Method and device for selecting representative day for line loss calculation and electronic equipment

Abstract

The invention discloses a representative day selection method and device for line loss calculation and electronic equipment. The method comprises the following steps: acquiring a plurality of operation data of a power system in different time periods; the multiple pieces of operation data are analyzed, multiple line loss influence factors are obtained, and the line loss influence factors are used for representing the data types of the operation data which can influence the line loss calculation result of the power system; guiding the factor evaluation model to select a plurality of key influence factors from the plurality of line loss influence factors based on the plurality of operation data; and determining a representative day from the plurality of time periods based on the plurality of key influence factors and the plurality of operation data, the representative day being used for performing line loss calculation on the power system. The technical problem that the accuracy of reflecting the line loss rate of the court by the selected representative day in the related technology is low is solved.

Description

Technical Field

[0001] The present invention relates to the field of power systems, and in particular to a method and device for selecting a representative day for line loss calculation, and electronic equipment. Background Art

[0002] Line loss calculation is a basic and critical task in the operation and management of power systems. It is not only directly related to the economic benefits of power companies, but also affects the operating efficiency and reliability of power grids. Line loss refers to the loss of electric energy during the transmission of electricity from power plants to users, mainly including resistance loss, reactance loss, transformer loss and power electronic equipment loss. In power systems, accurate calculation of line loss is of great significance for power grid planning, power dispatching, cost accounting and energy efficiency improvement. When calculating line loss, the selection of typical daily load curves is particularly critical. The representative daily load curve refers to the load curve of a certain day in a specific time period that can reflect the load characteristics in that time period. In power system analysis, representative daily load curves are often used to predict and analyze power demand, plan power grid construction, evaluate the operating status of power equipment, and calculate line losses.

[0003] The traditional method of selecting representative days mainly relies on statistical methods, such as selecting the daily load curve closest to the monthly average load rate or selecting the daily load curve with the largest load in the whole month. However, with the complexity of the power grid structure and the access of a high proportion of distributed power sources, the representative days selected by these methods are gradually unable to meet the line loss calculation requirements of the power system, resulting in the selected representative days reflecting the actual line loss rate of the substation with low accuracy.

[0004] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0005] The embodiments of the present invention provide a method and device for selecting a representative day for line loss calculation, and an electronic device, so as to at least solve the technical problem in the related art that the representative day selected has low accuracy in reflecting the line loss rate of the substation area.

[0006] According to one aspect of an embodiment of the present invention, there is provided a method for selecting a representative day for line loss calculation, comprising: acquiring a plurality of operating data of an electric power system in different time periods; analyzing the plurality of operating data to obtain a plurality of line loss influencing factors, wherein the line loss shadow factor is used to characterize the data type of the operating data that may affect the line loss calculation result of the electric power system; based on the plurality of operating data, guiding a factor evaluation model to select a plurality of key influencing factors from the plurality of line loss influencing factors; based on the plurality of key influencing factors and the plurality of operating data, determining a representative day from a plurality of time periods, wherein the representative day is used to perform line loss calculation on the electric power system.

[0007] Furthermore, based on multiple operating data, the guiding factor evaluation model selects multiple key influencing factors from multiple line loss influencing factors, including: performing data standardization processing on multiple operating data based on multiple line loss influencing factors to obtain standard operating data corresponding to different line loss influencing factors; performing sensitivity analysis on multiple line loss influencing factors based on the standard operating data to obtain multiple sensitivity weights, and performing multi-factor correlation analysis on multiple line loss influencing factors based on the standard operating data to obtain multiple grey correlation weights, wherein different sensitivity weights are used to reflect the degree of influence of different line loss influencing factors on line loss calculation results, and different grey correlation weights are used to reflect the degree of correlation between different line loss influencing factors and line loss calculation results, and multiple line loss influencing factors, multiple sensitivity weights and multiple grey correlation weights correspond to each other one by one; multiple line loss influencing factors are evaluated based on multiple sensitivity weights and multiple grey correlation weights to obtain multiple evaluation indicators; and multiple key influencing factors are determined from multiple line loss influencing factors based on multiple evaluation indicators.

[0008] Furthermore, a sensitivity analysis is performed on multiple line loss influencing factors based on standard operating data to obtain multiple sensitivity weights, including: determining the variance, covariance and joint variance corresponding to different line loss influencing factors based on the standard operating data; summarizing the variance, covariance and joint variance corresponding to different line loss influencing factors to obtain the total variance; obtaining the partial variance corresponding to different line loss influencing factors based on the quotient of the variance corresponding to different line loss influencing factors and the total variance; summarizing the partial variances corresponding to different line loss influencing factors to obtain the total partial variance; obtaining multiple sensitivity weights based on the quotient of the absolute value of the partial variance corresponding to different line loss influencing factors and the total partial variance.

[0009] Furthermore, a multi-factor correlation analysis is performed on multiple line loss influencing factors based on standard operating data to obtain multiple grey correlation weights, including: based on the system identification of the power system, obtaining multiple reference operating data corresponding to the multiple line loss influencing factors, wherein the multiple line loss influencing factors correspond to the multiple reference operating data one-to-one; determining multiple correlations corresponding to the multiple line loss influencing factors based on the multiple reference operating data and the standard operating data corresponding to different reference operating data; obtaining a total correlation based on the sum of the multiple correlations; and obtaining multiple grey correlation weights based on the quotient of the multiple correlations and the total correlation.

[0010] Furthermore, based on multiple reference operating data and standard operating data corresponding to different reference operating data, multiple correlations corresponding to multiple line loss influencing factors are determined, including: obtaining the difference between different reference operating data and the corresponding standard operating data to obtain multiple difference values, wherein the multiple difference values ​​correspond one-to-one to the standard operating data; processing the maximum difference, the minimum difference, and multiple difference values ​​among the multiple difference values ​​according to a preset formula to obtain multiple correlation coefficients corresponding to different reference operating data; and determining the correlations corresponding to different line loss influencing factors based on the average value of the multiple correlation coefficients.

[0011] Furthermore, based on multiple key influencing factors and multiple operating data, representative days are determined from multiple time periods, including: obtaining multiple key operating data of multiple key influencing factors in different time periods from multiple operating data; performing cluster analysis on the multiple key operating data based on multiple data clustering centers corresponding to different time periods to obtain multiple clustering results; aggregating the multiple clustering results to obtain target clustering results; and screening representative days from multiple time periods based on the target clustering results.

[0012] Furthermore, the method also includes: using the elbow criterion to process multiple key operating data to determine the number of data clusters corresponding to different key influencing factors; based on the number of data clusters and multiple key operating data, determining multiple initial clustering centers corresponding to different time periods; clustering the multiple key operating data based on the multiple initial clustering centers to obtain multiple initial clustering results; adjusting the multiple initial clustering centers based on the multiple initial clustering results to obtain multiple data clustering centers corresponding to different time periods.

