Method and device for determining line loss influence index of transformer area and storage medium

Abstract

This application discloses a method, apparatus, and storage medium for determining line loss impact indicators for a transformer substation. The method includes: acquiring target sample data for multiple transformer substations, wherein the target sample data includes at least external environmental data and internal operational data for the corresponding substations; based on the target sample data for multiple substations, performing dimensionality reduction on multiple initial line loss impact indicators to obtain multiple candidate line loss impact indicators, and determining the target indicator values ​​corresponding to each of the candidate line loss impact indicators; based on the target indicator values ​​corresponding to each of the candidate line loss impact indicators, determining the target weights corresponding to each of the candidate line loss impact indicators; and based on the target weights corresponding to each of the candidate line loss impact indicators, selecting the target line loss impact indicator from the multiple candidate line loss impact indicators. This application solves the technical problem of inaccurate determination results of line loss impact indicators for transformer substations in related technologies.

Description

Technical Field

[0001] This application relates to the field of power systems, and more specifically, to a method, apparatus, and storage medium for determining line loss impact indicators for a transformer substation. Background Technology

[0002] With the expansion of distribution network scale, the proportion of line losses in transformer substations is increasing, making line loss management a core aspect of power supply companies' efforts to improve quality and efficiency. In related technologies, the identification of line loss impact indicators in transformer substations mainly relies on theoretical line loss calculations or single correlation analysis, which has the following defects and shortcomings: First, there is a lack of hierarchical data governance. Due to the multi-source heterogeneous data and numerous outliers in the acquired transformer substation data, the lack of hierarchical quality governance leads to low accuracy in identifying line loss impact indicators and results that are not reproducible. Second, differentiated analysis is not implemented according to the power supply zone type to which the transformer substation belongs (e.g., A+, A, B, C, D, E), resulting in poor adaptability of the determined line loss impact indicators. Third, the dimensions of line loss impact indicators are incomplete, failing to distinguish transformer substation characteristics and not covering all dimensions of static, dynamic, renewable energy, and environmental features. Fourth, relying solely on single correlation or sensitivity analysis cannot accurately reflect the contribution of line loss impact indicators. Therefore, related technologies suffer from the technical problem of inaccurate determination of line loss impact indicators in transformer substations.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This application provides a method, apparatus, and storage medium for determining line loss impact indicators of a transformer substation, thereby at least solving the technical problem of inaccurate determination results of line loss impact indicators of transformer substations in related technologies.

[0005] According to one aspect of the embodiments of this application, a method for determining line loss impact indicators for a transformer substation is provided, comprising: acquiring target sample data for multiple transformer substations, wherein the target sample data includes at least external environmental data and internal operating data of the corresponding transformer substations; performing dimensionality reduction on multiple initial line loss impact indicators based on the target sample data for multiple transformer substations to obtain multiple candidate line loss impact indicators, and determining target indicator values ​​corresponding to the multiple candidate line loss impact indicators respectively; determining target weights corresponding to the multiple candidate line loss impact indicators based on the target indicator values ​​corresponding to the multiple candidate line loss impact indicators respectively; and selecting a target line loss impact indicator from the multiple candidate line loss impact indicators based on the target weights corresponding to the multiple candidate line loss impact indicators.

[0006] According to another aspect of the embodiments of this application, a device for determining line loss impact indicators of a transformer substation is provided, comprising: a data acquisition module, configured to acquire target sample data of multiple transformer substations, wherein the target sample data includes at least external environmental data and internal operating data of the corresponding transformer substation; a first selection module, configured to perform dimensionality reduction on multiple initial line loss impact indicators based on the target sample data of multiple transformer substations to obtain multiple candidate line loss impact indicators, and determine the target indicator values ​​corresponding to the multiple candidate line loss impact indicators respectively; a first determination module, configured to determine the target weights corresponding to the multiple candidate line loss impact indicators respectively based on the target indicator values ​​corresponding to the multiple candidate line loss impact indicators respectively; and a second selection module, configured to select the target line loss impact indicator from the multiple candidate line loss impact indicators based on the target weights corresponding to the multiple candidate line loss impact indicators respectively.

[0007] According to another aspect of the embodiments of this application, a non-volatile storage medium is provided, which stores multiple instructions, any one of which is adapted to be loaded by a processor to determine the line loss impact index of a transformer area.

[0008] According to another aspect of the embodiments of this application, an electronic device is provided, including: one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement any one of the following methods for determining the line loss impact index of a transformer area.

[0009] According to another aspect of the embodiments of this application, a computer program product is provided, which, when executed on a data processing device, is suitable for executing the steps of a method for determining the line loss impact index of a transformer area.

[0010] In this embodiment, target sample data from multiple transformer substations is acquired, including at least external environmental data and internal operational data for each substation. Based on this data, multiple initial line loss impact indicators are dimensionality-reduced to obtain multiple candidate line loss impact indicators, and target indicator values ​​are determined for each candidate indicator. Based on these target indicator values, target weights are determined for each candidate indicator. Finally, a target line loss impact indicator is selected from the candidate indicators based on their respective target weights. This method achieves the goal of determining multiple candidate line loss impact indicators from the acquired target sample data and initial line loss impact indicators, and further selecting the target indicator based on their respective target weights. This improves the accuracy of the target line loss impact indicator determination for transformer substations, thereby solving the technical problem of inaccurate line loss impact indicator determination in related technologies. Attached Figure Description

[0011] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0012] Figure 1 This is a flowchart of a method for determining line loss impact indicators of a transformer substation according to an embodiment of this application;

[0013] Figure 2 This is a flowchart of an optional method for determining line loss impact indicators of a transformer substation, according to an embodiment of this application.

[0014] Figure 3 This is a schematic diagram of a device for determining the line loss impact index of a transformer area according to an embodiment of this application;

[0015] Figure 4 This is a structural diagram of an electronic device provided according to an embodiment of this application. Detailed Implementation

[0016] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0017] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0018] It should be noted that the information and data collected in this application (including but not limited to initial sample data) are information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of this data all comply with relevant laws, regulations, and standards, and necessary confidentiality measures have been taken. This process does not violate public order and good morals, and corresponding access points are provided for users to choose whether to authorize or refuse. For example, interfaces are established between this system and relevant users or organizations, providing users with corresponding access points to choose whether to agree to or refuse automated decision-making results; if the user chooses to refuse, the process proceeds to the expert decision-making stage.

