Power industry assessment index auditing verification method based on similarity analysis

Abstract

The embodiment of the invention provides a method for auditing and checking power industry assessment indexes based on similarity analysis, and belongs to the technical field of power index checking assistance. The method comprises the following steps: acquiring power consumption data of all large industrial electricity price type power supply users in a to-be-audited year; acquiring corresponding electricity utilization information according to the electricity utilization data; screening out to-be-identified large industrial electricity price type power supply users according to the electricity utilization information of all the large industrial electricity price type power supply users; acquiring a same-period similarity data set of the to-be-identified large industrial electricity price type power supply users; and screening target doubtful users according to the same-period similarity data set of the to-be-identified large industrial electricity price type power supply users. According to the method, the target doubtful users are screened by adopting the similarity data set, so that the target doubtful users can be efficiently and accurately screened out, subsequent work of auditing and checking is facilitated, and the auditing and checking efficiency and effect of the assessment indexes in the power industry are effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power index verification assistance, and in particular to a method for auditing and verifying power industry assessment indexes based on similarity analysis. Background Art

[0002] In order to adjust the assessment indicators, some units did not record the meter readings according to the facts and adjusted the electricity sales data. This means that in order to cope with the assessment and pursue short-term performance, some power supply units did not record the meter readings according to the facts and artificially adjusted the electricity sales data. This behavior may lead to the distortion of the enterprise's electricity sales information, affect the company's financial situation and operating results, and affect the accuracy of decision-making. It is a link that requires strict attention and review in marketing audits. By strengthening audit supervision and establishing a sound transcription system, relevant units can be encouraged to rectify the problems found, standardize transcription and electricity sales behaviors, and ensure the accuracy and fairness of assessment indicators.

[0003] At present, the recording of abnormal values ​​of electricity sales of some units can only be judged by the experience of auditors during the audit process. However, due to the large number of audit verifications, this method requires high ability of auditors and has low audit verification efficiency.

[0004] The inventors of the present application discovered during the process of implementing the present invention that the above-mentioned scheme in the prior art has the defect of low audit verification efficiency. Summary of the invention

[0005] The purpose of the embodiment of the present invention is to provide a method for auditing and verifying the assessment indicators of the electric power industry based on similarity analysis, and the method for auditing and verifying the assessment indicators of the electric power industry based on similarity analysis has the function of high auditing and verification efficiency.

[0006] In order to achieve the above purpose, an embodiment of the present invention provides a method for auditing and verifying assessment indicators of the power industry based on similarity analysis, comprising:

[0007] Obtain electricity consumption data of all large industrial electricity price power supply users in the year to be audited;

[0008] Acquiring corresponding electricity usage information according to the electricity usage data;

[0009] Filter out the large industrial electricity price power supply users to be identified based on the electricity consumption information of all large industrial electricity price power supply users;

[0010] Obtain the similarity data set of the large industrial electricity price power supply users to be identified in the same period;

[0011] The target suspicious users are screened based on the contemporaneous similarity dataset of the large industrial electricity price power supply users to be identified.

[0012] Optionally, the electricity consumption data of all large industrial electricity price power supply users in the audit year include:

[0013] Determine whether the operation period of the large industrial electricity price power supply user is greater than or equal to the preset period;

[0014] When it is determined that the operation period of the large industrial electricity price power supply user is greater than or equal to the preset period, obtaining the electricity consumption data of the large industrial electricity price power supply user;

[0015] When it is determined that the operation period of the large industrial electricity price type power supply user is less than the preset period, the acquisition of the large industrial electricity price type power supply user is abandoned.

[0016] Optionally, acquiring corresponding electricity usage information according to the electricity usage data includes:

[0017] According to formula (1), the annual proportion of electricity consumption of the large industrial electricity price power supply users in each month in the audit year is obtained.

[0018]

[0019] Among them, μ jk is the annual proportion of electricity consumption of the jth large industrial electricity price power supply user in the kth month of the audit year, w jk is the electricity consumption of the jth large industrial electricity price power supply user in the kth month of the audit year, w j is the annual electricity consumption of the jth large industrial electricity price power supply user in the audit year, where j is an integer number and k is an integer number;

[0020] According to formula (2), the annual proportion of electricity consumption of the large industrial electricity price power supply users in each month of the year before the audit year is obtained.