[0013] Furthermore, multiple operating data of the power system in multiple time periods are obtained, including: obtaining multiple initial operating data of the power system in multiple time periods, wherein the multiple initial operating data include multiple operating data; performing anomaly detection on the multiple initial operating data to obtain multiple data detection results, wherein the multiple data detection results are used to characterize whether there are rough data in the multiple initial operating data; removing the rough data contained in the multiple initial operating data based on the multiple data detection results to obtain multiple operating data.

[0014] According to another aspect of an embodiment of the present invention, there is also provided a device for selecting a representative day for line loss calculation, comprising: a data acquisition module, used to acquire a plurality of operating data of an electric power system in a plurality of time periods, wherein different time periods have a plurality of operating data of the same type; a factor analysis module, used to analyze the plurality of operating data to obtain a plurality of line loss influencing factors, wherein the line loss shadow factor is used to characterize the data type of the operating data that may affect the line loss calculation result of the electric power system; a factor selection module, used to input the plurality of operating data into a factor evaluation model based on the plurality of line loss influencing factors, and select a plurality of key influencing factors from the plurality of line loss influencing factors using the factor evaluation model; a representative day determination module, used to determine a representative day from a plurality of time periods based on the plurality of key influencing factors and the plurality of operating data, wherein the representative day is used to perform line loss calculation on the electric power system.

[0015] According to another aspect of an embodiment of the present invention, there is further provided an electronic device, comprising: a memory storing an executable program; and a processor for running the program, wherein the method in each embodiment of the present invention is executed when the program is running.

[0016] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium includes a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the methods in various embodiments of the present invention.

[0017] According to another aspect of an embodiment of the present invention, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the method in each embodiment of the present invention is implemented.

[0018] According to another aspect of an embodiment of the present invention, a computer program product is provided, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method in each embodiment of the present invention is implemented.

[0019] According to another aspect of the embodiments of the present invention, a computer program is further provided. When the computer program is executed by a processor, the methods in the embodiments of the present invention are implemented.

[0020] In an embodiment of the present invention, a method is adopted in which multiple operating data of an electric power system in different time periods are acquired; the multiple operating data are analyzed to obtain multiple line loss influencing factors; based on the multiple operating data, a factor evaluation model is guided to select multiple key influencing factors from the multiple line loss influencing factors; based on the multiple key influencing factors and the multiple operating data, a representative day is determined from multiple time periods. The management system can use the factor evaluation model to screen out key influencing factors with a high correlation with the line loss calculation results from multiple line loss influencing factors that will affect the line loss calculation results, and select representative days from different time periods according to the operating data corresponding to the key influencing factors. This can ensure the correlation between the selected representative days and the line loss calculation results of the electric power system, thereby achieving the purpose of improving the accuracy and reliability of the method for selecting representative days for substations, thereby achieving the technical effect that the selected representative days reflect the actual situation of the substation line loss to the greatest extent, and further solving the technical problem in the related art that the selected representative days reflect the line loss rate of the substation with low accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0022] Figure 1 is a flow chart of a method for selecting a representative day for line loss calculation according to an embodiment of the present invention;

[0023] Figure 2 is a schematic diagram of an optional Euclidean distance between station area data and cluster centers according to an embodiment of the present invention;

[0024] Figure 3 is a schematic diagram of an optional theoretical line loss calculation result according to an embodiment of the present invention;

[0025] Figure 4 is a detailed flow chart of a method for selecting a representative day for line loss calculation according to an embodiment of the present invention;

[0026] Figure 5 4 is a schematic diagram of a device for selecting a representative day for line loss calculation according to an embodiment of the present invention. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0029] According to an embodiment of the present invention, an embodiment of a method for selecting a representative day for line loss calculation is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0030] Figure 1 is a flow chart of a method for selecting a representative day for line loss calculation according to an embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:

[0031] Step S102, obtaining a plurality of operation data of the power system in different time periods.

[0032] The above-mentioned different time periods may be time periods that can reflect the change cycle of the operation data of the power system. For example, the above-mentioned time period may be a line loss calculation period. Specifically, the above-mentioned time period may be a plurality of time periods obtained by dividing a month. For example, it may be a half-day, a day, a week and other time periods. The above-mentioned operation data may be data reflecting the working status of the power system. For example, it may be system internal data and system external data of the power system, but not limited to this. Among them, the above-mentioned system internal data includes information such as equipment archive data, operation collection data, but not limited to this. The above-mentioned system external data may include at least one or more of the following: date type, weather type, month, maximum temperature, minimum temperature, cloud opacity, but not limited to this.

[0033] In an optional embodiment, considering that the representative day can help the power system better analyze the trend of line loss changes and formulate loss reduction strategies, if the representative day is not selected, the line loss analysis will lack an effective reference point, which may make it difficult for the power system to identify and solve the key factors that cause line loss, thereby reducing the accuracy of line loss calculation. Further considering that the selection of the representative day is affected by multiple factors such as load level and grid operation status, and these factors vary significantly in different time periods, if data from different time periods is missing, the representative day may be selected in the absence of data support, resulting in errors in the selection of the representative day. Therefore, the power management system (hereinafter referred to as the management system) can first obtain the system internal data and system external data of the power system in different time periods.

[0034] For example, in order to accurately select representative days, thereby accurately reflecting the line loss rate within a specific time period and improving the accuracy of line loss calculations, the management system can obtain internal data of the substation system such as equipment archive data, operation collection data and other information. At the same time, the management system can also obtain external system data such as date type, weather type, month, maximum temperature, minimum temperature, cloud opacity, etc.

[0035] Step S104: Analyze multiple operating data to obtain multiple line loss impact factors, wherein the line loss impact factor is used to characterize the data type of the operating data that may affect the line loss calculation result of the power system.

[0036] The above-mentioned line loss influencing factors may be factors that directly affect the line loss in the power system, for example, they may be the power supply radius, the number of users, and the cabling rate, but are not limited thereto.

[0037] In an optional embodiment, considering that the operation and dispatching of the power system need to be based on the real-time load level and equipment status, if the line loss influencing factors cannot be accurately and comprehensively grasped, the dispatching decision may be erroneous, and the power supply and demand cannot be effectively balanced, resulting in low power grid operation efficiency. Therefore, the management system can analyze the above-mentioned multiple operating data to obtain the above-mentioned multiple line loss factors. The management system can more accurately predict the line loss of the power system based on the above-mentioned multiple line loss influencing factors, thereby improving the accuracy of line loss calculation.

[0038] For example, in order to clarify the effects of different line loss influencing factors on the line loss calculation of the substation system, the management system can classify the line loss influencing factors of the substation system into the following five types through the analysis of multiple operation data of the substation: management, technology, measurement, new energy access, and environment. The analysis of the impact of the above five types of line loss influencing factors on line loss calculation can be shown in Table 1.