[0019] According to an embodiment of this application, a method embodiment for determining the line loss impact index of a transformer area is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0020] Figure 1 This is a flowchart of a method for determining line loss impact indicators of a transformer substation according to an embodiment of this application, such as... Figure 1 As shown, the method includes the following steps:

[0021] Step S102: Obtain target sample data for multiple transformer substations, wherein the target sample data includes at least the external environment data and internal operation data of the corresponding transformer substation.

[0022] It can be understood that multiple distribution substations refer to a collection of low-voltage distribution substations located within different power supply zones (e.g., the six power supply zones of A+, A, B, C, D, and E). In the general concept of this field, a distribution substation is the basic power supply unit of a distribution network. The use of multiple substations here is intended to reflect the comprehensiveness and diversity of the data collection objects, providing a basis for subsequent differential analysis.

[0023] Optionally, the target sample refers to a high-quality dataset obtained after preprocessing (such as cleaning and verification). Compared to the original collected data (i.e., the initial sample data), the target sample data is usable data after quality management. Its core characteristics are high quality and usability, excluding outliers, missing values, and logically incorrect data. External environment data refers to objective environmental factors that affect the operation of a distribution area but are not part of the internal physical structure or operating status of the power grid; these are uncontrolled variables. In power systems, this typically includes data on external disturbance factors such as weather, time, and holidays, used to assess the indirect impact of the external environment on line losses. Internal operating data refers to data reflecting the internal power grid architecture, equipment status, and electrical operating parameters of a distribution area; these are controlled or semi-controlled variables. Covering the physical attributes (such as topology and capacity) and operating status (such as load and power consumption) of the power grid, this data is the core basis for calculating line losses and identifying target line loss impact indicators.

[0024] In one alternative embodiment, multiple transformer substations belong to multiple power supply zones, and the multiple power supply zones include multiple power supply area types.

[0025] It can be understood that multiple transformer substations refer to a collection of substations located within multiple power supply zones. These multiple power supply zones include different power supply area types (e.g., six categories: A+, A, B, C, D, and E). By performing quantitative analysis on power supply zones of different power supply area types, it is possible to adapt to the significant differences in grid structure, load density, and operating characteristics of various substations, thereby improving the adaptability and rationality of the determination results of target line loss impact indicators.

[0026] In one optional embodiment, acquiring target sample data for multiple transformer substations includes: acquiring initial sample data for multiple transformer substations; performing a first-level screening of the initial sample data using sample integrity constraints and data validity constraints to obtain first sample data; performing a second-level screening of the first sample data using data rationality constraints to obtain second sample data; and performing a third-level screening of the second sample data using a clustering algorithm to obtain target sample data.

[0027] It is understandable that initial sample data refers to the raw, multi-source data set that has not undergone any cleaning or filtering process, including internal operational data and external environmental data. Logically, it is the preliminary state of the "target sample data," containing a large amount of noise, outliers, and missing values, and requires subsequent processing steps to be transformed into high-quality data.

[0028] Sample integrity constraints are constraints used to ensure that data fields are not null, unique, and conform to basic structural rules such as primary and foreign key relationships. They are a basic verification rule based on database structure and logical consistency, aiming to eliminate structurally missing or duplicate data caused by entry errors or system failures, ensuring basic data availability. Data validity constraints (e.g., operating years ≥ 0, power supply radius > 0, total length > 0, 0 < cable coverage ≤ 100%) are constraints used to ensure that numerical data is within a reasonable physical or business range. They are a verification rule based on numerical range logic, aiming to eliminate extreme outliers (such as negative numbers or excessively large values) caused by sensor failures or transmission errors, ensuring the physical validity of the data.

[0029] The first-level screening refers to the initial filtering process performed on the initial sample data based on sample integrity and data validity constraints. It is the foundational level in the data cleaning process, focusing on compliance checks of data structure and numerical ranges, and serves as a prerequisite for subsequent more complex screenings (such as rationality constraints and clustering). The first sample data refers to the dataset obtained after the first-level screening, which has removed structural missing data and obvious numerical anomalies. Compared to the initial sample data, its quality is improved, but it may still contain logical errors or data that violates business regulations.

[0030] Data rationality constraints refer to checking whether data conforms to specific business rules based on power business logic and physical laws (e.g., 0 < theoretical daily line loss rate ≤ 10%, |synchronous line loss rate - theoretical line loss rate| ≤ 5%, data collection success rate = 100%, number of users > 0, power supply area ≠ empty, 0 < trunk line cross-sectional area < 1250, power supply radius < total line length). Data rationality constraints are a deep verification rule based on domain knowledge, verifying not only simple numerical problems but also focusing on the logical relationships between data (e.g., input > output, synchronous and theoretical deviations) to discover hidden logical errors. Second-level filtering refers to a second filtering process on the first sample data based on data rationality constraints. It is an intermediate level in the data cleaning process, focusing on the consistency check of business logic, further eliminating data that, although numerically valid, has business logic errors. The second sample data refers to the data set obtained after the second-level filtering, after removing data with business logic violations. Compared to the first sample data, its business logic correctness is guaranteed, but it may still contain statistical outliers.

[0031] Clustering algorithms are used to identify outliers (outliers) that are far from the main distribution area of ​​the data, thus achieving refined data cleaning. Third-level filtering refers to a third filtering process based on clustering algorithms on the second sample data, typically involving the removal of outliers. It is a high-level step in the data cleaning process, focusing on anomaly detection in statistical distribution, aiming to remove outliers that, while numerically reasonable, are extreme in the overall distribution, thereby improving data quality and consistency.

[0032] It should be noted that the target sample data here refers to a high-quality, anomaly-free dataset obtained after three levels of screening (completeness, validity, rationality, and outlier removal from clustering). It is the final output of the entire data cleaning process and the input data for subsequent weight calculations, representing usable data after rigorous processing. By collecting initial sample data and combining it with three levels of screening, spikes and outliers in heterogeneous data can be effectively removed, solving the problems of uncontrollable data quality and non-reproducible analysis results.

[0033] Optionally, the six power supply zones of A+, A, B, C, D and E can be used as basic units to collect multi-source data (i.e. initial sample data) for low-voltage distribution areas to obtain initial sample data for multiple distribution areas. The collected multi-source data mainly includes two categories: internal network system data (i.e. internal operation data) and external environment data. The specific collection content is as follows: (1) Internal network system data collection. The internal network system data mainly comes from the distribution network management system and related business systems, including but not limited to: line loss related data, such as theoretical line loss data, daily line loss monitoring data, and synchronous line loss rate data; archive and topology data, such as equipment archive data (such as transformer, line parameters), and power grid topology data; operation monitoring data, such as load data, power data (including input power, sales power, etc.), and user collection success rate data. (2) External environment data collection. The external environment data mainly reflects the external natural and social environmental factors that affect the line loss of the distribution area, including but not limited to: meteorological data, such as weather type (such as sunny, cloudy, rainy and snowy), daily maximum temperature, and daily minimum temperature; time characteristic data, such as date type (which can be defined as working day or holiday). Through the above steps, a raw multi-source dataset covering the static characteristics, dynamic operating characteristics, new energy access status, external environment, and metering integrity of the transformer area can be obtained, providing comprehensive data support for subsequent data cleaning and determination of target line loss impact indicators.