[0021]

[0022] Among them, μ jk ′ is the annual proportion of electricity consumption of the jth large industrial electricity price power supply user in the kth month of the year before the audit year, w jk ′ is the electricity consumption of the jth large industrial electricity price power supply user in the kth month of the year before the audit year, w j ′ It is the annual electricity consumption of the jth large industrial electricity price category power supply user in the year before the year to be audited.

[0023] Optionally, screening out large industrial electricity price class power supply users to be identified based on electricity consumption information of all large industrial electricity price class power supply users includes screening using an isolation forest algorithm.

[0024] Optionally, obtaining a contemporaneous similarity dataset of large industrial electricity price power supply users to be identified includes:

[0025] The annual proportion of electricity consumption of each month of each large industrial electricity price power supply user in the audit year and the annual proportion of electricity consumption of each month in the year before the audit year are used to construct multiple monthly electricity consumption data groups;

[0026] Obtain a Euclidean distance set based on multiple monthly electricity consumption data sets of large industrial electricity price power supply users;

[0027] The Euclidean distance set is used as a contemporaneous similarity data set.

[0028] Optionally, obtaining the Euclidean distance set according to multiple monthly electricity consumption data groups of large industrial electricity price power supply users includes:

[0029] Construct a multidimensional space coordinate system, and map the values ​​in the electricity consumption data group of each month of the large industrial electricity price power supply users to the multidimensional space coordinate system;

[0030] According to formula (3), the Euclidean distance of each monthly electricity consumption data group is obtained:

[0031]

[0032] Wherein, d is the Euclidean distance of the electricity consumption data set, N is the spatial dimension, i is an integer number, and X i1 is the coordinate value of the mapping point X1 in the i-th dimension in the multi-dimensional space coordinate system of the median of the power consumption data set, X i2 The coordinate value of the mapping point X2 in the i-th dimension in the multi-dimensional space coordinate system of the median of the power consumption data set;

[0033] The Euclidean distance of each monthly electricity consumption data group of each large industrial electricity price power supply user is aggregated to obtain the Euclidean distance set.

[0034] Optionally, the target suspicious users are screened according to the similarity data set of the large industrial electricity price power supply users to be identified in the same period, including:

[0035] Using a clustering algorithm to cluster each value in the contemporaneous similarity data set, and obtaining a cluster center and a cluster area;

[0036] Target suspicious users are screened out according to the clustering area.

[0037] Optionally, screening target suspicious users according to the contemporaneous similarity data set of large industrial electricity price power supply users to be identified also includes:

[0038] According to formula (4), the threshold of the target suspicious user is obtained;

[0039] C=J·α, (4)

[0040] Among them, C is the threshold of the target suspicious user, J is the number of large industrial electricity price power supply users, and α is the preset selection ratio of the target suspicious users;

[0041] Determine whether the target suspicious users screened out by the cluster area are greater than or equal to the threshold of the target suspicious users;

[0042] In the case where it is determined that the target suspicious users screened out in the cluster area are greater than or equal to the threshold of the target suspicious users, selecting the target suspicious users of the threshold number from the target suspicious users screened out in the cluster area;

[0043] When it is determined that the number of target suspicious users screened out by the clustering area is less than the threshold of target suspicious users, all target suspicious users screened out by the clustering area are acquired.

[0044] On the other hand, the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, wherein the instructions are used to be read by a machine so that the machine executes any of the above methods.

[0045] Through the above technical scheme, the method for auditing and verifying the assessment indicators of the power industry based on similarity analysis provided by the present invention obtains the electricity consumption data of all large industrial electricity price-type power supply users in the year to be audited, and obtains the corresponding electricity consumption information based on the electricity consumption data, and then screens out the large industrial electricity price-type power supply users to be identified based on the electricity consumption information, that is, a preliminary screening is performed, and for the large industrial electricity price-type power supply users to be identified, their similarity data sets of the same period are obtained, and the final target suspicious users can be screened out based on the similarity data sets of the same period; the method of using the similarity data sets to screen the target suspicious users can efficiently and accurately screen out the target suspicious users, so as to facilitate the subsequent work of auditing and verification, and effectively improve the efficiency and effect of the auditing and verification of the assessment indicators of the power industry.