[0039] Table 1

[0040]

[0041]

[0042]

[0043] Step S106: Based on the multiple operating data, the guiding factor evaluation model selects multiple key influencing factors from the multiple line loss influencing factors.

[0044] The factor evaluation model may be a model for extracting a plurality of key influencing factors from a plurality of line loss influencing factors. The key influencing factors may be line loss influencing factors having a relatively high degree of influence on line loss calculation.

[0045] In an optional embodiment, considering that the analysis results may be affected by multiple factors, if all factors that may be related but have a weak impact are included in the analysis, a large amount of noise will be introduced, which will increase the complexity of the analysis model and reduce the accuracy and reliability of the analysis. Therefore, the management system can first analyze the above-mentioned multiple operating data, obtain multiple line loss influencing factors, and use the above-mentioned multiple line loss influencing factors as inputs to the factor evaluation model. The factor evaluation model can further analyze multiple line loss influencing factors that have a greater impact on line loss calculation, and the management system can use the above-mentioned multiple line loss influencing factors with a greater impact as multiple key influencing factors.

[0046] For example, in order to identify the key influencing factors that have a significant impact on the line loss calculation of the substation system from multiple line loss influencing factors, the management system can first analyze multiple operating data of the substation system to obtain multiple line loss influencing factors of the substation system. Then, the management system can input the above multiple line loss influencing factors into the factor evaluation model of the substation system. The above factor evaluation model can select the top few factors with higher weights as key influencing factors through operations such as sensitivity analysis, grey correlation analysis and comprehensive weight calculation, thereby providing more accurate decision-making basis for subsequent decisions.

[0047] Step S108, determining a representative day from multiple time periods based on multiple key influencing factors and multiple operating data, wherein the representative day is used to calculate line losses for the power system.

[0048] The representative day mentioned above may be a day selected from historical records that can reflect the typical operating status or load characteristics of a specific period.

[0049] In an optional embodiment, considering that the selection of representative days is affected by many complex factors, by analyzing the above-mentioned multiple key influencing factors and the corresponding multiple operating data, the management system can more comprehensively understand the operating characteristics of the system in different periods, thereby improving the accuracy and representativeness of the selection of representative days. Therefore, the management system can analyze the above-mentioned multiple key influencing factors and multiple operating data to obtain the operating characteristics of the power system in different time periods, and then the power system can further select a representative day based on the above-mentioned operating characteristics.

[0050] For example, in order to intelligently combine multiple factors to accurately select the representative day of line loss of the substation system, the management system can pre-build an analysis and evaluation model for multiple factors. Then, the management system can input multiple operating data corresponding to the above multiple key influencing factors in different time periods into the above analysis and evaluation model to evaluate the operating characteristics of the substation system in different time periods to obtain evaluation results, which can reflect the possibility of the above different time periods becoming representative days. Finally, the management system can select a representative day based on the above evaluation results.

[0051] In an embodiment of the present invention, a method is adopted in which multiple operating data of an electric power system in different time periods are acquired; the multiple operating data are analyzed to obtain multiple line loss influencing factors; based on the multiple operating data, a factor evaluation model is guided to select multiple key influencing factors from the multiple line loss influencing factors; based on the multiple key influencing factors and the multiple operating data, a representative day is determined from multiple time periods. The management system can use the factor evaluation model to screen out key influencing factors with a high correlation with the line loss calculation results from multiple line loss influencing factors that will affect the line loss calculation results, and select representative days from different time periods according to the operating data corresponding to the key influencing factors. This can ensure the correlation between the selected representative days and the line loss calculation results of the electric power system, thereby achieving the purpose of improving the accuracy and reliability of the method for selecting representative days for substations, thereby achieving the technical effect that the selected representative days reflect the actual situation of the substation line loss to the greatest extent, and further solving the technical problem in the related art that the selected representative days reflect the line loss rate of the substation with low accuracy.

[0052] Furthermore, based on multiple operating data, the guiding factor evaluation model selects multiple key influencing factors from multiple line loss influencing factors, including: performing data standardization processing on multiple operating data based on multiple line loss influencing factors to obtain standard operating data corresponding to different line loss influencing factors; performing sensitivity analysis on multiple line loss influencing factors based on the standard operating data to obtain multiple sensitivity weights, and performing multi-factor correlation analysis on multiple line loss influencing factors based on the standard operating data to obtain multiple grey correlation weights, wherein different sensitivity weights are used to reflect the degree of influence of different line loss influencing factors on line loss calculation results, and different grey correlation weights are used to reflect the degree of correlation between different line loss influencing factors and line loss calculation results, and multiple line loss influencing factors, multiple sensitivity weights and multiple grey correlation weights correspond to each other one by one; multiple line loss influencing factors are evaluated based on multiple sensitivity weights and multiple grey correlation weights to obtain multiple evaluation indicators; and multiple key influencing factors are determined from multiple line loss influencing factors based on multiple evaluation indicators.

[0053] The above-mentioned standardization processing may be a process for converting data of different ranges and dimensions to the same scale. The above-mentioned standard operating data may be operating data that has been standardized. The above-mentioned sensitivity analysis may be an analysis for determining the degree of influence of each influencing factor on the line loss rate. The above-mentioned sensitivity weight may be a quantitative indicator for measuring the degree of influence of the above-mentioned multiple line loss influencing factors on the line loss calculation results. The above-mentioned multi-factor association analysis may be an analysis for determining the correlation and weight coefficient between each line loss influencing factor and the line loss. The above-mentioned grey correlation weight may be a quantitative indicator for evaluating the degree of correlation between the line loss influencing factor and the line loss calculation result. The above-mentioned evaluation index may be an indicator for determining the above-mentioned multiple key influencing factors.

[0054] In an optional embodiment, considering that the determination of key influencing factors is closely related to the operation data of the power system, if the above operation data is not processed and analyzed, the selection results of key influencing factors may be biased, thereby causing inaccuracy in the subsequent decision-making process. Therefore, the management system can first perform data standardization on multiple operation data based on multiple line loss influencing factors to obtain standard operation data to convert data of different ranges and dimensions to the same scale. Then, the management system can perform sensitivity analysis on multiple line loss influencing factors based on standard operation data to obtain multiple sensitivity weights to determine the degree of influence of different line loss influencing factors on the line loss calculation results. The management system can also perform multi-factor correlation analysis on multiple line loss influencing factors based on standard operation data to obtain multiple gray correlation weights to determine the degree of correlation between different line loss influencing factors and line loss calculation results. Through the above processing and analysis, the management system can evaluate multiple line loss influencing factors based on multiple sensitivity weights and multiple gray correlation weights to obtain multiple evaluation indicators. Finally, the management system can determine multiple key influencing factors from multiple line loss influencing factors based on multiple evaluation indicators.

