Refined prediction method and system for transformer area load of electric power system

Abstract

The invention discloses a refined prediction method for power system transformer area loads. The method comprises the steps that historical data information of all transformer areas in a target power system is acquired; calculating to obtain a preliminary industry label of each transformer area, and summarizing and synthesizing to obtain a final industry label of each transformer area; obtaining a typical load curve of each transformer area by adopting a clustering algorithm, and clustering to obtain an expanded industry label under the final industry label; analyzing load characteristic change of each expansion industry label in different seasons and factors influencing the load change; generating a user portrait of each transformer area and performing correlation analysis; carrying out adaptive learning and evaluation on the load prediction model of each transformer area; and checking the load prediction model of each transformer area, and carrying out refined load prediction of the corresponding transformer area by adopting the finally checked load prediction model. The invention also discloses a system for realizing the fine prediction method of the power system area load. The method is higher in reliability and better in accuracy.

Description

Technical Field

[0001] The present invention belongs to the field of electrical automation, and in particular relates to a method and system for fine-grained prediction of load in a power system substation. Background Art

[0002] With the development of economy and technology and the improvement of people's living standards, electricity has become an indispensable secondary energy source in people's production and life, bringing endless convenience to people's production and life. Therefore, ensuring the stable and reliable supply of electricity has become one of the most important tasks of the power system.

[0003] At present, with the rapid development of new power systems, traditional low-voltage distribution substations are also developing towards a safer and smarter direction. In order to ensure the reliability, safety and intelligence of power system substations, the refined load prediction of power system substations is particularly important.

[0004] At present, most of the load forecasting schemes of power systems are based on the total amount of power systems, but no detailed load forecasting is carried out for power system substations. This situation makes the reliability and accuracy of load forecasting results of power system substations poor. Summary of the invention

[0005] One of the purposes of the present invention is to provide a method for fine-grained prediction of power system substation load with high reliability and good accuracy.

[0006] A second object of the present invention is to provide a system for implementing a refined prediction method for the power system substation load.

[0007] The refined prediction method for the power system area load provided by the present invention comprises the following steps:

[0008] S1. Obtain historical data information of each substation in the target power system;

[0009] S2. Calculate the preliminary industry labels for each area based on the historical data information obtained in step S1;

[0010] S3. Summarize the preliminary industry labels calculated in step S2 to obtain the final industry labels for each area;

[0011] S4. Based on the historical data information obtained in step S1, a clustering algorithm is used to obtain a typical load curve of each substation, and the typical load curve is clustered again to obtain an extended industry label under the final industry label;

[0012] S5. Analyze the load characteristic changes of each expansion industry label in different seasons and the factors affecting the load changes;

[0013] S6. Generate user portraits for each area based on the analysis results obtained in step S5, and perform correlation analysis;

[0014] S7. According to the data information obtained in step S6, adaptive learning and evaluation of the load forecasting model of each substation;

[0015] S8. Assess the load forecasting model of each substation obtained in step S7, and use the load forecasting model that finally passes the assessment to perform refined load forecasting for the corresponding substation.

[0016] Step S2, based on the historical data information obtained in step S1, calculates and obtains the preliminary industry label of each area, which specifically includes the following steps:

[0017] For public transformer areas:

[0018] The following formula is used to calculate the proportion of electricity consumption of each industry in the substation area:

[0019]

[0020] Where P i is the proportion of electricity consumption of the ith industry in the area; E i is the electricity consumption of the ith industry in the area; ∑ j E j It is the sum of electricity consumption of various industries in the area;

[0021] If the electricity consumption of the i-th industry in the substation accounts for P i The largest, then the preliminary industry label of the area is the industry label of the ith industry;

[0022] For the conversion area:

[0023] The preliminary industry label of the area is the industry label of the user;

[0024] The industry label described is a third-level industry label.