[0034] Optionally, K-MEANS clustering analysis can be used to cluster the analysis objects (i.e., the second sample data), and then the distance between each analysis object (i.e., each sample point) and the cluster center can be determined to identify outliers. Sample points that deviate significantly from the cluster center often indicate data quality issues. Assume the second sample data is... , Where m represents the number of sample points in the second sample data, and i represents the i-th sample point. This represents the dimension of each sample point. Perform K-MEANS cluster analysis:

[0035] Randomly select k data points and divide them into k clusters. The initial cluster centers of the k clusters are as follows: , where j represents the j-th initial cluster center. Based on the initial cluster centers of each cluster, the distance between each sample point and these initial cluster centers is calculated, and the clusters are re-divided according to the minimum distance until all sample points are assigned. The minimum distance function is:

[0036]

[0037] in, This represents the minimum distance corresponding to the i-th sample point.

[0038] Recalculate the cluster centers for each cluster:

[0039]

[0040] in, This is an indicator function. The value of this function is 1 when the i-th sample point is assigned to the j-th cluster, and 0 otherwise.

[0041] Repeat the steps of "calculating the distance from the sample point to the cluster center and re-dividing the clusters" and "recalculating the cluster centers" until convergence, at which point the algorithm terminates. Based on the obtained clustering results, outliers are identified, and the top 2% of sample points in each cluster that deviate from the cluster center are removed, making the remaining clusters more compact and better reflecting the classification characteristics of the data.

[0042] Optionally, outlier detection and screening can be performed not only using cluster analysis but also based on the autovariability of sample points. The measured values ​​of a power system are constantly changing, but regardless of how the measured data changes, the underlying relationships remain constant, for example... ,power With current The direct proportional relationship remains unchanged, where, This represents phase. However, when abnormal data occurs, load changes abruptly, or the system structure changes, this relationship will inevitably be disrupted. Therefore, based on the constraints of these laws, the correlation of power system measurements can be used to detect and filter out abnormal data.

[0043] Variance reflects how much data deviates from the mean. In a stable system, the deviation of each sample point's measurement from the mean should be within a certain range. If it exceeds this range, it indicates a significant change in the data for that sample point. Therefore, variance can be used to detect variations in sample points. The formula for the variance of sample point a is:

[0044]

[0045]

[0046] in, denoted as , where is the measurement variance of sample point a (reflecting the dispersion of the data at that point); N is the total number of sample points. The measured value is for sample point a; Let a be the expected value of the measured value at sample point a, i.e., the mean.

[0047] When N=1, the autovariance of sample point a is: The autovariance reflects the extent to which the measured value of sample point a deviates from the mean. Sample points with an autovariance greater than a preset autovariance threshold are removed.

[0048] Step S104: Based on the target sample data of multiple transformer areas, the dimensionality of multiple initial line loss impact indicators is reduced to obtain multiple candidate line loss impact indicators, and the target indicator values ​​corresponding to the multiple candidate line loss impact indicators are determined respectively.

[0049] It can be understood that the initial line loss impact index refers to the complete set of all potential indicators that may affect the line loss value, extracted from theoretical analysis, historical experience, or preliminary feature engineering. Candidate line loss impact indices are obtained after dimensionality reduction of multiple initial line loss impact indices. Dimensionality reduction refers to using algorithms such as Principal Component Analysis (PCA) to recombine high-dimensional and correlated initial indices into a few mutually orthogonal candidate line loss impact indices that retain most of the information of the initial line loss impact indices, thereby eliminating redundancy and improving the stability of subsequent analysis. The target index value refers to the specific numerical value corresponding to the candidate line loss impact index; it is the quantitative representation of the index and also serves as a bridge connecting qualitative indicators and quantitative calculations, directly inputting data for subsequent weight calculations.

[0050] Optionally, principal component analysis (PCA) can be used to obtain candidate line loss impact indicators. After standardizing the target sample data, PCA is performed, retaining initial line loss impact indicators with a cumulative variance contribution rate ≥ 85% as candidate line loss impact indicators. When there is a certain correlation between two initial line loss impact indicators, it can be interpreted that the information reflected by these two factors overlaps to some extent. Therefore, using PCA, by recombining the original numerous initial line loss impact indicators, a few key combined indicators are extracted from the multiple initial line loss impact indicators, making them reflect the impact on line loss as much as possible (cumulative variance contribution rate ≥ 85%), and these are used as new candidate line loss impact indicators.

[0051] Step S106: Based on the target index values ​​corresponding to the multiple candidate line loss impact indicators, determine the target weights corresponding to the multiple candidate line loss impact indicators.

[0052] It can be understood that the target weight refers to the numerical value calculated based on the target index value of the candidate line loss impact index, which is used to quantify the contribution of each index to line loss.

[0053] In one optional embodiment, the target weights of the multiple candidate line loss impact indicators are determined based on their respective target indicator values. This includes: determining initial weights for the multiple candidate line loss impact indicators based on their respective target indicator values; determining sensitivity coefficients for the multiple candidate line loss impact indicators based on their respective target indicator values, wherein the sensitivity coefficients characterize the degree of influence of the corresponding candidate line loss impact indicator on the line loss value; and determining target weights for the multiple candidate line loss impact indicators based on their respective initial weights and sensitivity coefficients.

[0054] The initial weight, as understood, is a coefficient calculated based on the target index value of candidate line loss indices, reflecting the degree of correlation between the index and the line loss value. In this embodiment, the initial weight is one of the two dimensions for determining the target weight, representing the inherent importance or static contribution of the index to line loss. It can be obtained through methods such as correlation analysis and does not involve the impact of dynamic fluctuations. The sensitivity coefficient, calculated based on the target index value of candidate line loss indices, characterizes the degree of influence of changes in the index on the fluctuation of the line loss value. In this embodiment, the sensitivity coefficient is the other dimension for determining the target weight, representing the dynamic sensitivity or fluctuation contribution of the index to line loss, i.e., the degree of drastic change in the line loss value when the index undergoes a small change. The target weight is obtained by weighted summation of the initial weight and the sensitivity coefficient. It integrates the static correlation and dynamic sensitivity of the index, aiming to more comprehensively and accurately reflect the overall contribution of the index to line loss. By integrating the initial weight reflecting the importance of the index with the sensitivity coefficient characterizing the degree of influence, the rationality and accuracy of the target weight determination result are improved, thereby improving the accuracy of the target line loss indices determination result.