[0055] For example, in order to unify data of different ranges and dimensions in the substation system to the same scale, the management system can first perform dimensionless data processing, including: calculating positive influencing factors and negative influencing factors. Among them, the calculation process of positive influencing factors can be shown as follows:

[0056]

[0057] The calculation process of reverse influencing factors can be shown as follows:

[0058]

[0059] Among them, h ij represents the standardized value of the influencing factor in the i-th row and j-th column; d ij is the influencing factor value of the i-th row and j-th column; max i (d ij ) is the maximum value of the influencing factor in the i-th row and j-th column, min i (d ij ) is the minimum value of the influencing factor in the i-th row and j-th column.

[0060] After the management system completes the standardization of the substation system operation data, it obtains the standard operation data of the substation system. The management system can use the Sobol sensitivity analysis method to perform sensitivity analysis and calculation on multiple line loss influencing factors based on the above standard operation data to determine the degree of influence of the above multiple influencing factors on the line loss rate, that is, to determine the sensitivity weight. The management system can also determine the correlation between line loss and the above multiple line loss influencing factors by establishing a gray correlation model, and further obtain the weight coefficients of the above multiple line loss influencing factors based on the above correlation, that is, to determine the gray correlation weight. Finally, the management system can determine the comprehensive weight of the above multiple influencing factors based on the above sensitivity weight and gray correlation weight, and extract the top 10 influencing factors ranked by the above comprehensive weight as key influencing factors for the selection of representative days.

[0061] Furthermore, a sensitivity analysis is performed on multiple line loss influencing factors based on standard operating data to obtain multiple sensitivity weights, including: determining the variance, covariance and joint variance corresponding to different line loss influencing factors based on the standard operating data; summarizing the variance, covariance and joint variance corresponding to different line loss influencing factors to obtain the total variance; obtaining the partial variance corresponding to different line loss influencing factors based on the quotient of the variance corresponding to different line loss influencing factors and the total variance; summarizing the partial variances corresponding to different line loss influencing factors to obtain the total partial variance; obtaining multiple sensitivity weights based on the quotient of the absolute value of the partial variance corresponding to different line loss influencing factors and the total partial variance.

[0062] In an optional embodiment, considering the large number of line loss influencing factors, if these line loss influencing factors are analyzed based on the same weight, it may not be possible to accurately identify which line loss influencing factors are most critical to the final line loss calculation results, so these key line loss influencing factors may be ignored during line loss calculation, resulting in reduced accuracy of line loss calculation results. Therefore, it is particularly important to calculate the sensitivity weights of the above-mentioned multiple line loss influencing factors. The management system can calculate the sensitivity weights of the above-mentioned multiple line loss influencing factors using the following steps: First, the management system can determine the variance, covariance, and joint variance corresponding to different line loss influencing factors based on standard operating data. Then, the management system can summarize the variance, covariance, and joint variance corresponding to the above-mentioned different line loss influencing factors to obtain the total variance. After obtaining the above-mentioned total variance, the management system can obtain the partial variance corresponding to the different line loss influencing factors based on the quotient of the variance corresponding to the different line loss influencing factors and the total variance, and then summarize the partial variances of the above-mentioned different line loss influencing factors to obtain the total partial variance. Finally, the management system may obtain the above-mentioned multiple sensitivity weights based on the quotient of the absolute value of the partial variance corresponding to the above-mentioned different line loss influencing factors and the total partial variance.

[0063] For example, in order to decompose the variance of the line loss rate into the contribution of each influencing factor and its interaction, so as to determine the degree of influence of each influencing factor on the line loss rate, the management system can use the Sobol sensitivity analysis method to perform sensitivity analysis calculation. The management system can set the analysis model as shown in the following formula:

[0064] y=f(x1,x2,…x d ).

[0065] Among them, y is the line loss rate, x1, x2, ... x d is the line loss influencing factor. Based on the above formula, the management system can obtain the total output variance as shown in the following formula:

[0066]

[0067] Where V(y) represents the total variance, d is the number of line loss influencing factors, V i is the variance of the i-th line loss influencing factor, V ij is the covariance of the i-th and j-th line loss influencing factors, V i,j,…,d is the joint variance of all line loss influencing factors. The total variance V(y) represents the impact of all line loss influencing factors on the line loss rate of the objective function, and the partial variance V i It represents the influence of the above-mentioned multiple line loss influencing factors on the line loss rate of the objective function, and the covariance V ij It represents the influence of the interaction between the above multiple influencing factors on the line loss rate of the objective function. The management system can define the proportion of the partial variance to the total variance as the sensitivity of the line loss influencing factor, as shown in the following formula:

[0068]

[0069] Among them, S i represents the sensitivity of the i-th line loss factor, V(Y|X i ) represents the variance of the model output when the value of the i-th line loss influencing factor is given, and V(Y) represents the total variance of the model output. Based on the above formula, the management system can derive the calculation process of the sensitivity weight as shown below:

[0070]

[0071] Among them, a i represents the sensitivity weight of the i-th line loss influencing factor, |S i | represents the absolute value of the sensitivity of the i-th line loss factor, represents the total sensitivity, that is, the sum of the absolute values ​​of the sensitivities of the 1st to jth line loss influencing factors.

[0072] Furthermore, a multi-factor correlation analysis is performed on multiple line loss influencing factors based on standard operating data to obtain multiple grey correlation weights, including: based on the system identification of the power system, obtaining multiple reference operating data corresponding to the multiple line loss influencing factors, wherein the multiple line loss influencing factors correspond to the multiple reference operating data one-to-one; determining multiple correlations corresponding to the multiple line loss influencing factors based on the multiple reference operating data and the standard operating data corresponding to different reference operating data; obtaining a total correlation based on the sum of the multiple correlations; and obtaining multiple grey correlation weights based on the quotient of the multiple correlations and the total correlation.

[0073] The system identifier may be an identifier for distinguishing different power systems. The reference operation data may be data for reflecting line loss rate characteristics. The correlation degree may be a correlation degree between the plurality of line loss influencing factors and the line loss calculation result.

[0074] In an optional embodiment, considering the large number of line loss influencing factors, if the importance of multiple line loss influencing factors is not judged, it may cause the power system resources to be misallocated, thereby reducing the accuracy of the power system line loss analysis. Therefore, it is particularly important to conduct a multi-factor correlation analysis on the above-mentioned multiple line loss influencing factors. The management system can calculate the above-mentioned multiple gray correlation weights according to the following steps: First, the management system can obtain the system identification of the current power system, and then obtain the multiple reference operating data corresponding to the above-mentioned multiple line loss influencing factors based on the above-mentioned system identification. Then, the management system can determine the multiple correlations corresponding to the multiple line loss influencing factors based on the above-mentioned multiple reference operating data and their corresponding standard operating data. Finally, the management system can sum the above-mentioned multiple correlations to obtain the total correlation, and the management system can calculate the above-mentioned multiple gray correlation weights based on the quotient of the above-mentioned multiple correlations and the total correlation.

