A method for extracting short-circuit current control rules based on historical calculation results
By using Gaussian mixture models and correlation coefficient screening, combined with power grid operation data, short-circuit current control rules can be quickly extracted, solving the problem of rapid and reliable short-circuit current control after the integration of new energy sources, and ensuring the safety and stability of the new power system.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NARI TECH CO LTD
- Filing Date
- 2022-12-29
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies struggle to quickly and reliably extract short-circuit current control rules for new power systems from massive historical calculation results. This is especially true after large-scale centralized and distributed renewable energy integration, where traditional methods are prone to time-consuming calculations, redundancy, or omissions.
By acquiring and preprocessing historical short-circuit current calculation results, Gaussian mixture model is used for clustering. Combined with generator set and line status, sets Sgen and Sline with high correlation coefficients are selected. Short-circuit current is calculated using typical operating mode data, and generator set and line control rules are extracted.
It enables rapid and reliable extraction of short-circuit current control rules, supports the adjustment of power grid regulation operation mode and preventive control decisions, and ensures the safe and stable operation of the new power system.
Smart Images
Figure CN116244613B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a method for extracting short-circuit current control rules based on historical calculation results, belonging to the field of power system automation technology. Background Technology
[0002] Large-scale centralized renewable energy integration has directly impacted short-circuit current levels, while massive distributed renewable energy integration has altered the power grid's source-load distribution, posing a significant challenge to traditional short-circuit current control strategies. Methods based on physical models for mining short-circuit current control rules face challenges such as the large volume and dispersed distribution of renewable energy unit data, and the small impact of individual renewable energy units, making it difficult to uncover short-circuit current control rules for novel power systems. The vast historical calculation results of short-circuit currents contain a wealth of short-circuit current control rule information, necessitating the integration of physical mechanisms and big data technologies to extract these rules.
[0003] The patent "Short-circuit current prevention and control method and device considering safety and stability limit constraints" (CN202110428472.7) proposes a short-circuit current prevention and control decision method considering safety and stability limit constraints. It calculates the control effect of candidate short-circuit current control measures on the short-circuit current of key circuit breakers, and selects the optimal short-circuit current prevention and control measures based on this. However, when there are a large number of new energy measures among the candidate measures, the calculation time is relatively long.
[0004] The patent "Method for Optimizing Transmission Network Structure for Adjusting Outgoing Line Schemes of 500 kV Substations" (CN202110441474.X) proposes a method for optimizing the transmission network structure for adjusting outgoing line schemes of 500 kV substations. For the initially screened outgoing line adjustment schemes, it is necessary to calculate the short-circuit current level under different outgoing line adjustment schemes. However, it fails to specify whether the short-circuit current level being evaluated is all circuit breakers in the nearby area or whether it relies on expert experience for screening, which poses a risk of calculation redundancy or omission.
[0005] The patent "An intelligent auxiliary decision-making method and system for limiting short-circuit current operation mode" (CN201911022581.8) proposes a decision-making method for limiting short-circuit current control measures based on machine learning. The machine learning prediction results are verified by simulation. When the prediction results are not applicable, conventional auxiliary decision-making methods are used. However, the inapplicability of machine learning predictions can be known in advance through short-circuit current rule mining. If there are no relevant rules in the historical data, the adaptability of the machine learning prediction results is random.
[0006] The information disclosed in this background section is intended only to enhance the understanding of the overall background of the invention and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention
[0007] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method for extracting short-circuit current control rules based on historical calculation results. By using massive historical operating mode data and historical short-circuit current calculation results, potentially effective candidate measures are first screened out. On this basis, physical mechanisms are integrated to further screen out effective short-circuit current control measures, thereby achieving rapid and reliable extraction of short-circuit current control rules. This supports the adjustment of operating modes and preventive control decisions of power grid regulation, and ensures the safe and stable operation of the new power system.