[0046] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

[0048] Figure 1 It is a flow chart of a method for auditing and verifying assessment indicators of the electric power industry based on similarity analysis according to an embodiment of the present invention;

[0049] Figure 2 It is a flowchart of obtaining electricity consumption data in a method for auditing and verifying assessment indicators of the power industry based on similarity analysis according to an embodiment of the present invention;

[0050] Figure 3 It is a flowchart of screening electricity consumption information in a method for auditing and verifying assessment indicators of the power industry based on similarity analysis according to an embodiment of the present invention;

[0051] Figure 4 It is a flow chart of obtaining electricity consumption information in a method for auditing and verifying assessment indicators of the power industry based on similarity analysis according to an embodiment of the present invention;

[0052] Figure 5 It is a flowchart of obtaining a contemporaneous similarity data set in a method for auditing and verifying electric power industry assessment indicators based on similarity analysis according to an embodiment of the present invention;

[0053] Figure 6 It is a flowchart of obtaining the Euclidean distance of a power consumption data group in a method for auditing and verifying power industry assessment indicators based on similarity analysis according to an embodiment of the present invention;

[0054] Figure 7 It is a flowchart of screening target suspicious users in a method for audit verification of assessment indicators in the power industry based on similarity analysis according to an embodiment of the present invention;

[0055] Figure 8 This is an example diagram of anomaly detection using an isolation forest in a method for audit verification of electric power industry assessment indicators based on similarity analysis according to an embodiment of the present invention;

[0056] Fig. 9 This is an example diagram of anomaly detection using Euclidean distance in a method for audit verification of electric power industry assessment indicators based on similarity analysis according to an embodiment of the present invention. DETAILED DESCRIPTION

[0057] The specific implementation of the embodiment of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the embodiment of the present invention, and is not used to limit the embodiment of the present invention.

[0058] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application are in compliance with the relevant provisions of national laws and regulations. In the embodiments of this application, some existing solutions in the industry such as certain software, components, and models may be mentioned, which should be considered as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of this application, but it does not mean that the applicant has or will necessarily use the solution.

[0059] Figure 1 The flowchart of the method for auditing and verifying the assessment indicators of the power industry based on similarity analysis according to one embodiment of the present invention. Figure 1 The audit and verification methods for the power industry assessment indicators may include:

[0060] In step S10, the electricity consumption data of all large industrial electricity price power supply users in the audit year are obtained. Among them, the large industrial electricity price power supply data is a part of the power industry with a large demand for electricity, that is, the change in electricity value is more obvious, so the electricity consumption of the large industrial electricity price class is used as the electricity consumption category for audit verification. Specifically, the acquisition of electricity consumption data of large industrial electricity price power supply users can be as follows: Figure 2 Specifically, Figure 2 In the step of acquiring the electricity consumption data, the step of acquiring the electricity consumption data may include:

[0061] In step S100, it is determined whether the operating period of the large industrial electricity price power supply users is greater than or equal to the preset period. Considering that the power consumption of large industrial users is related to the scale and period of operation, a preliminary judgment needs to be made on the operating period of large industrial electricity price power supply users. Specifically, large industrial electricity price power supply users that have reached full production and have been in stable operation for more than 3 years are generally used to judge the power consumption data to eliminate the errors caused by the large changes in output in the first few years of production, that is, the preset period includes greater than or equal to 3 years.

[0062] In step S101, when it is determined that the operation period of the large industrial electricity price power supply user is greater than or equal to the preset period, the power consumption data of the large industrial electricity price power supply user is obtained. Among them, if the large industrial electricity price power supply user has been in operation for 3 years or more, the power consumption data of the large industrial electricity price power supply user can be obtained.

