A method and system for electricity transaction data settlement based on big data

Abstract

The present invention relates to the field of power transaction data settlement, and specifically to a power transaction data settlement method and system based on big data; according to the characteristics of power transaction data, abnormal data in first basic data is screened, and abnormal areas are selected for subsequent further calculations; based on the two perspectives of area and user, a power consumption matrix and a power consumption change matrix are established, and the association is performed through a correlation matrix; the average value matrix and the standard deviation matrix are calculated according to the correlation matrix, so as to further construct a judgment matrix to represent the power consumption range of the user; by constructing an evaluation matrix and calculating an abnormal threshold, the abnormal index of each user in the evaluation matrix can be compared with the abnormal threshold to judge whether each user has an abnormality within a certain period of time; the problem that the accuracy of the collected data is difficult to guarantee in the current process of power transaction data settlement, resulting in poor accuracy of the power transaction data settlement result is solved.

Description

Technical Field

[0001] The present invention relates to the field of power transaction data settlement, and in particular to a power transaction data settlement method and system based on big data. Background Art

[0002] Electricity trading settlement is a multi-link, multi-step process, which mainly includes five core links: electricity metering, data management, settlement processing, account management and report output. Electricity metering is the basis of electricity settlement, ensuring the authenticity and accuracy of electricity sales. It is usually carried out by power companies or third-party institutions. The metering results are an important basis for electricity settlement. Data management involves the input, storage, processing and output of various data such as electricity, price, time period, etc., and requires the use of specialized software systems such as big data models to ensure the integrity, accuracy and security of the data. After that, it enters the settlement processing link and selects different settlement models for calculation according to the data type.

[0003] However, in the process of data management, due to environmental noise and some human illegal factors in the actual data collection process, the accuracy of the collected data is difficult to guarantee, which will lead to data errors. For example, some users reported that their actual electricity consumption was far less than the electricity consumption and electricity bill at the time of settlement, which was often caused by electricity theft or leakage. This situation needs to be investigated to prevent such incidents from continuing to occur. At the same time, it also ensures the accuracy of the electricity trading data settlement results to improve public satisfaction and reduce complaints. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a method and system for electricity trading data settlement based on big data, which solves the problem that the accuracy of collected data is difficult to ensure in the current process of electricity trading data settlement, resulting in poor accuracy of electricity trading data settlement results.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A method for electricity transaction data settlement based on big data, the method comprising the following steps:

[0007] S1. Obtaining first basic data of each region including regional pre-settlement electricity, regional electricity consumption, user electricity consumption and user pre-settlement electricity within a period of time;

[0008] S2. Eliminate abnormal data in the first basic data, calculate the power error rate of each area, and determine whether the power error rate is greater than the power error rate threshold;

[0009] If yes, mark the area as an abnormal area, and mark the first basic data of the abnormal area as fourth basic data;

[0010] If not, the area is marked as a normal area;

[0011] S3. Constructing a power consumption matrix of users in each abnormal area according to the fourth basic data, constructing a power consumption change matrix and a regional power consumption change matrix according to the user power consumption matrix, and constructing a correlation matrix between the power consumption change matrix and the regional power consumption change matrix;

[0012] S4, calculating the mean value matrix and the standard deviation matrix of the correlation matrix, and constructing the upper limit matrix and the lower limit matrix, and constructing the judgment matrix according to the upper limit matrix and the lower limit matrix;

[0013] S5. Construct an evaluation matrix based on the judgment matrix and the user's power consumption matrix, obtain the user's abnormal index based on the evaluation matrix, and determine whether the abnormal index is greater than the abnormal threshold;

[0014] If yes, the user is marked as an abnormal user and output;

[0015] If not, the user is marked as a normal user and entered into the settlement model.

[0016] Preferably, in step S2, the following steps are specifically included:

[0017] S21. Eliminate abnormal data in basic data;

[0018] S22, obtaining the actual time and theoretical time of first basic data collection in each area;

[0019] S23, calculating the ratio of the actual time number to the theoretical time number to obtain the acquisition rate;

[0020] S24, setting a collection rate threshold, and determining whether the collection rate of each area is greater than the collection rate threshold;

[0021] If yes, mark the corresponding area as the first area;

[0022] If not, the corresponding area is marked as the second area;

[0023] S25. Calculate the power error rate of each first area according to the third basic data; the calculation formula of the power error rate is:

[0024] ;

[0025] In the above formula, Indicates the power error rate, Indicates the amount of electricity to be settled. Indicates power consumption;

[0026] S26, determining whether the power error rate of each first area is greater than the power error rate threshold;

[0027] If yes, marking the first area as an abnormal area, and marking the third basic data of the abnormal area as fourth basic data;

[0028] If not, the third area is marked as a normal area.