[0025] Step S3 summarizes and synthesizes the preliminary industry labels calculated in step S2 to obtain the final industry labels of each area, which specifically includes the following steps:

[0026] Determine the preliminary industry label calculated in step S2:

[0027] For the preliminary industry label k:

[0028] If the number of zones corresponding to the preliminary industry label k is greater than the set value, the preliminary industry label k is used as the final industry label of the corresponding zone;

[0029] If the number of areas corresponding to the preliminary industry label k is not greater than the set value, the secondary industry label corresponding to the preliminary industry label k is used as the final industry label of the corresponding area.

[0030] Step S4, based on the historical data information obtained in step S1, uses a clustering algorithm to obtain a typical load curve of each substation, and clusters the typical load curve again to obtain an extended industry label under the final industry label, specifically including the following steps:

[0031] According to the historical data information obtained in step S1, clustering the historical load data of the substation area is performed using a clustering algorithm to obtain a typical load curve of each substation area;

[0032] Matching the final industry label obtained in step S3 with the typical load curve of each substation;

[0033] The typical load curves under each final industry label are clustered again using a clustering algorithm to obtain the extended industry labels under each final industry label.

[0034] The analysis of the load characteristic changes of each extended industry tag in different seasons and the factors affecting the load changes described in step S5 specifically includes the following steps:

[0035] Analyze the changes in load characteristics of various expansion industry labels in different seasons:

[0036] Set SeasonS a The typical load curve of a typical area is C a , Season S b The typical load curve of a typical area is C b , then calculate C a and C b The Manhattan distance D ab for n is C a and C b The number of elements in, c ak C a The kth element in c bk C b The kth element in ;

[0037] Set SeasonS a The load curves at different time periods are C ai (t), t is the time period; C ai (t) Load change rate in each period ΔC ai (t) represents the load change trend,

[0038] Analyze the factors affecting load changes:

[0039] The Pearson correlation coefficient r is calculated using the following formula: xy , thus quantifying the impact of meteorological factors on different industries:

[0040]

[0041] Where x i To collect the influencing factors data at the time of collection; For x i The sample mean of i is the load data; for l i The sample mean of ;

[0042] The meteorological factors include humidity, temperature, rainfall and wind speed.

[0043] Step S6 generates user portraits for each substation based on the analysis results obtained in step S5, and performs correlation analysis, which specifically includes the following steps:

[0044] Combining the clustering result obtained in step S4 and the analysis result obtained in step S5, a user profile of each substation is generated;

[0045] Conduct correlation analysis: Conduct correlation analysis on meteorological factors for the same historical period; the meteorological factors include humidity, temperature, rainfall and wind speed.

[0046] Step S7, based on the data information obtained in step S6, performs adaptive learning and evaluation of the load forecasting model of each substation, specifically includes the following steps:

[0047] Feature Importance Analysis:

[0048] Perform feature importance analysis on the prediction results of different industry labels, and evaluate the correlation between each feature and the target variable based on mutual information;

[0049] The entropy H(X) of the characteristic factor X is expressed as p(x) is the probability of taking the value x;

[0050] The entropy H(Y) of the load forecast Y is expressed as p(y) is the probability of taking the value y;

[0051] The joint entropy of X and Y H(X,Y) is expressed as p(x,y) is the joint probability distribution of random variables X and Y, that is, the probability of observing X = x and Y = y at the same time;

[0052] The mutual information of X and Y, I(X;Y), is expressed as I(X;Y)=H(X)+H(Y)-H(X,Y);

[0053] The larger the mutual information, the stronger the correlation between X and Y;

[0054] Through mutual information calculation, the main influencing factors of load forecasting in each industry label are evaluated;

[0055] Adaptive Learning and Evaluation of Load Forecasting Models:

[0056] For the substations where the number of samples is less than the set value, the load forecasting models used include the support vector machine model and the kernel support vector machine model;

[0057] Support vector machine model: Given training samples D = {(x1,l1),(x2,l2),...,(x m ,l m )},x m is the mth sample, l m is the label corresponding to the mth sample, construct the model f(x)=wφ(x)+b, where f(x) is the label corresponding to the sample, w is the weight coefficient, φ(x) is the feature space, and b is the bias term;