[0063] In step S102, if it is determined that the operation period of the large industrial electricity price power supply user is less than the preset period, the acquisition of the large industrial electricity price power supply user is abandoned. If the large industrial electricity price power supply user has been in operation for less than 3 years, the power consumption data of the large industrial electricity price power supply user is deleted or abandoned.

[0064] In step S100 to step S102, it is fully considered that the scale of large industrial electricity price users gradually changes in the first few years of production, and there are large fluctuations in electricity consumption. Therefore, in order to eliminate the error in the production years, the electricity consumption data of large industrial electricity price users whose production years meet the production years requirements are obtained, and large industrial electricity price users whose production years do not meet the production years requirements are removed. This method can reduce the interference of the production years on the screening results.

[0065] In step S11, the corresponding electricity consumption information is obtained according to the electricity consumption data. After obtaining the electricity consumption data of large industrial electricity price power supply users whose commissioning / operation period meets the preset period, the electricity consumption data can be converted to obtain the electricity consumption information of large industrial electricity price power supply users. Specifically, the electricity consumption information may include: user number, user name, address, electricity price category (large industrial electricity price category), commissioning date, electricity consumption in each month, and the proportion of electricity consumption in each month throughout the year.

[0066] In step S12, the large industrial electricity price class power supply users to be identified are screened out based on the electricity consumption information of all large industrial electricity price class power supply users. Among them, the electricity consumption information screening of large industrial electricity price class power supply users can be carried out in an isolation forest / isolation forest manner. The isolation forest can mine and analyze the isolated outliers of abnormal electricity consumption in the electricity consumption data of large industrial electricity price class power supply users for multiple months to identify key abnormal electricity consumption conditions. Specifically, the step of screening the electricity consumption information of large industrial electricity price class power supply users by the isolation forest can be as follows: Figure 3 Specifically, Figure 3 In the present invention, the screening step may include:

[0067] In step S120, the power consumption information of large industrial power price power users is extracted through the isolation forest algorithm to obtain change characteristics, which may include power consumption change characteristics, long-term trend change characteristics, and periodic (monthly) change characteristics.

[0068] In step S121, the number of trees and the capacity of each sub-sample of the power consumption change isolation forest algorithm are set according to the power consumption change characteristics, the long-term trend change characteristics and the periodic change characteristics. This step may further include:

[0069] 1. Randomly select one of the electricity consumption information data sets of large industrial electricity price power supply users and plasticity, such as the proportion of electricity consumption in each month throughout the year;

[0070] 2. Randomly select a value of this attribute and calculate the percentage of electricity consumption in each month throughout the year;

[0071] 3. Classify each record according to the attribute of the annual proportion of electricity consumption of each month. Put the records whose annual proportion of electricity consumption of each month is less than the annual proportion value of electricity consumption of each month in the left child node, and put the records whose annual proportion of electricity consumption of each month is greater than or equal to the annual proportion value of electricity consumption of each month in the right child node;

[0072] 4. Then recursively construct the left child node and the right child node until the condition is met: the incoming data set has only one record or multiple identical records;

[0073] 5. The height of the tree has reached the specified height.

[0074] In step S122, based on the electricity consumption change characteristics and the isolation forest algorithm with set parameters, the electricity consumption information mutation characteristics, long-term trend change characteristics and periodic change characteristics of large industrial electricity price power supply users are isolated and analyzed to obtain the electricity consumption information anomaly score level of large industrial electricity price power supply users.

[0075] In step S13, a contemporaneous similarity data set of large industrial electricity price power supply users to be identified is obtained. The contemporaneous similarity data of large industrial electricity price power supply users to be identified can be obtained by using the Euclidean distance method.

[0076] In step S14, the target suspicious user is screened according to the contemporaneous similarity data set of the large industrial electricity price class power supply users to be identified. After obtaining the contemporaneous similarity data set of the large industrial electricity price class power supply users to be identified, the target suspicious user can be screened according to the contemporaneous similarity data set.

[0077] In step S10 to step S14, the electricity consumption data of all large industrial electricity price power supply users in the audit year are first obtained, and the corresponding electricity consumption information can be obtained based on the electricity consumption data. The isolation / isolation forest algorithm is used to preliminarily screen the electricity consumption information of all large industrial electricity price power supply users to obtain the large industrial electricity price power supply users to be identified. Then, based on the electricity consumption information of the large industrial electricity price power supply users to be identified, a similarity data set for the same period is obtained, and the target suspicious users are screened out using the similarity data set for the same period, which can facilitate subsequent audit verification work.