[0029] Preferably, in step S21, the following steps are specifically included:

[0030] S211, determining in turn whether each pre-settlement electricity quantity in the basic data is less than 0;

[0031] If yes, marking the data as abnormal data and deleting it from the first basic data to generate second basic data;

[0032] If not, the data is marked as normal data;

[0033] S212, determining in turn whether the pre-settlement electricity data of each region in the second basic data exceeds the deviation threshold of the regional segmented electricity;

[0034] If yes, the pre-settlement electricity data of the area is marked as abnormal data and deleted from the second basic data to generate third basic data;

[0035] If not, the data is marked as normal data.

[0036] Preferably, in step S3, the following steps are specifically included:

[0037] S31. Constructing a power consumption matrix of users in each abnormal area according to the fourth basic data; the expression of the power consumption matrix is:

[0038] ;

[0039] In the above formula, Represents the electricity consumption matrix of users in any area, Indicates the electricity consumption of the mth user on the nth day;

[0040] S32, calculating the power consumption of each area on any day based on the power consumption matrix of users in each abnormal area;

[0041] S33, constructing a power consumption matrix of each abnormal area within a period of time; the expression of the power consumption matrix of the abnormal area within a period of time is:

[0042] ;

[0043] In the above formula, represents the power consumption matrix, Indicates the power consumption of the abnormal area on the nth day;

[0044] S34. Construct an electricity consumption change matrix according to the user's electricity consumption matrix; the expression of the electricity consumption change matrix is:

[0045] ;

[0046] in,

[0047] ;

[0048] In the above formula, Represents the user's electricity consumption change matrix, It represents the change in electricity consumption of the jth user on the i-th day in the abnormal area;

[0049] S35. Construct a regional power change matrix based on the power loss matrix of each abnormal area within a period of time; the expression of the regional power change matrix of the abnormal area within a period of time is:

[0050] ;

[0051] in,

[0052] ;

[0053] In the above formula, Represents the regional power change matrix of the abnormal area over a period of time, Indicated in Regional electricity consumption change value during the day;

[0054] S36. Construct a correlation matrix of the two based on the regional power change matrix of the abnormal area within a period of time and the power consumption change matrix of the area; the expression of the correlation matrix is:

[0055] ;

[0056] In the above formula, represents the incidence matrix, represents the electricity consumption change matrix, represents the regional electricity change matrix, Represents the identity matrix of order n-1.

[0057] Preferably, in step S32, the following steps are specifically included:

[0058] S321. Calculate the pre-settlement electricity of each region on any day; the calculation formula for the pre-settlement electricity is:

[0059] ;

[0060] In the above formula, represents the estimated electricity consumption of each region on the i-th day, represents the first The pre-settled electricity of power sources, there are F power sources in total;

[0061] S322, calculating the power consumption of each area on any day based on the power consumption matrix of users in each abnormal area; the calculation formula of the power consumption is:

[0062] ;

[0063] In the above formula, represents the power consumption of each abnormal area on the i-th day, It represents the estimated electricity consumption of each abnormal area on the i-th day. It represents the electricity consumption of the jth user on the i-th day, and the total number of users is m.

[0064] Preferably, in step S4, the following steps are specifically included:

[0065] S41, calculating the average value of each row element in the incidence matrix to construct an average value matrix; the expression of the average value matrix is:

[0066] ;

[0067] in,

[0068] ;

[0069] In the above formula, represents the mean value matrix, represents the i-th element in the mean value matrix, represents the average value of the elements in the n-1th row of the incidence matrix, represents the element in the i-th row and j-th column of the incidence matrix, and each row has a total of m elements;

[0070] S42, calculating the standard deviation of each row element in the incidence matrix to construct a standard deviation matrix; the expression of the standard deviation matrix is:

[0071] ;

[0072] in,

[0073] ;