[0008] To achieve the above objectives, the present invention is implemented using the following technical solution:
[0009] This invention discloses a method for extracting short-circuit current control rules based on historical calculation results, comprising the following steps:
[0010] Obtain the preprocessed historical calculation results of short-circuit current, including the historical calculation results of short-circuit current for each bus, the time scale of the historical calculation results of short-circuit current, and the start-up and shutdown status of generator units and the commissioning status of lines throughout the network;
[0011] Based on the preprocessed short-circuit current history calculation results, Gaussian mixture model is used for clustering to obtain Gaussian mixture clustering results;
[0012] Based on the Gaussian mixture clustering results, and using a preset interval threshold, set S is separated. bus set S bus Accurate and effective short-circuit current control rules can be extracted from all busbars in the system;
[0013] According to the set S bus The correlation coefficients between the start-up and shutdown status of generator units, the line commissioning status, and the short-circuit current of the entire network were calculated respectively, and the set of generator units S was obtained by screening. gen and line set S line ;
[0014] Data from typical operating modes with the entire power grid connected and all generating units in operation are collected to calculate set S. bus The short-circuit current of all busbars is denoted as the first short-circuit current.
[0015] Based on typical operating mode data with the entire power grid connected and all generating units in operation, let S... gen The generator sets in the set are shut down and the set S is calculated. bus The short-circuit current of all busbars is denoted as the second short-circuit current.
[0016] Based on typical operating mode data with the entire power grid connected and all generating units in operation, let S... line Line maintenance and calculation of set Sbus The short-circuit current of all busbars is denoted as the third short-circuit current.
[0017] Based on the first short-circuit current and the second short-circuit current, extract the generator set control rules for the short-circuit current;
[0018] Based on the first short-circuit current and the third short-circuit current, extract the line control rules for the short-circuit current.
[0019] Furthermore, the preprocessing of the short-circuit current history calculation results includes the following steps:
[0020] Obtain the historical calculation results of short-circuit current for each busbar and its corresponding time scale;
[0021] Based on the timescale of the historical short-circuit current calculation results, obtain the start-up and shutdown status of generator units and the commissioning status of lines across the entire network corresponding to the timescale;
[0022] The start-up and shutdown status of generator units and the operation status of lines across the entire network are merged into the historical calculation results of bus short-circuit current for the corresponding time period to obtain the preprocessed historical calculation results of bus short-circuit current for each bus. If there are only historical calculation results of bus short-circuit current or data on the start-up and shutdown status of generator units and the operation status of lines across the entire network for a certain time period, the data for that time period is removed.
[0023] Furthermore, in the generator set start-up and shutdown state, generator set start-up is recorded as 1 and shutdown is recorded as 0; in the line commissioning state, line commissioning is recorded as 1 and shutdown is recorded as 0.
[0024] Furthermore, the Gaussian mixture model includes a Gaussian mixture model based on diagonal covariance, a Gaussian mixture model based on spherical covariance, a Gaussian mixture model based on tied covariance, and a Gaussian mixture model based on full covariance.
[0025] The method for selecting the Gaussian mixture model is as follows:
[0026] A Gaussian mixture model based on diagonal covariance was adopted, with cluster numbers ranging from N1 to N2. The historical calculation results of short-circuit current were clustered, and the optimal number of clusters based on the diagonal covariance Gaussian mixture model was evaluated using the Bayesian information criterion.
[0027] A Gaussian mixture model based on spherical covariance was adopted, with cluster numbers ranging from N1 to N2. The historical calculation results of short-circuit current were clustered, and the optimal number of clusters based on the spherical covariance Gaussian mixture model was evaluated using the Bayesian information criterion.
[0028] A Gaussian mixture model based on tied covariance was adopted, with cluster numbers ranging from N1 to N2. The historical calculation results of short-circuit current were clustered, and the optimal number of clusters based on the tied covariance Gaussian mixture model was evaluated using the Bayesian information criterion.