[0058] Based on the principle of structural risk minimization, the objective function is set as

[0059]

[0060] Where R(x) is the objective function value; ||w|| is the weight coefficient; m is the number of samples; y i is the actual label value of the i-th sample; ε is the allowable deviation;

[0061] Introducing penalty coefficients and slack variables, the objective function is transformed into

[0062]

[0063]

[0064] In the formula is the objective function value; C is the regularization parameter; ξ i is the first slack variable, is the second slack variable; φ(x i ) is the final model function; m is the number of samples;

[0065] Using Lagrange duality theory to solve, we get

[0066]

[0067] In the formula is the first Lagrangian coefficient, a i is the second Lagrangian coefficient; K(x i,x) is the RBF kernel function;

[0068] For substations with volatility greater than the set value, the load forecasting models used include the random forest model, the gradient boosting model and the extreme random tree model.

[0069] Step S8 is to assess the load forecasting model of each substation obtained in step S7, and use the load forecasting model that finally passes the assessment to perform refined load forecasting for the corresponding substation, which specifically includes the following steps:

[0070] The following formula is used to calculate the accuracy Acc of the load forecasting model:

[0071]

[0072] Where n is the number of samples; Q is the load reference value; x i is the actual load value at the i-th moment; x f,i is the load forecast value at the i-th moment;

[0073] If the accuracy Acc is higher than or equal to the set value, the load forecasting model passes the assessment;

[0074] If the accuracy Acc is lower than the set value, the load forecasting model fails the assessment, and the process returns to step S1 to re-perform adaptive learning of the load forecasting model until the load forecasting model passes the assessment;

[0075] The load forecasting model that finally passes the assessment is used to carry out detailed load forecasting for the corresponding substation.

[0076] The present invention also provides a system for realizing the refined prediction method of the power system substation load, comprising a data acquisition module, a label calculation module, a label correction module, a data clustering module, a load analysis module, a user analysis module, a model learning module and a load prediction module; the data acquisition module, the label calculation module, the label correction module, the data clustering module, the load analysis module, the user analysis module, the model learning module and the load prediction module are connected in series in sequence; the data acquisition module is used to acquire the historical data information of each substation in the target power system, and upload the data information to the label calculation module; the label calculation module is used to calculate the preliminary industry label of each substation according to the received data information and the acquired historical data information, and upload the data information to the label correction module; the label correction module is used to summarize and synthesize the calculated preliminary industry labels according to the received data information, obtain the final industry labels of each substation, and upload the data information to the data clustering module; the data clustering module is used to summarize and synthesize the calculated preliminary industry labels according to the received data information, obtain the final industry labels of each substation, and upload the data information to the data clustering module The historical data information is obtained, and a clustering algorithm is used to obtain the typical load curve of each substation. The typical load curve is clustered again to obtain the extended industry label under the final industry label, and the data information is uploaded to the load analysis module; the load analysis module is used to analyze the load characteristic changes of each extended industry label in different seasons and the factors affecting the load change according to the received data information, and upload the data information to the user analysis module; the user analysis module is used to generate user portraits of each substation according to the received data information and the obtained analysis results, and perform correlation analysis, and upload the data information to the model learning module; the model learning module is used to perform adaptive learning and evaluation of the load forecasting model of each substation according to the received data information and the obtained data information, and upload the data information to the load forecasting module; the load forecasting module is used to assess the load forecasting model of each substation according to the received data information, and use the load forecasting model that finally passes the assessment to perform refined load forecasting for the corresponding substation.

[0077] The refined prediction method and system for the power system substation load provided by the present invention adaptively learns and evaluates the substation load prediction model based on the labels and portraits of the industry in which the substation is located; therefore, the present invention can not only realize the refined prediction of the power system substation load, but also has higher reliability and better accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 The figure is a schematic diagram of the method flow of the present invention.