[0074] In the above formula, Represents the standard deviation matrix. Each row of the standard deviation matrix has m elements. represents the element at row i and column j in the incidence matrix, represents the i-th element in the mean value matrix, The i-th element of the standard deviation matrix is ​​the standard deviation of the i-th row elements in the incidence matrix;

[0075] S43, constructing an upper limit matrix and a lower limit matrix of the correlation matrix according to the mean value matrix and the standard deviation matrix;

[0076] The expression of the upper limit matrix is: ;

[0077] The expression of the lower limit matrix is: ;

[0078] In the above formula, and denote the upper limit matrix and the lower limit matrix respectively, represents the limiting coefficient, represents the mean value matrix, represents the standard deviation matrix;

[0079] S44. Construct a judgment matrix according to the upper limit matrix and the lower limit matrix.

[0080] Preferably, in step S5, the following steps are specifically included:

[0081] S51, calculating the sum of the elements of each row in the judgment matrix to construct an evaluation matrix;

[0082] S52, calculating the abnormal threshold of each user; the calculation formula of the abnormal threshold is:

[0083] ;

[0084] In the above formula, represents the abnormal coefficient of each user, It represents the evaluation coefficient, which is generally 0.5, and n represents the total number of days corresponding to the fourth basic data;

[0085] S53, obtaining an abnormal index of any user according to the evaluation matrix, and determining whether the abnormal index is greater than an abnormal threshold;

[0086] If yes, the user is marked as an abnormal user and output;

[0087] If not, the user is marked as a normal user and entered into the settlement model.

[0088] Preferably, in step S44, the expression of the judgment matrix is:

[0089] ;

[0090] in,

[0091] ;

[0092] In the above formula, represents the judgment matrix, represents the element in the i-th row and j-th column of the judgment matrix, and denote the elements in the i-th row and j-th column of the upper limit matrix and the lower limit matrix, respectively. Represents the element at the i-th row and j-th column in the incidence matrix.

[0093] Preferably, in step S51, the expression of the evaluation matrix is:

[0094] ;

[0095] in,

[0096] ;

[0097] In the above formula, represents the evaluation matrix, It represents the j-th element of the evaluation matrix, that is, the abnormal index of the j-th element.

[0098] The technical solution also provides a system for implementing the power transaction data settlement method, the system comprising: a processor and a memory, the memory being used to store a computer program, and the computer program implementing the power transaction data settlement method when executed by the processor.

[0099] Compared with the prior art, the present invention provides a method and system for electricity transaction data settlement based on big data, which has the following beneficial effects:

[0100] 1. The present invention screens and removes abnormal data in the first basic data according to the characteristics of power trading data, and at the same time determines whether the power trading data of each area is reliable by calculating the power error rate and collection rate of each area, and selects areas with a higher probability of abnormal users for subsequent further calculations, so as to facilitate the subsequent accurate selection of abnormal users with higher accuracy and less computing power used.

[0101] 2. The present invention establishes a power consumption matrix and a power consumption change matrix based on the two perspectives of region and user, and associates the regional power consumption change matrix with the user's power consumption change matrix through an association matrix to explore the relationship between the two, thereby facilitating further comprehensive consideration of whether each user is an abnormal user.

[0102] 3. The present invention calculates the mean value matrix and the standard deviation matrix according to the correlation matrix, so as to further construct the upper limit matrix and the lower limit matrix of the upper and lower limits of the user's electricity consumption according to the mean value matrix and the standard deviation matrix, and further construct a judgment matrix to represent the user's electricity consumption range, so as to facilitate the subsequent judgment of whether the user's electricity consumption is abnormal.

[0103] 4. The present invention constructs an evaluation matrix to comprehensively count the number of times each user has abnormal electricity consumption within a period of time, and comprehensively calculates the abnormal threshold based on the duration involved, so that the abnormal index of each user in the evaluation matrix can be compared with the abnormal threshold to determine whether each user has an abnormality within a certain period of time, so that the abnormal users can be screened out for subsequent further processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0104] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0105] Figure 1 The figure is a flow chart of the method for settling electricity transaction data according to the present invention. DETAILED DESCRIPTION

[0106] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific implementation methods, so that the implementation process of how the present application uses technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly.