[0055] In one optional embodiment, the initial weights of the multiple candidate line loss impact indicators are determined based on the target indicator values ​​corresponding to the multiple candidate line loss impact indicators, including: normalizing the target indicator values ​​corresponding to the multiple candidate line loss impact indicators to obtain standard indicator values ​​corresponding to the multiple candidate line loss impact indicators; determining a sequence of correlation coefficients corresponding to the multiple candidate line loss impact indicators based on the standard indicator values ​​corresponding to the multiple candidate line loss impact indicators, wherein the sequence of correlation coefficients includes multiple correlation coefficients, which are used to represent the degree of correlation between the corresponding candidate line loss impact indicator and the line loss value under the corresponding target sample data conditions; and determining the initial weights of the multiple candidate line loss impact indicators based on the sequence of correlation coefficients corresponding to the multiple candidate line loss impact indicators.

[0056] Normalization, as we understand it, refers to the process of converting target indicator values ​​of different dimensions and orders of magnitude into dimensionless values ​​on a unified scale. It's a data preprocessing technique aimed at eliminating numerical scale differences between indicators caused by different units (e.g., length vs. electricity), making the subsequent correlation coefficients comparable. The standard indicator value refers to the dimensionless value of the candidate line loss impact indicator obtained after normalization; it's the direct output of the normalization operation. Compared to the target indicator value, the standard indicator value exists within a unified standard space and is the direct input for calculating the correlation coefficient. The correlation coefficient measures the degree of local correlation between the standard indicator value and the line loss value of a candidate line loss impact indicator under a single data point condition. It's a single element in the correlation coefficient sequence, quantifying the synchronicity or correlation between changes in the candidate line loss impact indicator and changes in line loss. Each correlation coefficient corresponds one-to-one with a sample point in the target sample data. The initial weight is the average of multiple correlation coefficients in the correlation coefficient sequence of each candidate line loss impact indicator, reflecting the overall correlation strength between the candidate line loss impact indicator and the line loss value. By determining the initial weights of candidate line loss impact indicators based on correlation coefficients, the nonlinear correlation between line loss values ​​and each candidate line loss impact indicator can be effectively captured, thereby objectively and accurately determining the initial weights of the indicators.

[0057] Optionally, grey relational analysis can be used to obtain the weighting coefficients of the initial line loss impact index, where the resolution coefficient is... The specific calculation steps are as follows:

[0058] (1) Determine the reference sequence reflecting the characteristics of system behavior and the comparison sequence affecting system behavior. The data sequence reflecting the characteristics of system line loss values ​​is called the reference sequence. The data sequence consisting of the target index values ​​of candidate line loss influencing indicators affecting system behavior is called the comparison sequence.

[0059] (2) Dimensionless processing is performed on the reference and comparison sequences. Because the physical meanings of the factors in the system are different, the dimensions of the data may also be different, making comparison difficult or leading to incorrect conclusions. Therefore, dimensionless data processing is necessary when performing grey relational analysis.

[0060] (3) Calculate the grey relational coefficient between the b-th candidate line loss impact index and the line loss value at the i-th sample point. The degree of correlation is essentially the degree of difference in the geometric shapes of the curves; the magnitude of the difference between the curves can be used as a measure of the degree of correlation. Correlation coefficient. It can be calculated using the following formula:

[0061]

[0062] in, The resolution coefficient, If the value is greater than 0, then 0.5 can be used. It is the second-order minimum difference; The maximum difference between the two levels; Let be the absolute difference between the candidate line loss influence index b and the line loss value at the i-th sample point.

[0063] (4) Calculate the correlation coefficient as the initial weight of the candidate line loss impact indicators. Since the correlation coefficient represents the degree of correlation between the candidate line loss impact indicators and the line loss value at each sample point, it has more than one value, and the information is too scattered to facilitate overall comparison. Therefore, it is necessary to concentrate the correlation coefficients corresponding to each sample point into a single value, i.e., calculate its average value, as a quantitative representation of the degree of correlation between the candidate line loss impact indicators and the line loss value. The correlation coefficient of the b-th candidate line loss impact indicator. The calculation formula is as follows:

[0064]

[0065] Where N is the total number of sample points, used to calculate the average.

[0066] (5) Correlation ranking. Arranging the correlation between multiple candidate line loss impact indicators and line loss values ​​in order of magnitude forms the correlation order, which reflects the "good and bad" relationship of each candidate line loss impact indicator with respect to the line loss value.

[0067] In one optional embodiment, the sensitivity coefficients corresponding to the multiple candidate line loss impact indicators are determined based on the target indicator values ​​corresponding to the multiple candidate line loss impact indicators, including: constructing a line loss prediction model based on the target indicator values ​​corresponding to the multiple candidate line loss impact indicators, wherein the line loss prediction model is used to characterize the mapping relationship between the target indicator values ​​corresponding to the multiple candidate line loss impact indicators and the line loss values; generating simulated indicator values ​​corresponding to the multiple candidate line loss impact indicators; obtaining predicted line loss values ​​by using the line loss prediction model based on the simulated indicator values ​​corresponding to the multiple candidate line loss impact indicators; and determining the sensitivity coefficients corresponding to the multiple candidate line loss impact indicators based on the predicted line loss values.

[0068] A line loss prediction model can be understood as a mathematical function or algorithmic structure whose input is the target index value of candidate line loss influencing indicators, and whose output is the corresponding line loss value. Simulated index values ​​refer to a set of virtual values ​​generated through random sampling or a specific algorithm (such as the Monte Carlo method). Unlike the target index values ​​(actual observed values), simulated index values ​​aim to cover the possible range and distribution characteristics of the indicators to observe the fluctuation of line loss values ​​under different input combinations, thereby quantifying the impact of candidate line loss influencing indicators. The predicted line loss value refers to the corresponding line loss result output by the line loss prediction model after inputting the simulated index values. The sensitivity coefficient is a coefficient determined based on the predicted line loss value, reflecting the degree of contribution of the indicator to the fluctuation of the line loss value. Compared to the initial weights (which focus on static correlation), the sensitivity coefficient focuses on reflecting the dynamic sensitivity of the indicator. By constructing a line loss prediction model and combining it with simulated index values, the independent contribution of the fluctuation of each candidate line loss influencing indicator to the line loss prediction result can be quantified, making up for the deficiency of static correlation analysis in reflecting dynamic sensitivity.

