are inconsistent, the electricity price information selected by the user will be updated, and the electricity price selected by the user will be corrected according to the current electricity price standard; According to the revised electricity price and electricity consumption, the electricity fee, basic electricity fee, power regulation electricity fee and other electricity fees are calculated respectively, and the electricity fee calculation formula is regenerated.

6. The method for handling abnormal electricity charges based on intelligent verification according to claim 1, characterized in that: The visualization technology is used to update the three-color annotation display of the electricity fee correction calculation results in real time, specifically: Use red, yellow, and blue to guide users and clearly display calculation results; The parameters marked in blue can be changed and serve as input values ​​for the electricity fee calculation formula; Corrected parameters are marked in red and displayed side by side with the original data for easy comparison by users; Yellow represents the output result value, which is displayed side by side with the original data to accurately show the changes in the associated results; According to the electricity fee composition structure, the electricity fee composition structure includes electricity fee, basic electricity fee, power regulation electricity fee, and other electricity fees, and the calculation process after the correction parameters are listed in detail.

7. The method for handling abnormal electricity charges based on intelligent verification according to claim 1, characterized in that: The input parameters of the electricity charge are obtained, and the contribution of the input parameters to the electricity charge is obtained based on the local sensitivity method. The input parameters are ranked in importance according to the contribution, specifically: Acquire input parameters affecting electricity charges, the input parameters including electricity price, electricity quantity, basic electricity charge and power regulation electricity charge; For all parameters in the input parameters, the between-group variance and total variance of all parameters were calculated based on local sensitivity analysis; Compare and analyze the inter-group variance and the total variance, and obtain the variance contribution of all parameters in the input parameters based on the comparison results; The input parameters of electricity charges are ranked in order of importance according to the variance contribution of all parameters from large to small.

8. The method for handling abnormal electricity charges based on intelligent verification according to claim 2, characterized in that: The specific method for obtaining the electricity price characteristics is as follows: Based on time series analysis, set the time window for electricity price data; For each time point, take the data in the time window before and after the time point, calculate the arithmetic mean of these data as the smoothing value of the current point; Obtain the actual electricity price and calculate the deviation between the actual electricity price and the smoothed value; The standard deviation of the smoothed value is used as a reference value. If the deviation value exceeds three times the standard deviation of the smoothed value, the deviation value is quantified as an electricity price feature.

9. The method for handling abnormal electricity charges based on intelligent verification according to claim 8, characterized in that: The specific calculation formula of the smoothing value is as follows: The specific calculation formula of the electricity price characteristics is as follows: In the formula, For time point The smoothed value of is the window size, For the The electricity price at a point in time, where =1,2,3,...,R, R is an integer, is the electricity price characteristic, For time point actual electricity price.

10. A system using the method for handling abnormal electricity charges based on intelligent verification as claimed in any one of claims 1 to 9, characterized in that: It includes an electricity fee anomaly detection module, an electricity fee calculation and correction module, an electricity price and electricity fee calculation and correction module, and a visualization and importance analysis module. There are connections between the modules: An electricity charge anomaly detection module, used to obtain first data of electricity charge anomaly, and perform anomaly detection on electricity charge calculation based on a neural network algorithm, wherein the first data of electricity charge anomaly includes electricity price characteristics, power regulation standard characteristics, and fixed ratio characteristics; The power calculation correction module is used to correct the power based on the power flow calculation method and energy balance algorithm when the electricity fee calculation is abnormal; The electricity price and electricity fee calculation and correction module is used to retrieve the user's historical electricity price information and verify the consistency with the current electricity price information, determine and update the user's electricity price, and recalculate the electricity fee based on the corrected electricity quantity and electricity price; The visualization and importance analysis module is used to update the three-color annotation display of the electricity charge correction calculation results in real time based on visualization technology, obtain the input parameters of the electricity charge, obtain the contribution of the input parameters to the electricity charge based on the local sensitivity method, and sort the importance of the input parameters according to the contribution.

Images