[0107] Those skilled in the art can understand that all or part of the steps in the following embodiments can be completed by instructing the relevant hardware through a program, so the present application can be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application can be in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0108] In the process of data governance management, data governance refers to the process of managing and governing the data assets of the trading center, aiming to ensure the security, integrity, reliability and compliance of the data. The content of data governance includes the management and monitoring of data collection, storage, processing, transmission and application processes, as well as data privacy protection, security protection and compliance management. The goal of data governance is to improve the quality, credibility and effectiveness of data, reduce data risks and damage, and promote the maximum utilization and value creation of data assets.

[0109] In order to solve the problem that the accuracy of collected data is difficult to ensure in the current process of power transaction data settlement, resulting in poor accuracy of power transaction data settlement results, the present invention provides a power transaction data settlement method based on big data, which gradually identifies users with abnormal power transaction data through an algorithm, so that the power transaction data of this batch of abnormal users can be selected, and the accuracy of the power transaction data is confirmed after investigating whether there are problems with the circuits of the abnormal users, and settlement is performed through a settlement model, so as to further ensure the accuracy of the settlement results and improve the satisfaction of the masses. The method comprises the following steps:

[0110] S1. Obtain the first basic data of each region within a period of time, including regional pre-settlement electricity, regional electricity consumption, user electricity consumption and user pre-settlement electricity; the acquisition method is generally to use a big data model to obtain a large amount of relevant electricity transaction data published on the website of the relevant department;

[0111] S2. Eliminate abnormal data in the first basic data, calculate the power error rate of each area, and determine whether the power error rate is greater than the power error rate threshold;

[0112] If yes, mark the area as an abnormal area, and mark the first basic data of the abnormal area as fourth basic data;

[0113] If not, the area is marked as a normal area; for some of the first basic data, the values ​​are obviously inconsistent with the actual ones, which need to be eliminated to prevent affecting the subsequent judgment of abnormal users. In addition, due to calculation errors and in order to improve the calculation speed, under the condition of limited computing power, each area is preliminarily screened to select the area most likely to have abnormal users for subsequent calculations. In step S2, the following steps are specifically included:

[0114] S21, remove abnormal data in the basic data; abnormal data is data that is obviously inconsistent with the actual situation. Here, according to the characteristics of power trading data, two dimensions are mainly defined to screen abnormal data. These two dimensions are the user's pre-settlement electricity and the regional pre-settlement electricity. In step S21, the following steps are specifically included:

[0115] S211, determining in turn whether each pre-settlement electricity quantity in the basic data is less than 0;

[0116] If yes, marking the data as abnormal data and deleting it from the first basic data to generate second basic data;

[0117] If not, the data is marked as normal data;

[0118] S212, determine in turn whether the pre-settlement electricity data of each region in the second basic data exceeds the deviation threshold of the regional segmentation electricity; in practice, 3 times is generally selected, that is, when the regional pre-settlement electricity data is greater than 3 times the regional distribution transformer capacity, it means that the pre-settlement electricity data of the region is abnormal and needs to be confirmed later. After confirming that there is no abnormality, it will be included in the calculation range;

[0119] If yes, the pre-settlement electricity data of the area is marked as abnormal data and deleted from the second basic data to generate third basic data;

[0120] If not, the data is marked as normal data.

[0121] S22, obtaining the actual time and theoretical time of the first basic data collection of each area; the theoretical time is the total number of times when the system needs to collect data, and the actual time is the total number of times when the system actually collects data;

[0122] S23, calculating the ratio of the actual time number to the theoretical time number to obtain the collection rate; due to the influence of factors such as noise, at a certain time, the power transaction data of a certain user may not be collected, so the actual time number of the area will decrease. Therefore, by calculating the collection rate, that is, the ratio of the actual time number to the theoretical time number, the reliability of the data is reflected. In practice, only the area with a collection rate greater than 95% is generally subject to subsequent calculation;

[0123] S24, setting a collection rate threshold, and determining whether the collection rate of each area is greater than the collection rate threshold; in practice, the collection rate threshold is generally set to 95%;

[0124] If yes, mark the corresponding area as the first area;

[0125] If not, the corresponding area is marked as the second area;

[0126] S25. Calculate the power error rate of each first area according to the third basic data; the calculation formula of the power error rate is:

[0127] ;

[0128] In the above formula, Indicates the power error rate, Indicates the amount of electricity to be settled. Indicates power consumption;

[0129] S26, determining whether the power error rate of each first area is greater than a power error rate threshold; the power error rate threshold is generally set to 0.25;

[0130] If yes, the first area is marked as an abnormal area, and the third basic data of the abnormal area is marked as the fourth basic data; the abnormal area is an area with a high probability of containing abnormal users;

[0131] If not, the third area is marked as a normal area.