[0069] In one optional embodiment, based on the predicted line loss value, the sensitivity coefficients corresponding to multiple candidate line loss influencing indicators are determined, including: based on the predicted line loss value, determining the total variance and the conditional variances corresponding to the multiple candidate line loss influencing indicators, wherein the total variance is used to characterize the impact of the common changes of multiple candidate line loss influencing indicators on the fluctuation of the predicted line loss value, and the conditional variance is used to characterize the impact of changes only in the corresponding candidate line loss influencing indicators on the fluctuation of the predicted line loss value; based on the total variance and the conditional variances corresponding to the multiple candidate line loss influencing indicators, the sensitivity coefficients corresponding to the multiple candidate line loss influencing indicators are determined.

[0070] It is understandable that the total variance is used to characterize the impact of the combined changes of multiple candidate line loss indices on the volatility of the predicted line loss value. Conditional variance is the variance of the predicted line loss value when only one candidate line loss indice is changed, while all other input indices remain constant. The volatility impact refers to the measure of the range of change in the output line loss value caused by the change of any candidate line loss indice; the greater the volatility, the more significant the impact of any candidate line loss indice on the output. The sensitivity coefficient is the ratio of conditional variance to total variance; the larger this value, the more critical any candidate line loss indice is. By determining the conditional variances corresponding to multiple candidate line loss indices, the independent volatility impact of a single indicator after excluding interference from other factors can be accurately quantified, thereby improving the rationality and accuracy of the target weight determination results.

[0071] Optionally, Sobol sensitivity analysis can be used to obtain the sensitivity coefficients of candidate line loss indices. Sobol sensitivity analysis is used to determine the degree of influence of input variables on output variables. Based on the idea of ​​total variance decomposition, the variance of the output variable is decomposed into the contributions of each input variable and their interactions, thereby determining the degree of influence of each input variable on the output variable. The steps of Sobol sensitivity analysis are as follows:

[0072] Let the line loss prediction model be:

[0073] Where d represents the number of candidate line loss influencing indicators.

[0074] The total variance can be written as:

[0075] in, Let be the variance of the b-th candidate line loss impact index. Let b and c be the covariance of the line loss impact indicators. This is the joint variance of all candidate line loss impact indicators. Total variance This indicates the degree of fluctuation in the predicted line loss value caused by all candidate line loss impact indicators; covariance. This indicates the impact of the interaction between candidate line loss indices on the volatility of the predicted line loss value.

[0076] The proportion of the conditional variance bias of each candidate line loss impact indicator to the total variance is defined as the sensitivity coefficient of the candidate line loss impact indicator, and its formula is as follows:

[0077]

[0078] in, Let be the conditional variance of the b-th candidate line loss impact index.

[0079] Step S108: Based on the target weights corresponding to the multiple candidate line loss impact indicators, select the target line loss impact indicator from the multiple candidate line loss impact indicators.

[0080] It is understandable that the target line loss impact index refers to the set of key indicators that have a significant impact on line loss, selected from the candidate line loss impact indices.

[0081] Optionally, the target weights corresponding to multiple candidate line loss impact indicators are sorted in descending order, and a predetermined number of candidate line loss impact indicators with higher target weights are selected as target line loss impact indicators, or candidate line loss impact indicators with target weights greater than a preset weight threshold are selected as target line loss impact indicators.

[0082] Alternatively, the target weight of each candidate line loss impact indicator can be calculated using the following formula:

[0083]

[0084] in, and These are the initial weights and the weights corresponding to the sensitivity coefficients, respectively, and satisfy the following conditions: When calculating the target weight, one can take... .

[0085] Optionally, multiple candidate line loss impact indicators can be sorted according to the target weight, and the top 4 to 6 items can be selected as the target line loss impact indicators.

[0086] Through the above steps S102 to S108, multiple candidate line loss indices can be determined by acquiring target sample data from multiple transformer areas and multiple initial line loss indices. Based on the target weights corresponding to the multiple candidate line loss indices, the target line loss indices can be further selected from the multiple candidate line loss indices. This achieves the technical effect of improving the accuracy of the determination of the target line loss indices for transformer areas, thereby solving the technical problem of inaccurate determination of line loss indices for transformer areas in related technologies.

[0087] Based on the above embodiments and optional embodiments, this application proposes an implementation method for determining the line loss impact index of a transformer substation. This optional implementation method uses a low-voltage power grid in a certain region as the data collection object, and the data collection range is typical daily data for each month within a predetermined time period. The initial sample data collected includes 427,432 sample points, covering 89 power supply zones.

[0088] Figure 2 This is a flowchart of an optional method for determining line loss impact indicators for a transformer substation, provided according to an embodiment of this application. Figure 2 As shown, the implementation steps of the method for determining the line loss impact index of an optional transformer area include:

[0089] Step S1: Collect multi-source data based on power supply zones. Using six power supply zones (A+, A, B, C, D, and E) as units, collect multi-source data for each transformer area, including internal network system data (i.e., internal operating data) and external environmental data. Internal network data includes theoretical line loss, daily line loss; equipment records, topology; load, power consumption, and data collection success rate; external data includes weather, temperature, and weekday / holiday information.

[0090] Step S2, three-level data cleaning.

[0091] Step S21, the first level of data filtering is based on database constraints (including sample integrity constraints and data validity constraints), including NotNull, Unique, Primary Key, Foreign Key, and Check (referring to checking the data according to data validity constraints, such as operating years ≥ 0, power supply radius > 0).

[0092] The initial sample data is validated for integrity and normalization using the constraints inherent in the database. Sample integrity constraints primarily ensure data not null, uniqueness, correctness of primary and foreign key relationships, and the legality of numerical ranges. Data validity constraints may include, but are not limited to, constraints on operating years, power supply radius, total line length, and cable coverage rate. Database constraints are shown in Table 1.

[0093] Table 1 Database Constraints

[0094]

[0095] Step S22, the second level of data filtering is based on data rationality constraints (such as business rules, power grid related design specifications, etc.), such as 0% to 10% of the distribution area, |synchronous-theoretical| ≤ 5%, and 100% data collection success rate, etc.

[0096] Data anomaly identification based on business rules mainly utilizes power system business knowledge, combined with relevant power grid design specifications, to detect and filter data that violates power system business rules in various business systems. Data rationality constraints are shown in Table 2.

[0097] Table 2 Detection rules based on data rationality constraints

[0098]

[0099] Step S23: The second level of data filtering is based on the K-means clustering algorithm, which removes the farthest 2% of outlier sample points.