[0075] For example, in order to calculate the correlation between line loss and each line loss influencing factor by establishing a grey correlation model, and to obtain the weight coefficient of each line loss influencing factor according to the correlation, the management system can adopt the grey correlation analysis method, and the steps of the method are as follows: First, the management system can obtain the system identification of the current substation system, and then obtain multiple reference operating data corresponding to multiple line loss influencing factors of the above substation system based on the above system identification. Then, the management system can determine the multiple correlations corresponding to the above multiple line loss influencing factors based on the above multiple reference operating data and their corresponding standard operating data. The calculation process of the above multiple correlations can be shown as follows:

[0076]

[0077] Among them, r i is the correlation degree of the i-th line loss influencing factor, and N represents the number of all line loss influencing factors. It represents the correlation coefficient of all line loss influencing factors, that is, the sum of the correlation coefficients of the 1st to the Nth line loss influencing factors.

[0078] At this point, the management system can further calculate the correlation weight, and the calculation process can be shown as follows:

[0079]

[0080] Among them, β i is the grey correlation weight of the i-th line loss influencing factor, r i is the correlation degree of the i-th line loss influencing factor. is the total correlation of all line loss influencing factors, that is, the sum of the correlations of the 1st to the nth line loss influencing factors.

[0081] Furthermore, based on multiple reference operating data and standard operating data corresponding to different reference operating data, multiple correlations corresponding to multiple line loss influencing factors are determined, including: obtaining the difference between different reference operating data and the corresponding standard operating data to obtain multiple difference values, wherein the multiple difference values ​​correspond one-to-one to the standard operating data; processing the maximum difference, the minimum difference, and multiple difference values ​​among the multiple difference values ​​according to a preset formula to obtain multiple correlation coefficients corresponding to different reference operating data; and determining the correlations corresponding to different line loss influencing factors based on the average value of the multiple correlation coefficients.

[0082] The preset formula may be a formula for calculating the correlation coefficient. The correlation coefficient may reveal the correlation and consistency between the different reference operation data and the standard operation data.

[0083] In an optional embodiment, considering that the calculation of the correlation coefficient is a prerequisite for the weight allocation in the grey correlation analysis, if the calculation of the correlation coefficient is inaccurate, the correlation of the above-mentioned multiple influencing factors cannot be calculated, and the weights cannot be further derived. Therefore, the management system can calculate the correlation coefficient and correlation of the above-mentioned multiple line loss influencing factors according to the following steps: The management system can first obtain the difference between different reference operating data and the corresponding standard operating data to obtain multiple difference values. Then, the management system can process the multiple differences and their maximum difference and minimum difference according to the above-mentioned preset formula, so as to obtain multiple correlation coefficients corresponding to the above-mentioned different reference operating data. Finally, the management system can average the above-mentioned multiple correlation coefficients, and the average value can be used as the correlation corresponding to different line loss influencing factors.

[0084] For example, in order to calculate multiple correlation coefficients corresponding to different reference operation data of the substation system, the reference operation data can be a reference series reflecting the characteristics of the line loss rate, and the standard operation data can be a comparative series affecting the line loss rate. The management system can first determine the reference series reflecting the characteristics of the line loss rate and the comparative series affecting the line loss rate, and then the management system can calculate the grey correlation coefficient of the reference series and the comparative series according to the following preset formula:

[0085]

[0086] Among them, ξ represents the grey relational coefficient, Δ(min) represents the minimum difference between two levels, Δ(max) represents the maximum difference between two levels, Δ oi (k) is the absolute difference between each point in the comparison series and each point in the reference series, and ρ is the resolution coefficient, which is usually 0.5.

[0087] Considering that the correlation coefficient is the correlation degree value between the comparison sequence and the reference sequence at each moment, there is more than one correlation coefficient. In order to facilitate the overall comparison, the management system can collect the correlation coefficients at each moment into one value, that is, calculate the average value, as the quantitative representation of the correlation degree between the comparison sequence and the reference sequence, that is, the correlation degree r i , the calculation method is shown in the above formula and will not be repeated here.

[0088] Furthermore, based on multiple key influencing factors and multiple operating data, representative days are determined from multiple time periods, including: obtaining multiple key operating data of multiple key influencing factors in different time periods from multiple operating data; performing cluster analysis on the multiple key operating data based on multiple data clustering centers corresponding to different time periods to obtain multiple clustering results; aggregating the multiple clustering results to obtain target clustering results; and screening representative days from multiple time periods based on the target clustering results.

[0089] The above key operation data may be data that has a significant impact on the operation of the power system. The above data cluster center may be a point within a cluster or a cluster that best represents the characteristics of all data points in the cluster.

[0090] In an optional embodiment, considering that the accuracy and reliability of line loss calculation are crucial to the operation and management of the power system, and line loss is affected by many factors, the traditional method of selecting representative days may not be able to fully reflect the actual situation of line loss, and there are limitations and deviations. Therefore, the management system can adopt a cluster analysis method to select representative days from multiple time periods. The management system can first obtain multiple key operation data of multiple key influencing factors in different time periods from multiple operation data as the basis for cluster analysis. Then, the management system can classify the key operation data, and determine multiple data clustering centers corresponding to different time periods by evaluating the cluster density, and perform cluster analysis on multiple key operation data based on the above multiple data clustering centers to obtain multiple clustering results. Finally, the management system can summarize the above multiple clustering results to obtain the target clustering results, and the management system can further screen representative days from multiple time periods based on the above target clustering results.

[0091] For example, in order to accurately select the representative day of line loss in the substation area, the management system can first perform K-means cluster analysis with the extracted key influencing factors as variables to divide the substation data into multiple clusters. Then, the management system can calculate the cluster center of each cluster and further determine the final cluster center. The management system can group the data by month and calculate the Euclidean distance between each day and the final cluster center. In order to make the line loss characteristics of the representative day of line loss relatively close to other dates and better reflect the average level of line loss in that month, the day closest to the cluster center in each month is used as the representative day.

[0092] Furthermore, the method also includes: using the elbow criterion to process multiple key operating data to determine the number of data clusters corresponding to different key influencing factors; based on the number of data clusters and multiple key operating data, determining multiple initial clustering centers corresponding to different time periods; clustering the multiple key operating data based on the multiple initial clustering centers to obtain multiple initial clustering results; adjusting the multiple initial clustering centers based on the multiple initial clustering results to obtain multiple data clustering centers corresponding to different time periods.

[0093] The elbow criterion may be a graphical method for determining the number of target clusters (number of clusters) in cluster analysis. The initial cluster centers may be pre-selected points representing the centers of each potential cluster group. The initial cluster results may be cluster results obtained after the data are first distributed based on the initial cluster centers.