[0132] S3, constructing a power consumption matrix of users in each abnormal area according to the fourth basic data, constructing a power consumption change matrix and a regional power consumption change matrix according to the user power consumption matrix, and constructing a correlation matrix of the power consumption change matrix and the regional power consumption change matrix; on the basis of the fourth basic data obtained after eliminating a large amount of abnormal data, comprehensively considering the user's power transaction data and the regional power transaction data to construct a correlation matrix, so as to facilitate further judgment according to the correlation matrix in the future. In step S3, the following steps are specifically included:

[0133] S31. Constructing a power consumption matrix of users in each abnormal area according to the fourth basic data; the expression of the power consumption matrix is:

[0134] ;

[0135] In the above formula, Represents the electricity consumption matrix of users in any area, Indicates the electricity consumption of the mth user on the nth day;

[0136] S32, according to the power consumption matrix of users in each abnormal area, the power loss of each area on any day is calculated; for some areas, there may be multiple power sources supplying electricity, and the power loss needs to be calculated based on the pre-settled power and power consumption of different power sources; in step S32, the following steps are specifically included:

[0137] S321. Calculate the pre-settlement electricity of each region on any day; the calculation formula for the pre-settlement electricity is:

[0138] ;

[0139] In the above formula, represents the estimated electricity consumption of each region on the i-th day, represents the first The pre-settled electricity of power sources, there are F power sources in total;

[0140] S322, calculating the power consumption of each area on any day based on the power consumption matrix of users in each abnormal area; the calculation formula of the power consumption is:

[0141] ;

[0142] In the above formula, represents the power consumption of each abnormal area on the i-th day, It represents the estimated electricity consumption of each abnormal area on the i-th day. It represents the electricity consumption of the jth user on the i-th day, and the total number of users is m.

[0143] S33, constructing a power consumption matrix of each abnormal area within a period of time; the expression of the power consumption matrix of the abnormal area within a period of time is:

[0144] ;

[0145] In the above formula, represents the power consumption matrix, Indicates the power consumption of the abnormal area on the nth day;

[0146] S34. Construct an electricity consumption change matrix according to the user's electricity consumption matrix; the expression of the electricity consumption change matrix is:

[0147] ;

[0148] in,

[0149] ;

[0150] In the above formula, Represents the user's electricity consumption change matrix, It represents the change in electricity consumption of the jth user on the i-th day in the abnormal area;

[0151] S35. Construct a regional power change matrix based on the power loss matrix of each abnormal area within a period of time; the expression of the regional power change matrix of the abnormal area within a period of time is:

[0152] ;

[0153] in,

[0154] ;

[0155] In the above formula, Represents the regional power change matrix of the abnormal area over a period of time, Indicated in Regional electricity consumption change value during the day;

[0156] S36. Construct a correlation matrix of the two based on the regional power change matrix of the abnormal area within a period of time and the power consumption change matrix of the area; the expression of the correlation matrix is:

[0157] ;

[0158] In the above formula, represents the incidence matrix, represents the electricity consumption change matrix, represents the regional electricity change matrix, Represents the identity matrix of order n-1.

[0159] S4, calculate the mean value matrix and standard deviation matrix of the correlation matrix, and construct the upper limit matrix and the lower limit matrix, and construct the judgment matrix according to the upper limit matrix and the lower limit matrix; the daily power consumption of different users in a period of time roughly presents a normal distribution, so a certain multiple of the standard deviation is used here to construct the range of user power consumption, generally 3 times the standard deviation, in step S4, specifically including the following steps:

[0160] S41, calculating the average value of each row element in the incidence matrix to construct an average value matrix; the expression of the average value matrix is:

[0161] ;

[0162] in,

[0163] ;

[0164] In the above formula, represents the mean value matrix, represents the i-th element in the mean value matrix, represents the average value of the elements in the n-1th row of the incidence matrix, represents the element in the i-th row and j-th column of the incidence matrix, and each row has a total of m elements;