[0100] After the three-stage cleaning process described above, the remaining valid samples (i.e., target samples) for the transformer area are 269,816 sample points. K-MEANS clustering analysis is used to cluster the analysis objects (i.e., the second sample data), and then the distance between each analysis object (i.e., each sample point) and the cluster center is determined to identify outliers. Sample points that deviate significantly from the cluster center often indicate data quality issues. Assume the second sample data is... , Where m represents the number of sample points in the second sample data, and i represents the i-th sample point. This represents the dimension of each sample point. Perform K-MEANS cluster analysis:

[0101] Randomly select k data points and divide them into k clusters. The initial cluster centers of the k clusters are as follows: , where j represents the j-th initial cluster center. Based on the initial cluster centers of each cluster, the distance between each sample point and these initial cluster centers is calculated, and the clusters are re-divided according to the minimum distance until all sample points are assigned. The minimum distance function is:

[0102]

[0103] in, This represents the minimum distance corresponding to the i-th sample point.

[0104] Recalculate the cluster centers for each cluster.

[0105]

[0106] in, This is an indicator function. The value of this function is 1 when the i-th sample point is assigned to the j-th cluster, and 0 otherwise.

[0107] Repeat the steps of "calculating the distance from the sample point to the cluster center and re-dividing the clusters" and "recalculating the cluster centers" until convergence, at which point the algorithm terminates. Based on the obtained clustering results, outliers are identified, and the top 2% of sample points in each cluster that deviate from the cluster center are removed, making the remaining clusters more compact and better reflecting the classification characteristics of the data.

[0108] Not only can cluster analysis be used to detect and screen outliers, but outliers can also be detected and screened based on the autovariance of sample points.

[0109] The measured values ​​of a power system are constantly changing, but regardless of how the measured data changes, the underlying relationships remain unchanged. For example... ,power With current The direct proportional relationship remains unchanged, where, This represents phase. However, when abnormal data occurs, load changes abruptly, or the system structure changes, this relationship will inevitably be disrupted. Therefore, based on the constraints of these laws, the correlation of power system measurements can be used to detect and filter out abnormal data.

[0110] Variance reflects how much data deviates from the mean. In a stable system, the deviation of each sample point's measurement from the mean should be within a certain range. If it exceeds this range, it indicates a significant change in the data for that sample point. Therefore, variance can be used to detect variations in sample points. The formula for the variance of sample point a is:

[0111]

[0112]

[0113] in, denoted as , where is the measurement variance of sample point a (reflecting the dispersion of the data at that point); N is the total number of sample points. The measured value is for sample point a; Let a be the expected value of the measured value at sample point a, i.e., the mean.

[0114] When N=1, the autovariance of sample point a is: The autovariance reflects the extent to which the measured value of sample point a deviates from the mean. Sample points with an autovariance greater than a preset autovariance threshold are removed.

[0115] Step S3: Extract initial line loss impact indicators for the transformer substation. After obtaining high-quality target sample data, 19 initial line loss impact indicators were extracted based on the characteristics of the low-voltage transformer substation, covering five major categories: static, dynamic, new energy, environmental, and metering dimensions. The initial line loss impact indicators are shown in Table 3.

[0116] Table 3 Initial Line Loss Impact Indicators

[0117]

[0118]

[0119] Step S4: Principal component analysis (PCA) is used to obtain candidate line loss impact indicators. After standardizing the target sample data, PCA is performed, and initial line loss impact indicators with a cumulative variance contribution rate ≥ 85% are retained as candidate line loss impact indicators. When there is a certain correlation between two initial line loss impact indicators, it can be interpreted that the information reflected by these two factors overlaps to some extent. Therefore, using PCA, the original numerous initial line loss impact indicators are recombined, and a few key combined indicators are extracted from the multiple initial line loss impact indicators to reflect the impact on line loss as much as possible (cumulative variance contribution rate ≥ 85%), which serve as new candidate line loss impact indicators.

[0120] Step S5, two-dimensional calculation (initial weight and sensitivity calculation).

[0121] Step S51: Use grey relational analysis to obtain the weighting coefficients of the initial line loss impact index, where the resolution coefficient is... The specific calculation steps are as follows:

[0122] (1) Determine the reference sequence reflecting the characteristics of system behavior and the comparison sequence affecting system behavior. The data sequence reflecting the characteristics of system line loss values ​​is called the reference sequence. The data sequence consisting of the target index values ​​of candidate line loss influencing indicators affecting system behavior is called the comparison sequence.

[0123] (2) Dimensionless processing is performed on the reference and comparison sequences. Because the physical meanings of the factors in the system are different, the dimensions of the data may also be different, making comparison difficult or leading to incorrect conclusions. Therefore, dimensionless data processing is necessary when performing grey relational analysis.

[0124] (3) Calculate the grey relational coefficient between the b-th candidate line loss impact index and the line loss value at the i-th sample point. The degree of correlation is essentially the degree of difference in the geometric shapes of the curves; the magnitude of the difference between the curves can be used as a measure of the degree of correlation. Correlation coefficient. It can be calculated using the following formula:

[0125]

[0126] in, The resolution coefficient, If the value is greater than 0, then 0.5 can be used. It is the second-order minimum difference; The maximum difference between the two levels; Let be the absolute difference between the candidate line loss influence index b and the line loss value at the i-th sample point.

[0127] (4) Calculate the correlation coefficient as the initial weight of the candidate line loss impact indicators. Since the correlation coefficient represents the degree of correlation between the candidate line loss impact indicators and the line loss value at each sample point, it has more than one value, and the information is too scattered to facilitate overall comparison. Therefore, it is necessary to concentrate the correlation coefficients corresponding to each sample point into a single value, i.e., calculate its average value, as a quantitative representation of the degree of correlation between the candidate line loss impact indicators and the line loss value. The correlation coefficient of the b-th candidate line loss impact indicator. The calculation formula is as follows:

[0128]

[0129] Where N is the total number of sample points, used to calculate the average.

[0130] (5) Correlation ranking. Arranging the correlation between multiple candidate line loss impact indicators and line loss values ​​in order of magnitude forms the correlation order, which reflects the "good and bad" relationship of each candidate line loss impact indicator with respect to the line loss value.

[0131] Step S52, Sobol analysis: Based on variance decomposition, obtain the sensitivity coefficients of candidate line loss influencing indicators.

[0132] Sobol sensitivity analysis is used to determine the degree of influence of input variables on output variables. Based on the idea of ​​total variance decomposition, the variance of the output variable is decomposed into the contributions of each input variable and their interactions, thereby determining the degree of influence of each input variable on the output variable. The steps of Sobol sensitivity analysis are as follows:

[0133] Let the line loss prediction model be:

[0134] Where d represents the number of candidate line loss influencing indicators.