[0078] In the traditional audit process, judgment can only be made based on the experience of the auditors. However, due to the large number of audit verifications, this method requires high capabilities of auditors and has low audit verification efficiency. In this embodiment of the present invention, the method of using similarity data sets to screen target suspicious users can efficiently and accurately screen out target suspicious users to facilitate subsequent audit verification work, effectively improving the efficiency and effect of audit verification of assessment indicators in the power industry.

[0079] In this embodiment of the present invention, after obtaining the power consumption data of large industrial power price power supply users, the power consumption data can be converted to obtain power consumption information. The specific conversion steps can be as follows: Figure 4 Specifically, Figure 4 In the example, the conversion step may include:

[0080] In step S110, the annual proportion of electricity consumption of large industrial electricity price power supply users in each month in the audit year is obtained according to formula (1):

[0081]

[0082] Among them, μ jk is the annual proportion of electricity consumption of the jth large industrial electricity price power supply user in the kth month of the audit year, w jk is the electricity consumption of the jth large industrial electricity price power supply user in the kth month of the audit year, w j It is the annual electricity consumption of the jth large industrial electricity price category power supply user in the audit year, where j is an integer number and k is an integer number.

[0083] In step S111, the annual proportion of electricity consumption of large industrial electricity price power supply users in each month of the previous year of the audit year is obtained according to formula (2):

[0084]

[0085] Among them, μ jk ′ is the annual proportion of electricity consumption of the jth large industrial electricity price power supply user in the kth month of the year before the audit year, w jk ′ is the electricity consumption of the jth large industrial electricity price power supply user in the kth month of the year before the audit year, w j ′ It is the annual electricity consumption of the j-th large industrial electricity price category power supply user in the year before the audit year.

[0086] In this embodiment of the present invention, when the large industrial electricity price class power supply users to be identified are screened out, the large industrial electricity price class power supply users to be identified can be further screened, that is, the similarity data set of the same period is obtained. The specific acquisition steps can be as follows: Figure 5 Specifically, Figure 5 In the step of acquiring the contemporaneous similarity dataset, the step of acquiring the contemporaneous similarity dataset may include:

[0087] In step S130, multiple monthly electricity consumption data groups are constructed by combining the annual proportion of electricity consumption of each month of each large industrial electricity price power supply user in the audit year with the annual proportion of electricity consumption of each month in the year before the audit year.

[0088] In step S131, a Euclidean distance set is obtained based on multiple monthly electricity consumption data sets of large industrial electricity price power supply users. The Euclidean distance calculation for each monthly electricity consumption data set may include the following: Figure 6 Specifically, Figure 6 In the calculation step, the step may include:

[0089] In step S1310, a multidimensional space coordinate system is constructed, and the values ​​in the electricity consumption data group of each month of the large industrial electricity price class power supply users are mapped to the multidimensional space coordinate system.

[0090] In step S1311, the Euclidean distance of each monthly electricity consumption data group is obtained according to formula (3):

[0091]

[0092] Among them, d is the Euclidean distance of the electricity consumption data set, N is the spatial dimension, i is an integer number, and X i1 is the coordinate value of the point X1 in the i-th dimension mapped to the median of the electricity consumption data set in the multidimensional space coordinate system, X i2 The coordinate value of point X2 in the i-th dimension is mapped to the median of the electricity consumption data set in the multidimensional space coordinate system. Specifically, Euclidean distance is often used to measure the similarity or difference between data points. By calculating the Euclidean distance between different data points, auditors can quickly identify abnormal data that deviates significantly from the normal pattern, thereby promptly discovering potential risks or fraudulent behavior. Specifically, Isolation Forest has identified key abnormal electricity consumption situations, and Euclidean distance can once again demonstrate abnormal data that deviates significantly from the normal pattern.