[0029] A Gaussian mixture model based on full covariance was adopted, with cluster numbers ranging from N1 to N2. The historical calculation results of short-circuit current were clustered, and the optimal number of clusters based on the full covariance Gaussian mixture model was evaluated using the Bayesian information criterion.
[0030] By comparing the short-circuit current distribution errors of the diagonal, spherical, tied, and full covariance Gaussian mixture models with the actual distribution errors, the Gaussian mixture model with the smallest evaluation score based on the Bayesian information criterion is selected as the Gaussian mixture model for short-circuit current clustering.
[0031] Furthermore, based on the Gaussian mixture clustering results, and using a preset interval threshold, a set S of bus lines is separated. bus ,include:
[0032] Based on the Gaussian mixture clustering results, the response is that at least one interval exists. If the probability of bus k distributed within this interval is less than a preset interval threshold ε, then bus k is included in set S. bus .
[0033] Furthermore, the expression for the probability distribution is as follows:
[0034]
[0035] Where p(x) is the probability density of short-circuit current x, K is the optimal cluster number, and ω i μ i and ∑ i Let be the weights, mean, and variance of the i-th class, respectively.
[0036] Furthermore, the generator set assembly S gen and line set S line The screening method is as follows:
[0037] According to the set S bus Calculate the correlation coefficient between the start-stop status of generator units and the short-circuit current of the entire network, and select generator units whose absolute value of the correlation coefficient is greater than the preset coefficient threshold η to obtain the generator unit set S. gen ;
[0038] According to the set S bus The correlation coefficient between the line's operational status and short-circuit current is calculated, and lines with an absolute value of the correlation coefficient greater than a preset threshold η are selected to obtain the line set S.line .
[0039] Furthermore, the preset coefficient threshold η is 0.7.
[0040] Furthermore, the expression for the correlation coefficient between the generator set's start-up and shutdown states and the short-circuit current is as follows:
[0041]
[0042] Where, r g,k This represents the correlation coefficient between the start-stop state of generator set g and the short-circuit current of bus k; n represents the number of historical data points. This indicates the start / stop status of generator set g in the j-th record; This represents the average value of generator set g under start-stop conditions; This represents the short-circuit current of the j-th busbar k; This represents the average short-circuit current at bus k.
[0043] Furthermore, the expression for the correlation coefficient between the line's operational status and the short-circuit current is as follows:
[0044]
[0045] Where, r l,k This represents the correlation coefficient between the operational status of line l and the short-circuit current of bus k; n represents the number of historical data points. This indicates the operational status of line l of the j-th line; This represents the average value of the operational status of line l; This represents the short-circuit current of the j-th busbar k; This represents the average short-circuit current at bus k.
[0046] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:
[0047] This invention integrates big data technology and physical mechanisms. By using massive amounts of historical operation data and historical short-circuit current calculation results, it filters out potentially effective candidate measures from a large number of candidate measures. Based on this, it integrates physical mechanisms to further filter out effective short-circuit current control measures, realizing the rapid and reliable extraction of short-circuit current control rules. This supports the adjustment of operation mode and prevention and control decisions of power grid regulation, and ensures the safe and stable operation of the new power system. Attached Figure Description
[0048] Figure 1 This is a flowchart of a short-circuit current control rule extraction method based on historical calculation results; Detailed Implementation
[0049] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.