[0135] The total variance can be written as:

[0136] in, Let be the variance of the b-th candidate line loss impact index. Let b and c be the covariance of the line loss impact indicators. This is the joint variance of all candidate line loss impact indicators. Total variance This indicates the degree of fluctuation in the predicted line loss value caused by all candidate line loss impact indicators; covariance. This indicates the impact of the interaction between candidate line loss indices on the volatility of the predicted line loss value.

[0137] The proportion of the conditional variance bias of each candidate line loss impact indicator to the total variance is defined as the sensitivity coefficient of the candidate line loss impact indicator, and its formula is:

[0138]

[0139] in, Let be the conditional variance of the b-th candidate line loss impact index.

[0140] Step S6: Variable weight fusion + sorting to identify target line loss impact indicators.

[0141] The sensitivity coefficient and initial weight of each candidate line loss impact indicator can be calculated through the above process. Then, the target weight of each candidate line loss impact indicator can be calculated using the following formula:

[0142]

[0143] in, and These are the initial weights and the weights corresponding to the sensitivity coefficients, respectively, and satisfy the following conditions: When calculating the target weight, one can take... .

[0144] Multiple candidate line loss impact indicators are sorted according to target weights, and the top 4 to 6 are selected as target line loss impact indicators.

[0145] The above optional implementation methods achieve at least the following effects: By constructing a three-level progressive data cleaning mechanism of "completeness-effectiveness-rationality-clustering", outliers, spikes and logical errors are effectively removed, improving the data quality and reliability of sample data and laying a solid data foundation for subsequent accurate identification of target line loss impact indicators; by integrating the initial weights reflecting the importance of the indicators with the sensitivity coefficients representing the degree of impact, the rationality and accuracy of the target weight determination results are improved, thereby improving the accuracy of the target line loss impact indicator determination results.

[0146] It should be noted that the steps shown in the flowchart in 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 may be executed in a different order than that shown here.

[0147] This embodiment also provides a device for determining the line loss impact index of a transformer area. This device is used to implement the above embodiments and preferred embodiments, and will not be repeated for details already described. As used below, the terms "module" and "device" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0148] According to an embodiment of this application, an apparatus embodiment for implementing a method for determining line loss impact indicators of a transformer substation is also provided. Figure 3 This is a schematic diagram of a device for determining the line loss impact index of a transformer area according to an embodiment of this application, as shown below. Figure 3 As shown, the device for determining the line loss impact index of the above-mentioned transformer area includes a data acquisition module 302, a first selection module 304, a first determination module 306, and a second selection module 308. The device will be described below.

[0149] The data acquisition module 302 is used to acquire target sample data from multiple transformer substations, wherein the target sample data includes at least the external environment data and internal operation data of the corresponding transformer substation.

[0150] The first selection module 304 is connected to the data acquisition module 302. It is used to reduce the dimensionality of multiple initial line loss impact indicators based on the target sample data of multiple transformer areas, obtain multiple candidate line loss impact indicators, and determine the target indicator values ​​corresponding to the multiple candidate line loss impact indicators respectively.

[0151] The first determining module 306, connected to the first selecting module 304, is used to determine the target weights corresponding to the multiple candidate line loss impact indicators based on the target indicator values ​​corresponding to the multiple candidate line loss impact indicators respectively.

[0152] The second selection module 308, connected to the first determination module 306, is used to select the target line loss impact indicator from multiple candidate line loss impact indicators based on the target weights corresponding to the multiple candidate line loss impact indicators.

[0153] The device for determining line loss impact indicators of a transformer substation provided in this application embodiment, by setting a data acquisition module 302, a first selection module 304, a first determination module 306, and a second selection module 308, achieves the purpose of determining multiple candidate line loss impact indicators by acquiring multiple target sample data of multiple transformer substations and multiple initial line loss impact indicators, and further selecting the target line loss impact indicator from the multiple candidate line loss impact indicators according to the target weights corresponding to the multiple candidate line loss impact indicators, thereby improving the accuracy of the determination result of the target line loss impact indicator of the transformer substation and solving the technical problem of inaccurate determination result of line loss impact indicators of transformer substations in related technologies.

[0154] It should be noted that the above modules can be implemented by software or hardware. For example, for the latter, it can be implemented in the following ways: the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.

[0155] It should be noted that the data acquisition module 302, the first selection module 304, the first determination module 306, and the second selection module 308 mentioned above correspond to steps S102 to S108 in the embodiments. The instances and application scenarios implemented by the above modules and their corresponding steps are the same, but they are not limited to the content disclosed in the above embodiments. It should be noted that the above modules, as part of the device, can run in a computer terminal.

[0156] It should be noted that the optional or preferred implementation methods of this embodiment can be found in the relevant descriptions in the embodiments, and will not be repeated here.

[0157] The device for determining the line loss impact index of the aforementioned transformer area may also include a processor and a memory. The data acquisition module 302, the first selection module 304, the first determination module 306, the second selection module 308, etc., are all stored in the memory as program units, and the processor executes the aforementioned program units stored in the memory to realize the corresponding functions.

[0158] The processor contains a core that retrieves the corresponding program unit from memory. One or more cores may be configured. Memory may include non-persistent memory in computer-readable media, random access memory (RAM), and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory includes at least one memory chip.

[0159] This application provides a non-volatile storage medium storing a program that, when executed by a processor, implements a method for determining the line loss impact index of a transformer area.

[0160] This application provides an electronic device. Figure 4 This is a structural diagram of an electronic device provided according to an embodiment of this application. For example... Figure 4 As shown, the electronic device may include: one or more ( Figure 4 (Only one is shown in the document) Processor 402, memory 404, memory controller, and peripheral interface, wherein the peripheral interface is connected to the radio frequency module, audio module, and display. The electronic device includes a processor, memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs the following steps: acquiring target sample data from multiple transformer substations, wherein the target sample data includes at least the external environment data and internal operating data of the corresponding transformer substations; based on the target sample data from multiple transformer substations, performing dimensionality reduction on multiple initial line loss impact indicators to obtain multiple candidate line loss impact indicators, and determining the target indicator values ​​corresponding to each of the multiple candidate line loss impact indicators; based on the target indicator values ​​corresponding to each of the multiple candidate line loss impact indicators, determining the target weights corresponding to each of the multiple candidate line loss impact indicators; and based on the target weights corresponding to each of the multiple candidate line loss impact indicators, selecting the target line loss impact indicator from the multiple candidate line loss impact indicators. The device in this document can be a server, PC, etc.