[0165] S42, calculating the standard deviation of each row element in the incidence matrix to construct a standard deviation matrix; the expression of the standard deviation matrix is:

[0166] ;

[0167] in,

[0168] ;

[0169] In the above formula, Represents the standard deviation matrix. Each row of the standard deviation matrix has m elements. represents the element at row i and column j in the incidence matrix, represents the i-th element in the mean value matrix, The i-th element of the standard deviation matrix is ​​the standard deviation of the i-th row elements in the incidence matrix;

[0170] S43, constructing an upper limit matrix and a lower limit matrix of the correlation matrix according to the mean value matrix and the standard deviation matrix;

[0171] The expression of the upper limit matrix is: ;

[0172] The expression of the lower limit matrix is: ;

[0173] In the above formula, and denote the upper limit matrix and the lower limit matrix respectively, represents the limiting coefficient, represents the mean value matrix, represents the standard deviation matrix;

[0174] S44. Construct a judgment matrix according to the upper limit matrix and the lower limit matrix; the expression of the judgment matrix is:

[0175] ;

[0176] in,

[0177] ;

[0178] In the above formula, represents the judgment matrix, represents the element in the i-th row and j-th column of the judgment matrix, and denote the elements in the i-th row and j-th column of the upper limit matrix and the lower limit matrix, respectively. Represents the element at the i-th row and j-th column in the incidence matrix.

[0179] S5. Construct an evaluation matrix based on the judgment matrix and the user's power consumption matrix, obtain the user's abnormal index based on the evaluation matrix, and determine whether the abnormal index is greater than the abnormal threshold;

[0180] If yes, the user is marked as an abnormal user and output;

[0181] If not, the user is marked as a normal user and input into the settlement model. According to the type of data and the purpose of settlement, the enterprise will construct different settlement models. At this time, the data of the verified normal user is input into the settlement model to complete the settlement. For abnormal users, the enterprise will investigate the user to further determine whether the user's electricity consumption is abnormal. After obtaining the user's judgment matrix, the user's electricity consumption range is obtained. At this time, the evaluation matrix is ​​constructed to determine whether the user's electricity consumption is abnormal on each day. Then, the abnormal index and abnormal threshold are calculated to achieve the purpose of comprehensively judging whether the user's electricity trading data is abnormal. In step S5, the following steps are specifically included:

[0182] S51, calculate the sum of the elements in each column of the judgment matrix to construct an evaluation matrix; the calculation expression of the evaluation matrix is:

[0183] ;

[0184] in,

[0185] ;

[0186] In the above formula, represents the evaluation matrix, It represents the j-th element of the evaluation matrix, that is, the abnormal index of the j-th element.

[0187] S52, calculating the abnormal threshold of each user; the calculation formula of the abnormal threshold is:

[0188] ;

[0189] In the above formula, represents the abnormal coefficient of each user, It represents the evaluation coefficient, which is generally 0.5. The range of the evaluation coefficient is between 0 and 1. The larger the value, the higher the proportion of abnormal users included in the subsequent judgment results, but the possibility of missing abnormal users is also higher. n represents the total number of days corresponding to the fourth basic data;

[0190] S53, obtaining an abnormal index of any user according to the evaluation matrix, and determining whether the abnormal index is greater than an abnormal threshold;

[0191] If yes, the user is marked as an abnormal user and output;

[0192] If not, the user is marked as a normal user and entered into the settlement model.

[0193] During the implementation of the present invention, the abnormal data in the first basic data is first eliminated and several areas with a high probability of abnormal users are gradually selected. Then, based on the abnormal areas and the fourth basic data, the changes in power consumption of regional power trading data and user power trading data are comprehensively considered, and an association matrix is ​​constructed. Then, according to the principle of the normal function, the upper limit matrix and the lower limit matrix are calculated according to the association matrix, so as to construct a judgment matrix. Then, according to the evaluation matrix, the user's abnormal index is obtained and the abnormal threshold is calculated, so as to intuitively judge whether the corresponding user is abnormal by comparing the abnormal index and the abnormal threshold. For users judged to be normal, the user data is input into the settlement model in the power trading settlement process during the power trading settlement process. For abnormal users, the user information will be output, so as to facilitate the investigation of their power consumption, so as to give a final judgment on whether there is an abnormality, so as to perform subsequent further calculations on the abnormal users.