[0050] Example
[0051] This embodiment discloses a method for extracting short-circuit current control rules based on historical calculation results, including the following steps:
[0052] Obtain the preprocessed historical calculation results of short-circuit current, including the historical calculation results of short-circuit current for each bus, the time scale of the historical calculation results of short-circuit current, and the start-up and shutdown status of generator units and the commissioning status of lines throughout the network;
[0053] Based on the preprocessed short-circuit current history calculation results, Gaussian mixture model is used for clustering to obtain Gaussian mixture clustering results;
[0054] Based on the Gaussian mixture clustering results, and using a preset interval threshold, set S is separated. bus set S bus Accurate and effective short-circuit current control rules can be extracted from all busbars in the system;
[0055] According to set S bus The correlation coefficients between the start-up and shutdown status of generator units, the line commissioning status, and the short-circuit current of the entire network were calculated respectively, and the set of generator units S was obtained by screening. gen and line set S line ;
[0056] Data from typical operating modes with the entire power grid connected and all generating units in operation are collected to calculate set S. bus The short-circuit current of all busbars is denoted as the first short-circuit current.
[0057] Based on typical operating mode data with the entire power grid connected and all generating units in operation, let S... gen The generator sets in the set are shut down and the set S is calculated. bus The short-circuit current of all busbars is denoted as the second short-circuit current.
[0058] Based on typical operating mode data with the entire power grid connected and all generating units in operation, let S... line Line maintenance and calculation of set S bus The short-circuit current of all busbars is denoted as the third short-circuit current.
[0059] Based on the first short-circuit current and the second short-circuit current, extract the generator set control rules for the short-circuit current;
[0060] Based on the first and third short-circuit currents, extract the line control rules for the short-circuit current.
[0061] The technical concept of this invention is as follows: By integrating big data technology and physical mechanisms, and using massive historical operation data and historical short-circuit current calculation results, potentially effective candidate measures are screened from a large number of candidate measures. On this basis, physical mechanisms are integrated to further screen effective short-circuit current control measures, thereby achieving rapid and reliable extraction of short-circuit current control rules, supporting the adjustment of operation mode and prevention and control decisions of power grid regulation, and ensuring the safe and stable operation of the new power system.
[0062] The specific steps are as follows:
[0063] Step 1.
[0064] Obtain the timescale of the historical calculation results of short-circuit current and the historical calculation results of short-circuit current for each bus.
[0065] Step two.
[0066] The start-up and shutdown status of generator units and the commissioning status of lines in the whole network are obtained from the historical data of power grid operation mode and the time scale of all historical calculation results of short-circuit current. Among them, generator unit start-up is recorded as 1, shutdown is recorded as 0, line commissioning is recorded as 1, and shutdown is recorded as 0.
[0067] Step 3.
[0068] The historical calculation results of bus short-circuit current with the same time scale are merged with the start-up and shutdown status of generator units and the commissioning status of lines in the whole network into a unified record, and the merged historical calculation results of bus short-circuit current are obtained.
[0069] If a certain time point only contains historical calculation results of bus short-circuit current or data on the start-up and shutdown status of generator units and the commissioning status of lines across the entire network, then the data for that time point will be removed.
[0070] Step four.
[0071] The historical short-circuit currents of each busbar after merging were clustered using a Gaussian mixture model, yielding the Gaussian mixture clustering results. Let I be the historical short-circuit current data for busbar k. k The historical maximum value of the short-circuit current of bus k is The historical minimum value is
[0072] Gaussian mixture models include Gaussian mixture models based on diagonal covariance, Gaussian mixture models based on spherical covariance, Gaussian mixture models based on tied covariance, and Gaussian mixture models based on full covariance.
[0073] The selection method for Gaussian mixture models is as follows:
[0074] A Gaussian mixture model based on diagonal covariance was adopted, with cluster numbers ranging from N1 to N2. The historical calculation results of short-circuit current were clustered, and the optimal number of clusters based on the diagonal covariance Gaussian mixture model was evaluated using the Bayesian information criterion.
[0075] A Gaussian mixture model based on spherical covariance was adopted, with cluster numbers ranging from N1 to N2. The historical calculation results of short-circuit current were clustered, and the optimal number of clusters based on the spherical covariance Gaussian mixture model was evaluated using the Bayesian information criterion.