[0161] This application also provides a computer program product, which, when executed on a data processing device, is suitable for executing an initialization program with the following method steps: acquiring target sample data for multiple transformer substations, wherein the target sample data includes at least external environmental data and internal operating data of the corresponding transformer substations; based on the target sample data for multiple transformer substations, performing dimensionality reduction on multiple initial line loss impact indicators to obtain multiple candidate line loss impact indicators, and determining the target indicator values ​​corresponding to the multiple candidate line loss impact indicators respectively; based on the target indicator values ​​corresponding to the multiple candidate line loss impact indicators respectively, determining the target weights corresponding to the multiple candidate line loss impact indicators respectively; and based on the target weights corresponding to the multiple candidate line loss impact indicators respectively, selecting the target line loss impact indicator from the multiple candidate line loss impact indicators.

[0162] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0163] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0164] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0165] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0166] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0167] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0168] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0169] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0170] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0171] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for determining the line loss impact index of a transformer substation, characterized in that, include: Acquire target sample data from multiple transformer substations, wherein the target sample data includes at least the external environment data and internal operation data of the corresponding transformer substation; Based on the target sample data of the multiple transformer areas, the dimensionality of multiple initial line loss impact indicators is reduced to obtain multiple candidate line loss impact indicators, and the target indicator values ​​corresponding to the multiple candidate line loss impact indicators are determined respectively. Based on the target index values ​​corresponding to the multiple candidate line loss impact indicators, the target weights corresponding to the multiple candidate line loss impact indicators are determined. Based on the target weights corresponding to the multiple candidate line loss impact indicators, the target line loss impact indicator is selected from the multiple candidate line loss impact indicators.

2. The method according to claim 1, characterized in that, The acquisition of target sample data from multiple transformer substations includes: Obtain initial sample data for the multiple transformer substations; The initial sample data is subjected to a first-level screening using sample integrity constraints and data validity constraints to obtain the first sample data. By applying data rationality constraints, the first sample data is subjected to a second level of screening to obtain the second sample data; A clustering algorithm is used to perform a third-level screening on the second sample data to obtain the target sample data.

3. The method according to claim 1, characterized in that, The step of determining the target weights corresponding to the multiple candidate line loss impact indicators based on their respective target indicator values ​​includes: Based on the target index values ​​corresponding to the multiple candidate line loss impact indicators, the initial weights corresponding to the multiple candidate line loss impact indicators are determined. Based on the target index values ​​corresponding to the multiple candidate line loss impact indicators, the sensitivity coefficients corresponding to the multiple candidate line loss impact indicators are determined, wherein the sensitivity coefficients are used to characterize the degree of influence of the corresponding candidate line loss impact indicators on the line loss value. Based on the initial weights corresponding to the multiple candidate line loss impact indicators and the sensitivity coefficients corresponding to the multiple candidate line loss impact indicators, the target weights corresponding to the multiple candidate line loss impact indicators are determined.

4. The method according to claim 3, characterized in that, The step of determining the initial weights corresponding to the multiple candidate line loss impact indicators based on their respective target indicator values ​​includes: The target index values ​​corresponding to the multiple candidate line loss impact indicators are normalized to obtain the standard index values ​​corresponding to the multiple candidate line loss impact indicators. Based on the standard index values ​​corresponding to the multiple candidate line loss impact indicators, a correlation coefficient sequence corresponding to the multiple candidate line loss impact indicators is determined. The correlation coefficient sequence includes multiple correlation coefficients, which are used to represent the degree of correlation between the corresponding candidate line loss impact indicator and the line loss value under the corresponding target sample data conditions. Based on the correlation coefficient sequences corresponding to the multiple candidate line loss impact indicators, the initial weights corresponding to the multiple candidate line loss impact indicators are determined.

5. The method according to claim 3, characterized in that, The step of determining the sensitivity coefficients corresponding to the multiple candidate line loss impact indicators based on their respective target indicator values ​​includes: Based on the target index values ​​corresponding to the multiple candidate line loss impact indicators, a line loss prediction model is constructed, wherein the line loss prediction model is used to characterize the mapping relationship between the target index values ​​and line loss values ​​corresponding to the multiple candidate line loss impact indicators. Generate simulated index values ​​corresponding to the multiple candidate line loss impact indicators; Based on the simulated index values ​​corresponding to the multiple candidate line loss impact indicators, the predicted line loss value is obtained using the line loss prediction model. Based on the predicted line loss value, the sensitivity coefficients corresponding to the multiple candidate line loss impact indicators are determined respectively.

6. The method according to claim 5, characterized in that, The step of determining the sensitivity coefficients corresponding to the plurality of candidate line loss impact indicators based on the predicted line loss value includes: Based on the predicted line loss value, the total variance and the conditional variances corresponding to the multiple candidate line loss influencing indicators are determined respectively. The total variance is used to characterize the impact of the common changes of the multiple candidate line loss influencing indicators on the fluctuation of the predicted line loss value, and the conditional variance is used to characterize the impact of the changes of only the corresponding candidate line loss influencing indicators on the fluctuation of the predicted line loss value. Based on the total variance and the conditional variances corresponding to the multiple candidate line loss impact indicators, the sensitivity coefficients corresponding to the multiple candidate line loss impact indicators are determined.

7. The method according to any one of claims 1 to 6, characterized in that, The multiple transformer substations belong to multiple power supply zones, and the multiple power supply zones include multiple power supply area types.

8. A device for determining the line loss impact index of a transformer substation, characterized in that, include: The data acquisition module is used to acquire target sample data from multiple transformer substations, wherein the target sample data includes at least the external environment data and internal operation data of the corresponding transformer substation. The first selection module is used to reduce the dimensionality of multiple initial line loss impact indicators based on the target sample data of the multiple transformer areas, obtain multiple candidate line loss impact indicators, and determine the target indicator values ​​corresponding to the multiple candidate line loss impact indicators respectively. The first determining module is used to determine the target weights corresponding to the multiple candidate line loss impact indicators based on the target indicator values ​​corresponding to the multiple candidate line loss impact indicators respectively; The second selection module is used to select the target line loss impact indicator from the multiple candidate line loss impact indicators based on the target weights corresponding to the multiple candidate line loss impact indicators.

9. A non-volatile storage medium, characterized in that, The non-volatile storage medium stores multiple instructions, which are adapted to be loaded by a processor and executed by the method for determining the line loss impact index of a transformer area as described in any one of claims 1 to 7.

10. An electronic device, characterized in that, include: One or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the method for determining the line loss impact index of a transformer area as described in any one of claims 1 to 7.

Images