[0094] In an optional embodiment, considering the increase in the number of clusters in the cluster analysis, over-segmentation may occur, resulting in unclear differences between clusters, and the clustering model becomes too complex, resulting in overfitting, which reduces the accuracy of the analysis. Therefore, the management system can first use the elbow criterion to process multiple key operating data to determine the number of data clusters corresponding to different key influencing factors. Then, the management system can determine multiple initial clustering centers corresponding to different time periods based on the number of data clusters and multiple key operating data. The management system can further cluster multiple key operating data based on the multiple initial clustering centers to obtain multiple initial clustering results. Finally, the management system can adjust the above-mentioned multiple initial clustering centers based on the above-mentioned multiple initial clustering results to obtain multiple data clustering centers corresponding to different time periods.

[0095] For example, in order to select the representative day of line loss in the substation area, the management system can first use the elbow criterion to determine the number of clusters K. The specific determination process is as follows: Assume that the load data set X is divided into K categories. According to experience, K is generally taken as For each K value, perform traditional k-means clustering and record the corresponding SSE, then draw a graph of the relationship between k and SSE, and finally select the K corresponding to the elbow as the target number of clusters. The above SSE formula is as follows:

[0096]

[0097] Among them, q i represents the cluster center, c i represents the cluster, x i It is c i Then, the management system can extract key operating data from multiple key influencing factors of the substation system, and then use the above key operating data as variables to perform K-means clustering analysis. The line loss sample set can be defined as: P{p1,p2,…p N}, the management system can select the number K of target clusters for clustering. Each cluster has a center. The management system can calculate the Euclidean distance between the remaining samples and the center of the cluster according to the following formula, and classify the sample into the cluster with the smallest distance:

[0098]

[0099] Among them, D(X,Y) represents the Euclidean distance, x i is the data in the dataset, y i is the cluster center of the corresponding indicator. The management system can also adjust the above cluster centers. The management system can iterate the initial cluster centers using the error square sum function criterion so that the cluster center with the minimum error square sum function is used as the final cluster center. The above error square sum function is as follows:

[0100]

[0101] Among them, n j is the number of samples in the jth class, m j is the sample mean of the jth class, indicating the data cluster center. The management system can select the class with the least number of clusters as the cluster center based on the cluster density. Cluster density is a measure of the intra-cluster variance. The smaller the variance, the higher the identity of the data set. Given a data set X, its intra-cluster variance is defined as:

[0102]

[0103] in, is the vector x i and The distance between them, N is the total number of samples in X, is the sample mean.

[0104] Specifically, The calculation can be shown as follows:

[0105]

[0106] For the clustering output results c1, c2, …, c n , the cluster density is defined as:

[0107]

[0108] Among them, C represents the density of clustering, M is the number of clusters, var(X i ) is the i-th cluster X i The internal variance of .

[0109] Furthermore, multiple operating data of the power system in multiple time periods are obtained, including: obtaining multiple initial operating data of the power system in multiple time periods, wherein the multiple initial operating data include multiple operating data; performing anomaly detection on the multiple initial operating data to obtain multiple data detection results, wherein the multiple data detection results are used to characterize whether there are rough data in the multiple initial operating data; removing the rough data contained in the multiple initial operating data based on the multiple data detection results to obtain multiple operating data.

[0110] The initial operation data may refer to the original operation data of the power system. The rough data may refer to data containing a rough error.

[0111] In an optional embodiment, considering that the initial operation data of the power system may contain distorted data, which are sample points far away from the cluster center, and may cause errors in the selection of representative days and the weight calculation of each characteristic index, the management system needs to remove the distorted data. The management system can first obtain multiple initial operation data of the power system in multiple time periods, and perform anomaly detection on the multiple initial operation data to check whether the multiple initial operation data contain data with gross errors. If the above data detection results show that the multiple initial operation data contain gross data, the management system can remove the gross data based on the above data detection results to obtain multiple operation data.

[0112] For example, in order to ensure the accuracy of the operating data, the management system can use the Laida criterion to identify the days of line loss rate distortion. Assuming that a set of test data contains only random errors, the management system can calculate the average value and residual error of the initial operating data based on the following formula:

[0113]

[0114] Among them, B ′ is the initial operation data, B j ′ s is the jth data in the initial running data, is the average value of the initial running data, n is the total number of data contained in the initial running data, T js is the residual error of the jth data. The management system can further calculate the standard deviation according to the following formula:

[0115]

[0116] Where δ is the standard deviation, n is the total number of data contained in the initial running data, when the residual error satisfies the following formula, it is considered to be a gross error rather than a random error. j ′ s Data should be removed:

[0117] |T js |>3δ.

[0118] For ease of understanding, an embodiment of the present invention is described by taking a certain substation as an example. The management system can obtain the equipment archive data of the current substation, for example, the operation collection data and other information in May 2023, as well as system external data, including: date type, weather type, month, maximum temperature, minimum temperature, cloud opacity and other information. Then, the management system performs sensitivity and gray correlation analysis on the 40 influencing factors, and selects the top ten influencing factors ranked by weight as key factors. The results are shown in Table 2 below:

[0119] Table 2

[0120]

[0121] The management system identifies the days with distorted line loss rate in the substation based on the above formula, and no distorted day data is found. Therefore, the management system can perform cluster analysis on the substation data based on the extracted key influencing factors. The number of clusters is determined to be 3 according to the elbow method, and the substation is divided into 3 categories. The initial cluster centers are shown in Table 3 below:

[0122] Table 3

[0123]

[0124] The management system can iterate the initial cluster center using the error square sum function criterion to make the cluster center with the minimum error square sum function as the final cluster center. The final cluster center after iteration is shown in Table 4 below:

[0125] Table 4

[0126]

[0127] The management system can select the cluster with the least number of clusters as the cluster center according to the cluster density, and thus select the third cluster center as the final cluster center. Furthermore, the management system can calculate the Euclidean distance between the data of the station area and the final cluster center every day, and take the day with the closest distance to the cluster center as the representative day. Figure 2 is a schematic diagram of an optional Euclidean distance between the station area data and the cluster center according to an embodiment of the present invention, wherein the horizontal axis represents the specific date of the month, and the vertical axis represents the Euclidean distance between the station area data and the cluster center. Figure 2 As shown in the figure, May 22 is closest to the cluster center at 19.01, so May 22 is used as the representative day for calculating the line loss in the substation area. In order to verify the rationality of the selection of the representative day, the theoretical line loss of the substation area for one consecutive month is calculated. Figure 3 is a schematic diagram of an optional theoretical line loss calculation result according to an embodiment of the present invention, wherein the horizontal axis represents the specific date of the month, and the vertical axis represents the theoretical line loss rate. Figure 3 As shown, according to the above theoretical line loss calculation results, the theoretical line loss rate on the representative day May 22 is 1.896%, and the average line loss rate this month is 1.912%. The selected representative day can reflect the actual line loss rate of the substation this month. It can be seen that the method provided by the present invention has certain practical significance and application value.