[0194] Corresponding to the big data-based power transaction data settlement method provided in the above-mentioned embodiment, the present embodiment also provides a system for implementing the power transaction data settlement method. Since the power transaction data settlement system provided in the present embodiment corresponds to the power transaction data settlement method provided in the above-mentioned embodiment, the implementation method of the aforementioned power transaction data settlement method is also applicable to the power transaction data settlement system provided in the present embodiment, and will not be described in detail in the present embodiment.

[0195] The system includes a processor and a memory, wherein the memory is used to store a computer program, and when the computer program is executed by the processor, a method for settling electric power transaction data is implemented.

[0196] The above implementation methods have been described in detail. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for settling power transaction data based on big data, characterized in that: The method comprises the following steps: S1. Obtaining first basic data of each region including regional pre-settlement electricity, regional electricity consumption, user electricity consumption and user pre-settlement electricity within a period of time; S2. Remove abnormal data from the first basic data, calculate the power error rate of each area, and determine whether the power error rate is greater than the power error rate threshold; the calculation formula of the power error rate is: ； In the above formula, Indicates the power error rate, Indicates the amount of electricity to be settled. Indicates power consumption; If yes, mark the area as an abnormal area, and mark the first basic data of the abnormal area as fourth basic data; If not, the area is marked as a normal area; S3, constructing a power consumption matrix of users in each abnormal area according to the fourth basic data, constructing a power consumption change matrix and a regional power consumption change matrix according to the user power consumption matrix, and constructing a correlation matrix between the power consumption change matrix and the regional power consumption change matrix; in step S3, specifically including the following steps: S31. Constructing a power consumption matrix of users in each abnormal area according to the fourth basic data; the expression of the power consumption matrix is: ； In the above formula, Represents the electricity consumption matrix of users in any area, Indicates the electricity consumption of the mth user on the nth day; S32, calculating the power consumption of each area on any day based on the power consumption matrix of users in each abnormal area; S33, constructing a power consumption matrix of each abnormal area within a period of time; the expression of the power consumption matrix of the abnormal area within a period of time is: ； In the above formula, represents the power consumption matrix, Indicates the power consumption of the abnormal area on the nth day; S34. Construct an electricity consumption change matrix according to the user's electricity consumption matrix; the expression of the electricity consumption change matrix is: ； in, ； In the above formula, Represents the user's electricity consumption change matrix, It represents the change in electricity consumption of the jth user on the i-th day in the abnormal area; S35. Construct a regional power change matrix based on the power loss matrix of each abnormal area within a period of time; the expression of the regional power change matrix of the abnormal area within a period of time is: ； in, ； In the above formula, Represents the regional power change matrix of the abnormal area over a period of time, Indicated in Regional electricity consumption change value during the day; S36. Construct a correlation matrix of the two based on the regional power change matrix of the abnormal area within a period of time and the power consumption change matrix of the area; the expression of the correlation matrix is: ； In the above formula, represents the incidence matrix, represents the electricity consumption change matrix, represents the regional electricity change matrix, represents the identity matrix of order n-1; S4, calculating the mean value matrix and the standard deviation matrix of the correlation matrix, and constructing the upper limit matrix and the lower limit matrix, and constructing the judgment matrix according to the upper limit matrix and the lower limit matrix; in step S4, specifically including the following steps: S41, calculating the average value of each row element in the incidence matrix to construct an average value matrix; the expression of the average value matrix is: ； in, ； In the above formula, represents the mean value matrix, represents the i-th element in the mean value matrix, represents the average value of the elements in the n-1th row of the incidence matrix, represents the element in the i-th row and j-th column of the incidence matrix, and each row has a total of m elements; S42, calculating the standard deviation of each row element in the incidence matrix to construct a standard deviation matrix; the expression of the standard deviation matrix is: ； in, ； In the above formula, Represents the standard deviation matrix. Each row of the standard deviation matrix has m elements. represents the element at row i and column j in the incidence matrix, represents the i-th element in the mean value matrix, The i-th element of the standard deviation matrix is ​​the standard deviation of the i-th row elements in the incidence matrix; S43, constructing an upper limit matrix and a lower limit matrix of the correlation matrix according to the mean value matrix and the standard deviation matrix; The expression of the upper limit matrix is: ; The expression of the lower limit matrix is: ; In the above formula, and denote the upper limit matrix and the lower limit matrix respectively, represents the limiting coefficient, represents the mean value matrix, represents the standard deviation matrix; S44. Construct a judgment matrix according to the upper limit matrix and the lower limit matrix; the expression of the judgment matrix is: ； in, ； In the above formula, represents the judgment matrix, represents the element in the i-th row and j-th column of the judgment matrix, and denote the elements in the i-th row and j-th column of the upper limit matrix and the lower limit matrix, respectively. represents the element in the i-th row and j-th column of the incidence matrix; S5. Construct an evaluation matrix based on the judgment matrix and the user's power consumption matrix, obtain the user's abnormal index based on the evaluation matrix, and determine whether the abnormal index is greater than the abnormal threshold; the expression of the evaluation matrix is: ； in, ； In the above formula, represents the evaluation matrix, represents the j-th element of the evaluation matrix, that is, the abnormal index of the j-th element; If yes, the user is marked as an abnormal user and output; If not, the user is marked as a normal user and entered into the settlement model.