[0079] Figure 2 Schematic diagram of the functional modules of the system of the present invention. DETAILED DESCRIPTION

[0080] like Figure 1 The method flow chart of the method of the present invention is shown as follows: the refined prediction method of the power system substation load disclosed by the present invention comprises the following steps:

[0081] S1. Obtain historical data information of each substation in the target power system;

[0082] S2. Based on the historical data information obtained in step S1, the preliminary industry labels of each area are calculated; specifically, the steps include:

[0083] For public transformer areas:

[0084] The following formula is used to calculate the proportion of electricity consumption of each industry in the substation area:

[0085]

[0086] Where P i is the proportion of electricity consumption of the ith industry in the area; E i is the electricity consumption of the ith industry in the area; ∑ j E j It is the sum of electricity consumption of various industries in the area;

[0087] If the electricity consumption of the i-th industry in the area accounts for P i The largest, then the preliminary industry label of the area is the industry label of the ith industry;

[0088] For the conversion area:

[0089] The preliminary industry label of the area is the industry label of the user;

[0090] The industry label mentioned is a third-level industry label;

[0091] S3. Summarize and synthesize the preliminary industry labels calculated in step S2 to obtain the final industry labels of each area; specifically, the steps include:

[0092] Determine the preliminary industry label calculated in step S2:

[0093] For the preliminary industry label k:

[0094] If the number of zones corresponding to the preliminary industry label k is greater than the set value, the preliminary industry label k is used as the final industry label of the corresponding zone;

[0095] If the number of zones corresponding to the preliminary industry label k is not greater than the set value, the secondary industry label corresponding to the preliminary industry label k is used as the final industry label of the corresponding zone;

[0096] S4. Based on the historical data information obtained in step S1, a clustering algorithm is used to obtain a typical load curve of each substation, and the typical load curve is clustered again to obtain an extended industry label under the final industry label; specifically, the following steps are included:

[0097] According to the historical data information obtained in step S1, clustering the historical load data of the substation area is performed using a clustering algorithm to obtain a typical load curve of each substation area;

[0098] Matching the final industry label obtained in step S3 with the typical load curve of each substation;

[0099] The typical load curves under each final industry label are clustered again using a clustering algorithm to obtain the extended industry labels under each final industry label;

[0100] S5. Analyze the load characteristic changes of each extended industry label in different seasons and the factors affecting the load changes; specifically, the steps include:

[0101] Analyze the changes in load characteristics of various expansion industry labels in different seasons:

[0102] Set SeasonS a The typical load curve of a typical area is C a , Season S b The typical load curve of a typical area is C b , then calculate C a and C b The Manhattan distance D ab for n is C a and C b The number of elements in, c ak C a The kth element in bk C b The kth element in ;

[0103] Set SeasonS a The load curves at different time periods are C ai (t), t is the time period; C ai (t) Load change rate in each period ΔC ai (t) represents the load change trend,

[0104] Analyze the factors affecting load changes:

[0105] The Pearson correlation coefficient r is calculated using the following formula: xy , thus quantifying the impact of meteorological factors on different industries:

[0106]

[0107] Where x i To collect the influencing factors data at the time of collection; For x i The sample mean of i is the load data; for l i The sample mean of ;

[0108] Said meteorological factors include humidity, temperature, rainfall and wind speed;

[0109] S6. Based on the analysis results obtained in step S5, generate user portraits for each area and perform correlation analysis; specifically comprising the following steps:

[0110] Combining the clustering result obtained in step S4 and the analysis result obtained in step S5, a user profile of each substation is generated;

[0111] Conduct correlation analysis: Conduct correlation analysis on meteorological factors for the same period in history; the meteorological factors include humidity, temperature, rainfall and wind speed;

[0112] S7. According to the data information obtained in step S6, adaptive learning and evaluation of the load forecasting model of each substation is performed; specifically comprising the following steps:

[0113] Feature Importance Analysis:

[0114] Perform feature importance analysis on the prediction results of different industry labels, and evaluate the correlation between each feature and the target variable based on mutual information;

[0115] The entropy H(X) of the characteristic factor X is expressed as p(x) is the probability of taking the value x;

[0116] The entropy H(Y) of the load forecast Y is expressed as p(y) is the probability of taking the value y;

[0117] The joint entropy of X and Y H(X,Y) is expressed as p(x,y) is the joint probability distribution of random variables X and Y, that is, the probability of observing X = x and Y = y at the same time;

[0118] The mutual information of X and Y, I(X;Y), is expressed as I(X;Y)=H(X)+H(Y)-H(X,Y);

[0119] The greater the mutual information, the stronger the correlation between X and Y; conversely, the weaker it is. This indicator can be used to perform correlation analysis on multidimensional data to find out the main influencing factors of load forecasting in each industry label area, providing strong support for subsequent modeling and forecasting.

[0120] Through mutual information calculation, the main influencing factors of load forecasting in each industry label are evaluated;

[0121] Adaptive Learning and Evaluation of Load Forecasting Models:

[0122] For the substations where the number of samples is less than the set value, the load forecasting models used include the support vector machine model and the kernel support vector machine model;

[0123] Support vector machine model: Given training samples D = {(x1,l1),(x2,l2),...,(x m ,l m )},x m is the mth sample, l m is the label corresponding to the mth sample, construct the model f(x)=wφ(x)+b, where f(x) is the label corresponding to the sample, w is the weight coefficient, φ(x) is the feature space, and b is the bias term;

[0124] Based on the principle of structural risk minimization, the objective function is set as

[0125]

[0126] Where R(x) is the objective function value; ||w|| is the weight coefficient; m is the number of samples; y i is the actual label value of the i-th sample; ε is the allowable deviation;

[0127] Introducing penalty coefficients and slack variables, the objective function is transformed into

[0128]

[0129]

[0130] In the formula is the objective function value; C is the regularization parameter; ξ i is the first slack variable, is the second slack variable; φ(x i ) is the final model function; m is the number of samples;

[0131] Using Lagrange duality theory to solve, we get

[0132]

[0133] In the formula is the first Lagrangian coefficient, a i is the second Lagrangian coefficient; K(x i ,x) is the RBF kernel function;

[0134] For the substations with volatility greater than the set value, the load forecasting models used include the random forest model, the gradient boosting model and the extreme random tree model;

[0135] The decision tree is a supervised learning algorithm suitable for classification and regression problems. It can extract decision rules based on a series of load-related factors and target values. The input of the decision tree model is the data of the training set. It continuously calculates the size of different values ​​under different influencing factors and uses the least squares method to find the optimal split point with the smallest square error, that is, to solve:

[0136]

[0137] Where c1 is the average value of the target variable in set R1(j,s); c2 is the average value of the target variable in set R2(j,s); y i is the target variable value of the i-th sample; x i is the i-th sample; R1(j,s) satisfies x (j) ≤s, and R1(j,s)={x|x (j) ≤s}, x (j) is the jth characteristic variable of x; R2(j,s) satisfies x (j) ≥s, and R2(j,s)={x|x (j) ≥s}; j is the jth characteristic variable in the data set; s is the split point;

[0138] Thus, the calculated values ​​in different areas are obtained, and the above steps are repeated for the sub-areas until the square error is minimized. Finally, the area is divided into several areas to form a regression decision tree. The random forest model selects the characteristics corresponding to the load in the data set and the load value at the current moment by random sampling with replacement based on the decision tree. Different decision trees correspond to different data sets. By constructing multiple decision trees and determining the final prediction based on the prediction results of each decision tree, the possibility of large errors in the prediction of a single decision tree is reduced. Random forest is like a way to make a decision after obtaining the load prediction output values ​​of all single decision trees, so it is better than decision trees in terms of accuracy.