[0076] A Gaussian mixture model based on tied covariance was adopted, with cluster numbers ranging from N1 to N2. The historical calculation results of short-circuit current were clustered, and the optimal number of clusters based on the tied covariance Gaussian mixture model was evaluated using the Bayesian information criterion.
[0077] A Gaussian mixture model based on full covariance was adopted, with cluster numbers ranging from N1 to N2. The historical calculation results of short-circuit current were clustered, and the optimal number of clusters based on the full covariance Gaussian mixture model was evaluated using the Bayesian information criterion.
[0078] By comparing the short-circuit current distribution errors of the diagonal, spherical, tied, and full covariance Gaussian mixture models with the actual distribution errors, the Gaussian mixture model with the smallest Bayesian information criterion score is selected as the Gaussian mixture model for short-circuit current clustering.
[0079] Step 5.
[0080] Based on the Gaussian mixture clustering results, if at least one interval exists... If the probability of bus k being distributed within this interval is less than a preset interval threshold ε, then bus k is included in set S. bus set S bus Accurate and effective short-circuit current control rules can be extracted from all busbars in the system;
[0081] The interval The identification method is as follows:
[0082] 5-1) Order
[0083] 5-2) Obtaining short-circuit current based on Gaussian mixture model The probability distribution is given by the following formula:
[0084]
[0085] In the above formula, p(x) is the probability density of short-circuit current x, K is the optimal cluster number, and ω iμ i and ∑ i Let be the weights, mean, and variance of the i-th class, respectively.
[0086] 5-3) If If there exists at least one short-circuit current with a probability distribution less than ε, then there must exist... The probability that the busbar k is distributed in this interval is less than ε.
[0087] Step Six.
[0088] For set S bus Based on the historical calculation results of the bus short-circuit current, the correlation coefficients between the start-up and shutdown status of generator units and the commissioning status of lines in the entire network and the short-circuit current are calculated respectively. Generator units and lines with an absolute value of correlation coefficient greater than η are selected and denoted as sets S. gen and S line In this embodiment, the preset coefficient threshold η is 0.7.
[0089] The expression for the correlation coefficient between the generator set's start-up and shutdown status and the short-circuit current is as follows:
[0090]
[0091] Where, r g,k This represents the correlation coefficient between the start-stop state of generator set g and the short-circuit current of bus k; n represents the number of historical data points. This indicates the start / stop status of generator set g in the j-th record; This represents the average value of generator set g under start-stop conditions; This represents the short-circuit current of the j-th busbar k; This represents the average short-circuit current at bus k.
[0092] The expression for the correlation coefficient between the line's operational status and short-circuit current is as follows:
[0093]
[0094] Where, r l,k This represents the correlation coefficient between the operational status of line l and the short-circuit current of bus k; n represents the number of historical data points. This indicates the operational status of line l of the j-th line; This represents the average value of the operational status of line l; This represents the short-circuit current of the j-th busbar k; This represents the average short-circuit current at bus k.
[0095] Step 7.
[0096] Data from typical operating modes with the entire power grid connected and all generating units in operation are collected to calculate set S. busThe short-circuit current of all busbars is denoted as the first short-circuit current I. 0 .
[0097] Step 8.
[0098] Based on typical operating mode data with the entire power grid connected and all generating units in operation, let S... gen The generator sets in the set are shut down and the set S is calculated. bus The short-circuit current of all busbars is denoted as the second short-circuit current I. g .
[0099] Step Nine.
[0100] Based on typical operating mode data with the entire power grid connected and all generating units in operation, let S... line Line maintenance and calculation of set S bus The short-circuit current of all busbars is denoted as the third short-circuit current I. l .
[0101] Step 10.
[0102] If generator set g is shut down, the following conditions are met: in, This represents the first short-circuit current of bus k; The second short-circuit current of bus k is represented by α; α represents a coefficient. Indicates the value of the second interval; Let represent the first interval value; then we get the rule: when generator set g is out of service, the short-circuit current of bus k decreases significantly; when generator set g is in service, the short-circuit current of bus k increases significantly.