[0093] In step S1312, the Euclidean distances of each monthly electricity consumption data group of each large industrial electricity price class power supply user are aggregated to obtain a Euclidean distance set.

[0094] In step S132, the Euclidean distance set is used as the contemporaneous similarity data set.

[0095] In step S130 to step S132, the annual proportion of electricity consumption of each large industrial electricity price power supply user in each month of the audit year and the annual proportion of the same period are taken as a group to construct the electricity consumption data group of each month. Then, the corresponding Euclidean distance is calculated based on the electricity consumption data group of each month of the large industrial electricity price power supply user, and the Euclidean distance set is summarized, and the Euclidean distance set is used as the similarity data set for the same period. By obtaining the Euclidean distance, abnormal users can be further screened out, and the accuracy of the audit verification is further improved.

[0096] In this embodiment of the present invention, after obtaining the contemporaneous similarity data set of the large industrial electricity price power supply users to be identified, the target suspicious users can be screened according to the contemporaneous similarity data set. The specific screening steps can be as follows: Figure 7 Specifically, Figure 7 In the present invention, the screening step may include:

[0097] In step S140, a clustering algorithm is used to cluster each value in the contemporaneous similarity data set, and the cluster center and cluster area are obtained. The identification of the cluster area can be obtained by presetting a distance threshold, and the points / values ​​outside the cluster area are also outliers (target suspicious users).

[0098] In step S141, target suspicious users are screened out according to the clustering areas.

[0099] In step S142, the threshold of the target suspicious user is obtained according to formula (4);

[0100] C=J·α, (4)

[0101] Wherein, C is the threshold of the target suspicious user, J is the number of large industrial electricity price power supply users, and α is the preset selection ratio of the target suspicious user. Specifically, the preset selection ratio for the target suspicious user may include 5%.

[0102] In step S143, it is determined whether the target suspicious users screened out by the cluster area are greater than or equal to the threshold of the target suspicious users.

[0103] In step S144, when it is determined that the target suspicious users screened out in the cluster area are greater than or equal to the threshold of the target suspicious users, a threshold number of target suspicious users are selected from the target suspicious users screened out in the cluster area. If the target suspicious users outside the cluster area are greater than or equal to the threshold of the target suspicious users, it means that there are many target suspicious users detected by the cluster, and multiple target suspicious users with the farthest distance can be selected according to the threshold number of target suspicious users.

[0104] In step S145, when it is determined that the number of target suspicious users screened out by the clustering area is less than the target suspicious user threshold, all target suspicious users screened out by the clustering area are obtained. If the number of target suspicious users outside the clustering area is less than the target suspicious user threshold, it means that the number of target suspicious users detected by the clustering is small, and all target suspicious users outside the clustering area can be selected.

[0105] In step S140 to step S145, a clustering algorithm is first used to perform a clustering operation on each value in the similarity data set of the same period, and then the cluster center and the cluster area can be obtained. Points outside the cluster area are obtained, that is, the target suspicious users screened out by the cluster area. At the same time, according to the preset selection ratio of the target suspicious users, the threshold of the target suspicious users is calculated, and the threshold is compared with the target suspicious users screened out by the cluster area. If the threshold is larger, all target suspicious users outside the cluster area are selected, otherwise multiple target suspicious users with the farthest distance outside the cluster area are selected according to the threshold. In this way, the target suspicious users can be effectively and accurately screened out, reducing the influence of interference from partial values ​​during cluster screening, and in the case of a large amount of data, the target suspicious users with the highest probability can also be streamlined. Specifically, the present invention uses the isolation forest algorithm to identify outliers on the electricity consumption information group label of each large industrial electricity price class power supply user, and obtains the large industrial electricity price class power supply users to be identified, and then uses the Euclidean distance method to screen the large industrial electricity price class power supply users with low similarity (high distance value) as "suspicious large industrial electricity price class power supply users", that is, the target suspicious users.

[0106] In this embodiment of the present invention, for example, the output parameters of the large industrial electricity price class power supply user map to be identified may include: number of points: 12296, number of base estimators (i.e., decision trees): 100, parameter for controlling the proportion of abnormal points: 0.01, Figure 8 This is an example diagram of an anomaly detection algorithm based on isolation / isolation forest. Figure 8 It can be seen that the large industrial electricity price power supply users associated with abnormalpoints have audit doubt information, but it takes further processing to obtain the actual series of audit doubt information. Specifically, some data of the audit doubt information can be shown in Table 1.