[0139] S8. The load forecasting model of each substation obtained in step S7 is evaluated, and the load forecasting model that finally passes the evaluation is used to perform a refined load forecast for the corresponding substation; specifically, the following steps are included:

[0140] The following formula is used to calculate the accuracy Acc of the load forecasting model:

[0141]

[0142] Where n is the number of samples; Q is the load reference value; xi is the actual load value at the i-th moment; x f,i is the load forecast value at the i-th moment;

[0143] If the accuracy Acc is higher than or equal to the set value, the load forecasting model passes the assessment;

[0144] If the accuracy Acc is lower than the set value, the load forecasting model fails the assessment, and then returns to step S1 to re-perform adaptive learning of the load forecasting model until the load forecasting model passes the assessment; at this time, the low accuracy may be caused by factors such as algorithm inapplicability, incorrect industry labels, and missing load data; algorithm optimization is carried out for situations where the algorithm is not applicable; incorrect labels are mainly due to incorrect portrait results and incorrect initial labels, and user portrait algorithm optimization and initial industry label verification are carried out; load data supplementation is carried out for missing load data;

[0145] The load forecasting model that finally passes the assessment is used to carry out detailed load forecasting for the corresponding substation.

[0146] like Figure 2The functional module schematic diagram of the system of the present invention is shown as follows: the system disclosed in the present invention for realizing the refined prediction method of the load of the power system substation includes a data acquisition module, a label calculation module, a label correction module, a data clustering module, a load analysis module, a user analysis module, a model learning module and a load prediction module; the data acquisition module, the label calculation module, the label correction module, the data clustering module, the load analysis module, the user analysis module, the model learning module and the load prediction module are connected in series in sequence; the data acquisition module is used to obtain the historical data information of each substation in the target power system, and upload the data information to the label calculation module; the label calculation module is used to calculate the preliminary industry label of each substation according to the received data information and the acquired historical data information, and upload the data information to the label correction module; the label correction module is used to summarize and synthesize the calculated preliminary industry labels according to the received data information, obtain the final industry labels of each substation, and upload the data information to the data clustering module; the data clustering module is used to collect the received industry labels according to the received Data information, based on the historical data information obtained, a clustering algorithm is used to obtain the typical load curve of each substation, and the typical load curve is clustered again to obtain the extended industry label under the final industry label, and the data information is uploaded to the load analysis module; the load analysis module is used to analyze the changes in load characteristics of each extended industry label in different seasons and the factors affecting the load change according to the received data information, and upload the data information to the user analysis module; the user analysis module is used to generate user portraits of each substation according to the received data information and the obtained analysis results, and perform correlation analysis, and upload the data information to the model learning module; the model learning module is used to perform adaptive learning and evaluation of the load forecasting model of each substation according to the received data information and the obtained data information, and upload the data information to the load forecasting module; the load forecasting module is used to assess the load forecasting model of each substation based on the received data information, and use the load forecasting model that finally passes the assessment to perform refined load forecasting for the corresponding substation.

Claims

1. A refined prediction method for power system area load, comprising the following steps: S1. Obtain historical data information of each substation in the target power system; S2. Calculate the preliminary industry labels for each area based on the historical data information obtained in step S1; S3. Summarize the preliminary industry labels calculated in step S2 to obtain the final industry labels for each area; S4. Based on the historical data information obtained in step S1, a clustering algorithm is used to obtain a typical load curve of each substation, and the typical load curve is clustered again to obtain an extended industry label under the final industry label; S5. Analyze the load characteristic changes of each expansion industry label in different seasons and the factors affecting the load changes; S6. Generate user portraits for each area based on the analysis results obtained in step S5, and perform correlation analysis; S7. According to the data information obtained in step S6, adaptive learning and evaluation of the load forecasting model of each substation; S8. Assess the load forecasting model of each substation obtained in step S7, and use the load forecasting model that finally passes the assessment to perform refined load forecasting for the corresponding substation.