[0103] Step 11.
[0104] If the short-circuit current of bus k is after line l is shut down satisfy in, This represents the first short-circuit current of bus k; The third short-circuit current of busbar k is represented; α represents a coefficient. Indicates the value of the second interval; Let represent the first interval value; then the rule is: when line l is out of service, the short-circuit current of bus k decreases significantly; when line l is in service, the short-circuit current of bus k increases significantly.
[0105] In summary, this invention integrates historical data on power grid operation modes and historical short-circuit current calculations using time-scaled methods. Based on the Bayesian information criterion, it selects the optimal covariance type and optimal number of clusters for the Gaussian mixture clustering model of short-circuit current. Based on the probability distribution of short-circuit current, it determines that accurate short-circuit current control rules can be extracted from historical data. Combined with typical operating mode data of the entire power grid, reliable short-circuit current control rules are selected to support short-circuit current control during power grid regulation. This invention can control the short-circuit current level of new power systems, improving the efficiency and reliability of decision-making processes such as mode adjustment and preventative control that consider short-circuit current.
[0106] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.
Claims
1. A method for extracting short-circuit current control rules based on historical calculation results, characterized in that, Includes the following steps, Obtain the preprocessed historical calculation results of short-circuit current, including the historical calculation results of short-circuit current for each bus, the time scale of the historical calculation results of short-circuit current, and the start-up and shutdown status of generator units and the commissioning status of lines throughout the network; Based on the preprocessed short-circuit current history calculation results, Gaussian mixture model is used for clustering to obtain Gaussian mixture clustering results; Based on the Gaussian mixture clustering results, and using a preset interval threshold, sets are separated. ,gather Accurate and effective short-circuit current control rules can be extracted from all busbars in the system; According to the set The correlation coefficients between the start-up and shutdown status of generator units, the line commissioning status, and the short-circuit current of the entire network were calculated respectively, and the set of generator units was obtained by screening. and route set ; Data from typical operating modes with the entire power grid connected and all generating units in operation are collected to calculate the set. The short-circuit current of all busbars is denoted as the first short-circuit current. Based on typical operating mode data with the entire power grid connected and all generating units in operation, respectively, let The generator set is shut down and the set is calculated. The short-circuit current of all busbars is denoted as the second short-circuit current. Based on typical operating mode data with the entire power grid connected and all generating units in operation, respectively, let Line maintenance and calculation set The short-circuit current of all busbars is denoted as the third short-circuit current. Based on the first short-circuit current and the second short-circuit current, extract the generator set control rules for the short-circuit current; Based on the first short-circuit current and the third short-circuit current, extract the line control rules for the short-circuit current.
2. The method for extracting short-circuit current control rules based on historical calculation results according to claim 1, characterized in that, The preprocessing of the short-circuit current history calculation results includes the following steps: Obtain the historical calculation results of short-circuit current for each busbar and its corresponding time scale; Based on the timescale of the historical short-circuit current calculation results, obtain the start-up and shutdown status of generator units and the commissioning status of lines across the entire network corresponding to the timescale; The start-up and shutdown status of generator units and the operation status of lines across the entire network are merged into the historical calculation results of bus short-circuit current for the corresponding time period to obtain the preprocessed historical calculation results of bus short-circuit current for each bus. If there are only historical calculation results of bus short-circuit current or data on the start-up and shutdown status of generator units and the operation status of lines across the entire network for a certain time period, the data for that time period is removed.
3. The method for extracting short-circuit current control rules based on historical calculation results according to claim 2, characterized in that, In the generator set start-up and shutdown state, generator set start-up is recorded as 1 and shutdown is recorded as 0; in the line commissioning state, line commissioning is recorded as 1 and shutdown is recorded as 0.