2. The data settlement method according to claim 1, characterized in that: In step S2, the following steps are specifically included: S21. Eliminate abnormal data in basic data; S22, obtaining the actual time and theoretical time of first basic data collection in each area; S23, calculating the ratio of the actual time number to the theoretical time number to obtain the acquisition rate; S24, setting a collection rate threshold, and determining whether the collection rate of each area is greater than the collection rate threshold; If yes, mark the corresponding area as the first area; If not, the corresponding area is marked as the second area; S25. Calculate the power error rate of each first area according to the third basic data; the calculation formula of the power error rate is: ； In the above formula, Indicates the power error rate, Indicates the amount of electricity to be settled. Indicates power consumption; S26, determining whether the power error rate of each first area is greater than the power error rate threshold; If yes, marking the first area as an abnormal area, and marking the third basic data of the abnormal area as fourth basic data; If not, the first area is marked as a normal area.

3. The data settlement method according to claim 2, characterized in that: In step S21, the following steps are specifically included: S211, determining in turn whether each pre-settlement electricity quantity in the basic data is less than 0; If yes, marking the data as abnormal data and deleting it from the first basic data to generate second basic data; If not, the data is marked as normal data; S212, determining in turn whether the pre-settlement electricity data of each region in the second basic data exceeds the deviation threshold of the regional segmented electricity; If yes, the pre-settlement electricity data of the area is marked as abnormal data and deleted from the second basic data to generate third basic data; If not, the data is marked as normal data.

4. The data settlement method according to claim 1, characterized in that: In step S32, the following steps are specifically included: S321. Calculate the pre-settlement electricity of each region on any day; the calculation formula for the pre-settlement electricity is: ； In the above formula, represents the estimated electricity consumption of each region on the i-th day, represents the first The pre-settled electricity of power sources, there are F power sources in total; S322, calculating the power consumption of each area on any day based on the power consumption matrix of users in each abnormal area; the calculation formula of the power consumption is: ； In the above formula, represents the power consumption of each abnormal area on the i-th day, It represents the estimated electricity consumption of each abnormal area on the i-th day. It represents the electricity consumption of the jth user on the i-th day, and the total number of users is m.

5. The data settlement method according to claim 1, characterized in that: In step S5, the following steps are specifically included: S51, calculating the sum of the elements of each row in the judgment matrix to construct an evaluation matrix; S52, calculating the abnormal threshold of each user; the calculation formula of the abnormal threshold is: ； In the above formula, represents the abnormal coefficient of each user, represents the evaluation coefficient, Indicates the total number of days corresponding to the fourth basic data; S53, obtaining an abnormal index of any user according to the evaluation matrix, and determining whether the abnormal index is greater than an abnormal threshold; If yes, the user is marked as an abnormal user and output; If not, the user is marked as a normal user and entered into the settlement model.

6. A settlement system for implementing the power transaction data settlement method according to any one of claims 1 to 5, characterized in that: It comprises a processor and a memory, wherein the memory is used to store a computer program, and when the computer program is executed by the processor, the method for settling power transaction data as described in any one of claims 1 to 5 is implemented.

Images