2. The method for fine-grained prediction of power system load according to claim 1 is characterized in that Step S2, based on the historical data information obtained in step S1, calculates and obtains the preliminary industry label of each area, which specifically includes the following steps: For public transformer areas: The following formula is used to calculate the proportion of electricity consumption of each industry in the substation area: Where P i is the proportion of electricity consumption of the ith industry in the area; E i is the electricity consumption of the ith industry in the area; ∑ j E j It is the sum of electricity consumption of various industries in the area; If the electricity consumption of the i-th industry in the area accounts for P i The largest, then the preliminary industry label of the area is the industry label of the ith industry; For the conversion area: The preliminary industry label of the area is the industry label of the user; The industry label described is a third-level industry label.

3. The method for fine-grained prediction of power system load according to claim 2 is characterized in that Step S3 summarizes and synthesizes the preliminary industry labels calculated in step S2 to obtain the final industry labels of each area, which specifically includes the following steps: Determine the preliminary industry label calculated in step S2: For the preliminary industry label k: If the number of zones corresponding to the preliminary industry label k is greater than the set value, the preliminary industry label k is used as the final industry label of the corresponding zone; If the number of areas corresponding to the preliminary industry label k is not greater than the set value, the secondary industry label corresponding to the preliminary industry label k is used as the final industry label of the corresponding area.

4. The method for fine-grained prediction of power system load according to claim 3 is characterized in that Step S4, based on the historical data information obtained in step S1, uses a clustering algorithm to obtain a typical load curve of each substation, and clusters the typical load curve again to obtain an extended industry label under the final industry label, specifically including the following steps: According to the historical data information obtained in step S1, clustering the historical load data of the substation area is performed using a clustering algorithm to obtain a typical load curve of each substation area; Matching the final industry label obtained in step S3 with the typical load curve of each substation; The typical load curves under each final industry label are clustered again using a clustering algorithm to obtain the extended industry labels under each final industry label.

5. The method for fine-grained prediction of power system load according to claim 4 is characterized in that The analysis of the load characteristic changes of each extended industry tag in different seasons and the factors affecting the load changes described in step S5 specifically includes the following steps: Analyze the changes in load characteristics of various expansion industry labels in different seasons: Set SeasonS a The typical load curve of a typical area is C a , Season S b The typical load curve of a typical area is C b , then calculate C a and C b The Manhattan distance D ab for n is C a and C b The number of elements in, c ak C a The kth element in c bk C b The kth element in ; Set SeasonS a The load curves at different time periods are C ai (t), t is the time period; C ai (t) Load change rate in each period ΔC ai (t) represents the load change trend, Analyze the factors affecting load changes: The Pearson correlation coefficient r is calculated using the following formula: xy , thus quantifying the impact of meteorological factors on different industries: Where x i To collect the influencing factors data at the time of collection; For x i The sample mean of i is the load data; for l i The sample mean of ; The meteorological factors include humidity, temperature, rainfall and wind speed.

6. The method for fine-grained prediction of power system load according to claim 5 is characterized in that Step S6 generates user portraits for each substation based on the analysis results obtained in step S5, and performs correlation analysis, which specifically includes the following steps: Combining the clustering result obtained in step S4 and the analysis result obtained in step S5, a user profile of each substation is generated; Conduct correlation analysis: Conduct correlation analysis on meteorological factors for the same historical period; the meteorological factors include humidity, temperature, rainfall and wind speed.