4. The method for extracting short-circuit current control rules based on historical calculation results according to claim 1, characterized in that, The Gaussian mixture models include Gaussian mixture models based on diagonal covariance, Gaussian mixture models based on spherical covariance, Gaussian mixture models based on tied covariance, and Gaussian mixture models based on full covariance. The method for selecting the Gaussian mixture model is as follows: A Gaussian mixture model based on diagonal covariance was adopted, with the number of clusters being respectively ~ The historical calculation results of short-circuit current are clustered, and the optimal number of clusters based on the diagonal covariance Gaussian mixture model is evaluated using the Bayesian information criterion. A Gaussian mixture model based on spherical covariance was adopted, with the number of clusters being respectively ~ The historical calculation results of short-circuit current are clustered, and the optimal number of clusters based on the spherical covariance Gaussian mixture model is evaluated using the Bayesian information criterion. A Gaussian mixture model based on tied covariance was adopted, with the number of clusters being respectively ~ The historical calculation results of short-circuit current are clustered, and the optimal number of clusters based on the Tied covariance Gaussian mixture model is evaluated using the Bayesian information criterion. A Gaussian mixture model based on full covariance was adopted, with the number of clusters being respectively ~ The historical calculation results of short-circuit current are clustered, and the optimal number of clusters based on the full covariance Gaussian mixture model is evaluated using the Bayesian information criterion. By comparing the short-circuit current distribution errors of the diagonal, spherical, tied, and full covariance Gaussian mixture models with the actual distribution errors, the Gaussian mixture model with the smallest evaluation score of the Bayesian information criterion is selected as the Gaussian mixture model for short-circuit current clustering.
5. The method for extracting short-circuit current control rules based on historical calculation results according to claim 1, characterized in that, Based on the Gaussian mixture clustering results, and using a preset interval threshold, a set of busbars is separated. ,include: Based on the Gaussian mixture clustering results, the response is that at least one interval exists. busbar The probability of the distribution in this interval is less than the preset interval threshold. Then the busbar Include in set .
6. The method for extracting short-circuit current control rules based on historical calculation results according to claim 5, characterized in that, The expression for the probability distribution is as follows: ； in, The short-circuit current is The probability density, To achieve the optimal number of clusters, , and The first The class's weight, mean, and variance.
7. The method for extracting short-circuit current control rules based on historical calculation results according to claim 1, characterized in that, The generator set collection and route set The screening method is as follows: According to the set Calculate the correlation coefficient between the start-up and shutdown status of generator units across the entire network and the short-circuit current, and select generators whose absolute correlation coefficient value is greater than a preset threshold. The generator sets are obtained. ; According to the set Calculate the correlation coefficient between the line's operational status and the short-circuit current, and select lines with an absolute correlation coefficient value greater than a preset threshold. The routes are obtained, resulting in a set of routes. .
8. The method for extracting short-circuit current control rules based on historical calculation results according to claim 7, characterized in that, The preset coefficient threshold It is 0.
7.
9. The method for extracting short-circuit current control rules based on historical calculation results according to claim 7, characterized in that, The expression for the correlation coefficient between the generator set start-up and shutdown status and the short-circuit current is as follows: ； in, Indicates generator set Start-stop status and bus The correlation coefficient of short-circuit current; Indicates the quantity of historical data; Indicates the first Record generator set Start-stop status; Indicates generator set The average value of start-stop states; Indicates the first busbar Short-circuit current; Indicates busbar The average value of the short-circuit current.
10. The method for extracting short-circuit current control rules based on historical calculation results according to claim 7, characterized in that, The expression for the correlation coefficient between the line's operational status and the short-circuit current is as follows: ； in, Indicates the line Commissioning status and busbar The correlation coefficient of short-circuit current; Indicates the quantity of historical data; Indicates the first Line Operational status; Indicates the line The average value of the operational status; Indicates the first busbar Short-circuit current; Indicates busbar The average value of the short-circuit current.