[0128] Figure 4 is a detailed flow chart of a method for selecting a representative day for line loss calculation according to an embodiment of the present invention. Figure 4As shown, the management system first collects the operation data of the substation system. Based on the collected operation data, it studies the factors affecting the line loss in the substation from five aspects: management, technology, measurement, new energy access, and environment, and analyzes the mechanism of its influence on the line loss. Then, the management system identifies the days with distorted line loss rate in the substation, and removes the identified distorted days to ensure the accuracy of the subsequent selection of representative days. Furthermore, the management system can standardize the above operation data, and based on the standardized data, extract the key influencing factors of the substation through sensitivity weight calculation and gray correlation weight calculation. Finally, the management system can select the representative day of line loss in the substation by calculating the distance between the substation data and the cluster center.

[0129] According to an embodiment of the present invention, an embodiment of a device for selecting a representative day for line loss calculation is provided. It should be noted that the device can be used to execute the method for selecting a representative day for line loss calculation. The specific implementation method and application scenario are the same as those in the above embodiment and will not be described in detail here.

[0130] Figure 5 is a schematic diagram of a device for selecting a representative day for line loss calculation according to an embodiment of the present invention. Figure 5 As shown, the device comprises:

[0131] The data acquisition module 502 is used to acquire a plurality of operation data of the power system in a plurality of time periods, wherein different time periods have a plurality of operation data of the same type.

[0132] The factor analysis module 504 is used to analyze multiple operating data to obtain multiple line loss impact factors, wherein the line loss impact factor is used to characterize the data type of the operating data that will affect the line loss calculation result of the power system.

[0133] The factor selection module 506 is used to input multiple operating data into the factor evaluation model based on multiple line loss influencing factors, and select multiple key influencing factors from the multiple line loss influencing factors using the factor evaluation model.

[0134] The representative day determination module 508 is used to determine a representative day from multiple time periods based on multiple key influencing factors and multiple operating data, wherein the representative day is used to calculate line losses for the power system.

[0135] Furthermore, the data acquisition module is also used to: based on multiple operating data, guide the factor evaluation model to select multiple key influencing factors from multiple line loss influencing factors, including: performing data standardization processing on multiple operating data based on multiple line loss influencing factors to obtain standard operating data corresponding to different line loss influencing factors; performing sensitivity analysis on multiple line loss influencing factors based on standard operating data to obtain multiple sensitivity weights, and performing multi-factor correlation analysis on multiple line loss influencing factors based on standard operating data to obtain multiple grey correlation weights, wherein different sensitivity weights are used to reflect the degree of influence of different line loss influencing factors on line loss calculation results, and different grey correlation weights are used to reflect the degree of correlation between different line loss influencing factors and line loss calculation results, and multiple line loss influencing factors, multiple sensitivity weights and multiple grey correlation weights correspond to each other one by one; multiple line loss influencing factors are evaluated based on multiple sensitivity weights and multiple grey correlation weights to obtain multiple evaluation indicators; and multiple key influencing factors are determined from multiple line loss influencing factors based on multiple evaluation indicators.

[0136] Furthermore, the factor analysis module is also used to: perform sensitivity analysis on multiple line loss influencing factors based on standard operating data to obtain multiple sensitivity weights, including: determining the variance, covariance and joint variance corresponding to different line loss influencing factors based on the standard operating data; summarizing the variance, covariance and joint variance corresponding to different line loss influencing factors to obtain the total variance; obtaining the partial variance corresponding to different line loss influencing factors based on the quotient of the variance corresponding to different line loss influencing factors and the total variance; summarizing the partial variances corresponding to different line loss influencing factors to obtain the total partial variance; obtaining multiple sensitivity weights based on the quotient of the absolute value of the partial variance corresponding to different line loss influencing factors and the total partial variance.

[0137] Furthermore, the factor analysis module is also used to: perform multi-factor correlation analysis on multiple line loss influencing factors based on standard operating data to obtain multiple grey correlation weights, including: based on the system identification of the power system, obtaining multiple reference operating data corresponding to multiple line loss influencing factors, wherein the multiple line loss influencing factors correspond one-to-one to the multiple reference operating data; based on the multiple reference operating data, and the standard operating data corresponding to different reference operating data, determining multiple correlations corresponding to the multiple line loss influencing factors; based on the sum of the multiple correlations, obtaining the total correlation; based on the quotient of the multiple correlations and the total correlation, obtaining multiple grey correlation weights.

[0138] Furthermore, the factor selection module is also used to: determine multiple correlations corresponding to multiple line loss influencing factors based on multiple reference operating data and standard operating data corresponding to different reference operating data, including: obtaining the difference between different reference operating data and the corresponding standard operating data to obtain multiple difference values, wherein the multiple difference values ​​correspond one-to-one to the standard operating data; processing the maximum difference, the minimum difference, and multiple difference values ​​among the multiple difference values ​​according to a preset formula to obtain multiple correlation coefficients corresponding to different reference operating data; and determining the correlations corresponding to different line loss influencing factors based on the average value of the multiple correlation coefficients.

[0139] Furthermore, the representative day determination module is also used to: determine the representative day from multiple time periods based on multiple key influencing factors and multiple operating data, including: obtaining multiple key operating data of multiple key influencing factors in different time periods from multiple operating data; performing cluster analysis on multiple key operating data based on multiple data clustering centers corresponding to different time periods to obtain multiple clustering results; summarizing multiple clustering results to obtain target clustering results; and screening representative days from multiple time periods based on the target clustering results.

[0140] Furthermore, the factor analysis module is also used to: process multiple key operating data using the elbow criterion to determine the number of data clusters corresponding to different key influencing factors; determine multiple initial clustering centers corresponding to different time periods based on the number of data clusters and multiple key operating data; cluster multiple key operating data based on the multiple initial clustering centers to obtain multiple initial clustering results; adjust the multiple initial clustering centers based on the multiple initial clustering results to obtain multiple data clustering centers corresponding to different time periods.

[0141] Furthermore, the data acquisition module is also used to: acquire multiple operating data of the power system in multiple time periods, including: acquiring multiple initial operating data of the power system in multiple time periods, wherein the multiple initial operating data include multiple operating data; performing anomaly detection on the multiple initial operating data to obtain multiple data detection results, wherein the multiple data detection results are used to characterize whether there are rough data in the multiple initial operating data; removing the rough data contained in the multiple initial operating data based on the multiple data detection results to obtain multiple operating data.

[0142] An embodiment of the present application further provides an electronic device, comprising: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods in various embodiments of the present invention when running.