[0107] Table 1 Partial data of audit doubt information

[0108]

[0109]

[0110] In the power marketing audit scenario of checking whether to adjust the assessment indicators, the Euclidean distance algorithm can be used to analyze user electricity consumption behavior, detect abnormal electricity consumption patterns, etc. By calculating the Euclidean distance between different users' electricity consumption data, the similarity of user electricity consumption behavior can be analyzed to identify potential abnormal users or electricity consumption patterns. The data dimension adds the annual proportion of electricity consumption in November and December for each major industrial electricity price power supply user. Fig. 9 This is an example diagram of an anomaly detection algorithm based on the Euclidean distance algorithm.

[0111] After obtaining the abnormal points according to the Euclidean distance, the information of the abnormal large industrial electricity price power supply users is output, as shown in Table 2.

[0112] Table 2 List of target suspicious large industrial electricity price power supply users

[0113]

[0114]

[0115] In the above-mentioned list of target suspicious large industrial electricity price power supply users, these large industrial electricity price power supply users with low similarity (high distance value) as suspicious users can help auditors identify key transaction objects, discover potential risk points and optimize audit strategies, thereby improving the efficiency and accuracy of the audit. In the example: "Shanghai QinX Industrial Co., Ltd., Shanghai BaiX Technology Co., Ltd., Shanghai Pudong ZiX Printing Factory Co., Ltd., etc." can be used as key audit objects.

[0116] On the other hand, the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, and the instructions are used to be read by a machine so that the machine executes any of the above methods.

[0117] Through the above technical scheme, the method for auditing and verifying the assessment indicators of the power industry based on similarity analysis provided by the present invention obtains the electricity consumption data of all large industrial electricity price-type power supply users in the year to be audited, and obtains the corresponding electricity consumption information based on the electricity consumption data, and then screens out the large industrial electricity price-type power supply users to be identified based on the electricity consumption information, that is, a preliminary screening is performed, and for the large industrial electricity price-type power supply users to be identified, their similarity data sets of the same period are obtained, and the final target suspicious users can be screened out based on the similarity data sets of the same period; the method of using the similarity data sets to screen the target suspicious users can efficiently and accurately screen out the target suspicious users, so as to facilitate the subsequent work of auditing and verification, and effectively improve the efficiency and effect of the auditing and verification of the assessment indicators of the power industry. Specifically, the present invention conducts deep data mining on electricity consumption information data of large industrial electricity price power supply users to generate audit clue analysis information to check whether the meter readings were not recorded truthfully for the purpose of adjusting assessment indicators, or whether the electricity sales volume was adjusted. This provides technical support for auditing the phenomenon of adjusting electricity sales volume by adjusting assessment indicators in power marketing, helps auditors to quickly identify abnormal data and potential risks, and can also improve the accuracy and efficiency of audits, providing strong guarantees and new technical means for the health and compliance of corporate power marketing operations.

[0118] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0119] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0120] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0121] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0122] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0123] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0124] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0125] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0126] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for auditing and verifying assessment indicators of the power industry based on similarity analysis, characterized in that: include: Obtain electricity consumption data of all large industrial electricity price power supply users in the year to be audited; Acquiring corresponding electricity usage information according to the electricity usage data; Filter out the large industrial electricity price power supply users to be identified based on the electricity consumption information of all large industrial electricity price power supply users; Obtain the similarity data set of the large industrial electricity price power supply users to be identified in the same period; The target suspicious users are screened based on the contemporaneous similarity dataset of the large industrial electricity price power supply users to be identified.

2. The method according to claim 1, characterized in that The electricity consumption data of all large industrial electricity price power supply users in the audit year include: Determine whether the operation period of the large industrial electricity price power supply user is greater than or equal to the preset period; When it is determined that the operation period of the large industrial electricity price power supply user is greater than or equal to the preset period, obtaining the electricity consumption data of the large industrial electricity price power supply user; When it is determined that the operation period of the large industrial electricity price type power supply user is less than the preset period, the acquisition of the large industrial electricity price type power supply user is abandoned.