7. The method for fine-grained prediction of power system load according to claim 6 is characterized in that Step S7, based on the data information obtained in step S6, performs adaptive learning and evaluation of the load forecasting model of each substation, specifically includes the following steps: Feature Importance Analysis: Perform feature importance analysis on the prediction results of different industry labels, and evaluate the correlation between each feature and the target variable based on mutual information; The entropy H(X) of the characteristic factor X is expressed as p(x) is the probability of taking the value x; The entropy H(Y) of the load forecast Y is expressed as p(y) is the probability of taking the value y; The joint entropy of X and Y H(X,Y) is expressed as p(x,y) is the joint probability distribution of random variables X and Y; The mutual information of X and Y, I(X;Y), is expressed as I(X;Y)=H(X)+H(Y)-H(X,Y); The larger the mutual information, the stronger the correlation between X and Y; Through mutual information calculation, the main influencing factors of load forecasting in each industry label are evaluated; Adaptive Learning and Evaluation of Load Forecasting Models: For the substations where the number of samples is less than the set value, the load forecasting models used include the support vector machine model and the kernel support vector machine model; Given training samples D = {(x1,l1),(x2,l2),...,(x m ,l m )},x m is the mth sample, l m is the label corresponding to the mth sample, construct the model f(x)=wφ(x)+b, where f(x) is the label corresponding to the sample, w is the weight coefficient, φ(x) is the feature space, and b is the bias term; Based on the principle of structural risk minimization, the objective function is set as Where R(x) is the objective function value; ||w|| is the weight coefficient; m is the number of samples; y i is the actual label value of the i-th sample; ε is the allowable deviation; Introducing penalty coefficients and slack variables, the objective function is transformed into In the formula is the objective function value; C is the regularization parameter; ξ i is the first slack variable, is the second slack variable; φ(x i ) is the final model function; m is the number of samples; Using Lagrange duality theory to solve, we get In the formula is the first Lagrangian coefficient, a i is the second Lagrangian coefficient; K(x i ,x) is the RBF kernel function; For substations with volatility greater than the set value, the load forecasting models used include the random forest model, the gradient boosting model and the extreme random tree model.

8. The method for fine-grained prediction of power system load according to claim 7 is characterized in that Step S8 is to assess the load forecasting model of each substation obtained in step S7, and use the load forecasting model that finally passes the assessment to perform refined load forecasting for the corresponding substation, which specifically includes the following steps: The following formula is used to calculate the accuracy Acc of the load forecasting model: Where n is the number of samples; Q is the load reference value; x i is the actual load value at the i-th moment; x f,i is the load forecast value at the i-th moment; If the accuracy Acc is higher than or equal to the set value, the load forecasting model passes the assessment; If the accuracy Acc is lower than the set value, the load forecasting model fails the assessment, and the process returns to step S1 to re-perform adaptive learning of the load forecasting model until the load forecasting model passes the assessment; The load forecasting model that finally passes the assessment is used to carry out detailed load forecasting for the corresponding substation.

9. A system for implementing the refined prediction method for power system area load according to any one of claims 1 to 8, characterized in that It includes a data acquisition module, a label calculation module, a label correction module, a data clustering module, a load analysis module, a user analysis module, a model learning module and a load forecasting module; the data acquisition module, the label calculation module, the label correction module, the data clustering module, the load analysis module, the user analysis module, the model learning module and the load forecasting module are connected in series in sequence; the data acquisition module is used to obtain the historical data information of each substation in the target power system, and upload the data information to the label calculation module; the label calculation module is used to calculate the preliminary industry label of each substation according to the received data information and the acquired historical data information, and upload the data information to the label correction module; The label correction module is used to summarize and synthesize the preliminary industry labels calculated according to the received data information, obtain the final industry labels of each area, and upload the data information to the data clustering module; The data clustering module is used to obtain the typical load curve of each substation based on the received data information and the acquired historical data information by using a clustering algorithm, and cluster the typical load curve again to obtain the extended industry label under the final industry label, and upload the data information to the load analysis module; The load analysis module is used to analyze the load characteristic changes of each extended industry tag in different seasons and the factors affecting the load changes based on the received data information, and upload the data information to the user analysis module; The user analysis module is used to generate user portraits of each substation based on the received data information and the obtained analysis results, and to perform correlation analysis, and upload the data information to the model learning module; The model learning module is used to perform adaptive learning and evaluation of the load forecasting model of each substation based on the received data information, and upload the data information to the load forecasting module; The load forecasting module is used to assess the load forecasting models of each substation based on the received data information, and use the load forecasting model that finally passes the assessment to perform refined load forecasting for the corresponding substation.

Images