[0143] An embodiment of the present application further provides a computer-readable storage medium, which includes a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the methods in various embodiments of the present invention.

[0144] An embodiment of the present application further provides a computer program product, including a computer program, which implements the methods in various embodiments of the present invention when executed by a processor.

[0145] An embodiment of the present application further provides a computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium is used to store a computer program, and when the computer program is executed by a processor, the method in each embodiment of the present invention is implemented.

[0146] The embodiments of the present application further provide a computer program, which implements the methods in the above-mentioned embodiments of the present invention when executed by a processor.

[0147] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0148] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0149] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0150] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0151] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0152] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for selecting a representative day for line loss calculation, characterized in that: include: Obtain multiple operating data of the power system in different time periods; Analyze the multiple operating data to obtain multiple line loss impact factors, wherein the line loss impact factor is used to characterize the data type of the operating data that will affect the line loss calculation result of the power system; Based on the plurality of operation data, guiding the factor evaluation model to select a plurality of key influencing factors from the plurality of line loss influencing factors; Based on the multiple key influencing factors and the multiple operating data, a representative day is determined from the multiple time periods, wherein the representative day is used to calculate line losses for the power system.

2. The method according to claim 1, characterized in that Based on the plurality of operating data, the guiding factor evaluation model selects a plurality of key influencing factors from the plurality of line loss influencing factors, including: Performing data standardization processing on the multiple operating data based on the multiple line loss impact factors to obtain standard operating data corresponding to different line loss impact factors; Performing sensitivity analysis on the multiple line loss influencing factors based on the standard operation data to obtain multiple sensitivity weights, and performing multi-factor association analysis on the multiple line loss influencing factors based on the standard operation data to obtain multiple gray association weights, wherein different sensitivity weights are used to reflect the degree of influence of different line loss influencing factors on the line loss calculation result, and different gray association weights are used to reflect the degree of association between different line loss influencing factors and the line loss calculation result, and the multiple line loss influencing factors, the multiple sensitivity weights and the multiple gray association weights correspond to each other one by one; Evaluate the multiple line loss influencing factors based on the multiple sensitivity weights and the multiple grey relational weights to obtain multiple evaluation indicators; The plurality of key influencing factors are determined from the plurality of line loss influencing factors based on the plurality of evaluation indicators.

3. The method according to claim 2, characterized in that A sensitivity analysis is performed on the multiple line loss influencing factors based on the standard operation data to obtain multiple sensitivity weights, including: Determine the variance, covariance and joint variance corresponding to different line loss influencing factors based on the standard operation data; Summarizing the variances, covariances, and joint variances corresponding to different line loss influencing factors to obtain a total variance; Obtaining partial variances corresponding to different line loss impact factors based on the quotient of the variances corresponding to different line loss impact factors and the total variance; Summarizing the partial variances corresponding to different line loss influencing factors to obtain a total partial variance; The plurality of sensitivity weights are obtained based on the quotient of the absolute value of the partial variance corresponding to different line loss influencing factors and the total partial variance.

4. The method according to claim 2, characterized in that: Based on the standard operation data, a multi-factor correlation analysis is performed on the multiple line loss influencing factors to obtain multiple grey correlation weights, including: Based on the system identifier of the power system, acquiring a plurality of reference operating data corresponding to the plurality of line loss impact factors, wherein the plurality of line loss impact factors correspond to the plurality of reference operating data one-to-one; Determining a plurality of correlation degrees corresponding to the plurality of line loss influencing factors based on the plurality of reference operating data and the standard operating data corresponding to different reference operating data; Based on the sum of the multiple association degrees, a total association degree is obtained; The plurality of grey relational degree weights are obtained based on the quotient of the plurality of relational degrees and the total relational degree.

5. The method according to claim 4, characterized in that Determining multiple correlations corresponding to the multiple line loss influencing factors based on the multiple reference operating data and the standard operating data corresponding to different reference operating data includes: Obtaining differences between different reference operating data and the corresponding standard operating data to obtain a plurality of difference values, wherein the plurality of difference values ​​correspond one-to-one to the standard operating data; Processing the maximum difference, the minimum difference, and the multiple differences among the multiple differences according to a preset formula to obtain multiple correlation coefficients corresponding to different reference operation data; Based on the average values ​​of the multiple correlation coefficients, correlation degrees corresponding to different line loss influencing factors are determined.

6. The method according to claim 1, characterized in that Determining a representative day from the multiple time periods based on the multiple key influencing factors and the multiple operating data includes: Acquire, from the plurality of operating data, a plurality of key operating data of the plurality of key influencing factors in different time periods; Based on multiple data clustering centers corresponding to different time periods, cluster analysis is performed on the multiple key operation data to obtain multiple clustering results; Summarizing the multiple clustering results to obtain a target clustering result; The representative day is filtered from the multiple time periods based on the target clustering result.

7. The method according to claim 6, characterized in that The method further comprises: Processing the plurality of key operating data using an elbow criterion to determine the number of data clusters corresponding to different key influencing factors; Based on the number of data clusters and the multiple key operating data, determining multiple initial cluster centers corresponding to different time periods; Clustering the plurality of key operating data based on the plurality of initial clustering centers to obtain a plurality of initial clustering results; The multiple initial clustering centers are adjusted based on the multiple initial clustering results to obtain the multiple data clustering centers corresponding to different time periods.

8. The method according to claim 1, characterized in that Obtain multiple operating data of the power system in multiple time periods, including: Acquire a plurality of initial operation data of the power system in the plurality of time periods, wherein the plurality of initial operation data includes the plurality of operation data; Performing anomaly detection on the multiple initial operation data to obtain multiple data detection results, wherein the multiple data detection results are used to indicate whether there is rough data in the multiple initial operation data; The plurality of operation data are obtained by removing rough data included in the plurality of initial operation data based on the plurality of data detection results.

9. A device for selecting a representative day for line loss calculation, characterized in that: include: A data acquisition module, used for acquiring a plurality of operation data of the power system in a plurality of time periods, wherein the plurality of operation data of the same type are present in different time periods; A factor analysis module, used for analyzing the plurality of operating data to obtain a plurality of line loss influencing factors, wherein the line loss shadow factor is used to characterize the data type of the operating data that will affect the line loss calculation result of the power system; A factor selection module, configured to input the plurality of operating data into a factor evaluation model based on the plurality of line loss influencing factors, and select a plurality of key influencing factors from the plurality of line loss influencing factors using the factor evaluation model; A representative day determination module is used to determine a representative day from the multiple time periods based on the multiple key influencing factors and the multiple operating data, wherein the representative day is used to calculate line losses for the power system.

10. An electronic device, characterized in that: include: A memory storing an executable program; A processor, configured to run the program, wherein the program executes the method according to any one of claims 1 to 8 when running.

Images