3. The method according to claim 2, characterized in that Acquiring corresponding electricity usage information according to the electricity usage data includes: According to formula (1), the annual proportion of electricity consumption of the large industrial electricity price power supply users in each month in the audit year is obtained. Among them, μ jk is the annual proportion of electricity consumption of the jth large industrial electricity price power supply user in the kth month of the audit year, w jk is the electricity consumption of the jth large industrial electricity price power supply user in the kth month of the audit year, w j is the annual electricity consumption of the jth large industrial electricity price power supply user in the audit year, where j is an integer number and k is an integer number; According to formula (2), the annual proportion of electricity consumption of the large industrial electricity price power supply users in each month of the year before the audit year is obtained. Among them, μ jk ′ is the annual proportion of electricity consumption of the jth large industrial electricity price power supply user in the kth month of the year before the audit year, w jk ′ is the electricity consumption of the jth large industrial electricity price power supply user in the kth month of the year before the audit year, w j ′ It is the annual electricity consumption of the jth large industrial electricity price category power supply user in the year before the year to be audited.

4. The method according to claim 1, characterized in that The large industrial electricity price class power supply users to be identified are screened out according to the electricity consumption information of all large industrial electricity price class power supply users, including the screening by using the isolation forest algorithm.

5. The method according to claim 3, characterized in that: The similarity dataset of the large industrial electricity price power supply users to be identified includes: The annual proportion of electricity consumption of each month of each large industrial electricity price power supply user in the audit year and the annual proportion of electricity consumption of each month in the year before the audit year are used to construct multiple monthly electricity consumption data groups; Obtain a Euclidean distance set based on multiple monthly electricity consumption data sets of large industrial electricity price power supply users; The Euclidean distance set is used as a contemporaneous similarity data set.

6. The method according to claim 5, characterized in that The Euclidean distance set obtained based on multiple monthly electricity consumption data sets of large industrial electricity price power supply users includes: Construct a multidimensional space coordinate system, and map the values ​​in the electricity consumption data group of each month of the large industrial electricity price power supply users to the multidimensional space coordinate system; According to formula (3), the Euclidean distance of each monthly electricity consumption data group is obtained: Wherein, d is the Euclidean distance of the electricity consumption data set, N is the spatial dimension, i is an integer number, and X i1 is the coordinate value of the mapping point X1 in the i-th dimension in the multi-dimensional space coordinate system of the median of the power consumption data set, X i2 The coordinate value of the mapping point X2 in the i-th dimension in the multi-dimensional space coordinate system of the median of the power consumption data set; The Euclidean distance of each monthly electricity consumption data group of each large industrial electricity price power supply user is aggregated to obtain the Euclidean distance set.

7. The method according to claim 5, characterized in that The target suspicious users are screened based on the similarity data set of the large industrial electricity price power supply users to be identified, including: Using a clustering algorithm to cluster each value in the contemporaneous similarity data set, and obtaining a cluster center and a cluster area; Target suspicious users are screened out according to the clustering area.

8. The method according to claim 7, characterized in that The target suspicious users screened based on the similarity dataset of the large industrial electricity price power supply users to be identified also include: According to formula (4), the threshold of the target suspicious user is obtained; C=J·α, (4) Among them, C is the threshold of the target suspicious user, J is the number of large industrial electricity price power supply users, and α is the preset selection ratio of the target suspicious users; Determine whether the target suspicious users screened out by the cluster area are greater than or equal to the threshold of the target suspicious users; In the case where it is determined that the target suspicious users screened out in the cluster area are greater than or equal to the threshold of the target suspicious users, selecting the target suspicious users of the threshold number from the target suspicious users screened out in the cluster area; When it is determined that the number of target suspicious users screened out by the clustering area is less than the threshold of target suspicious users, all target suspicious users screened out by the clustering area are acquired.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and the instructions are used to be read by a machine so as to enable the machine to execute the method according to any one of claims 1 to 8.

Images