Power distribution network typical operation mode acquisition method considering voltage characteristics

Abstract

The invention discloses a power distribution network typical operation mode acquisition method considering voltage characteristics, and belongs to the technical field of power grid operation mode arrangement, and the method comprises the steps: S1, building a power grid model based on a power grid architecture in a power distribution network; s2, obtaining an estimated value of a state variable by using a state estimation algorithm based on historical operation data of the power distribution network and a power grid model, and obtaining a key element, a first load rate of the key element, a key line and a first voltage of the key line based on the estimated value of the state variable; s3, taking the normalized first load rate and the first voltage as clustering analysis features to obtain a feature matrix; and S4, taking the feature matrix as the input of a clustering algorithm to obtain a power distribution network process operation mode, and determining the power distribution network process operation mode meeting a preset condition as a typical operation mode of the power distribution network. The technical problem that the acquisition accuracy is difficult to improve while the acquisition efficiency of the typical operation mode of the power distribution network is guaranteed is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid operation mode arrangement, and specifically to a method for obtaining typical operation modes of a distribution network considering voltage characteristics. Background Art

[0002] With the gradual development of a new power system, corresponding dispatching for different operation scenarios is an important way to improve the stability and security of the power grid system. For example, a method, system, device, and medium for operating and dispatching a power system with the patent number CN114219216A, wherein the method includes: obtaining operation mode data of the power system; classifying the operation mode data by using a trained decision tree classifier to obtain typical operation scenario types; obtaining an optimized dispatching model corresponding to the typical operation scenario types, and using the optimized dispatching model to control the operation and dispatching of the power system; wherein, the typical operation scenario types are obtained through clustering analysis, and the optimized dispatching model includes a deep reinforcement learning model and a new power system model. By clustering, multiple typical operation scenarios are divided, and the corresponding optimized dispatching model is retrieved for different typical operation scenarios to control the dispatching, that is, a deep reinforcement learning optimized dispatching scheme is designed for different typical scenarios, effectively improving the performance of the dispatching decision-making scheme. The above scheme obtains typical operation scenarios by clustering the operation mode data. However, during the clustering analysis process, the errors existing in obtaining the operation mode data are not considered, and the local optimum situation existing in the clustering analysis is not considered. Therefore, the obtained typical operation scenarios are not accurate. At the same time, all operation mode data are processed to obtain typical operation scenarios. Therefore, the efficiency of obtaining typical operation scenarios is low. Summary of the Invention

[0003] Aiming at the technical problem that it is difficult to improve the accuracy of the typical operation mode of the distribution network while ensuring the acquisition efficiency of the typical operation mode of the distribution network in the prior art, the present invention provides a method for obtaining the typical operation mode of the distribution network considering voltage characteristics. By the characteristics of the state estimation algorithm, the estimated values of the state variables are obtained, eliminating the errors existing in obtaining historical operation data. According to the estimated values, key components, the first load rates of the key components, key lines, and the first voltages of the key lines are obtained. The normalized first load rates and first voltages are used as clustering analysis features to obtain a feature matrix. The feature matrix is used as the input of the clustering algorithm. By obtaining key components, key lines, and corresponding electrical signals through the estimated values for clustering analysis, the clustering analysis efficiency is improved. By determining the process operation mode of the distribution network that meets the preset conditions as the typical operation mode of the distribution network, the existing local optimum situation is avoided. The technical problem of being difficult to improve the accuracy of the typical operation mode of the distribution network while ensuring the acquisition efficiency of the typical operation mode of the distribution network is solved.

[0004] To solve the above technical problems, the present invention provides a method for obtaining typical operating modes of a distribution network considering voltage characteristics, including the following steps: S1: Establish a power grid model based on the power grid architecture in the distribution network; S2: Based on the historical operation data of the distribution network and the power grid model, use a state estimation algorithm to obtain the estimated values of state variables, and obtain key components, the first load rates of key components, key lines, and the first voltages of key lines based on the estimated values of state variables; S3: Use the normalized first load rates and first voltages as clustering analysis features to obtain a feature matrix; S4: Use the feature matrix as the input of a clustering algorithm to obtain the process operating modes of the distribution network, and determine the process operating modes of the distribution network that meet the preset conditions as the typical operating modes of the distribution network.

[0005] After adopting the above technical solutions, the present invention has the following advantages: Considering that during the process of obtaining historical operation data, due to the influence of internal and external factors, there will be differences between the obtained historical operation data and the actual historical operation data. The state estimation algorithm can more accurately estimate the state of the system by combining the mathematical model of the system and historical operation data, thereby reducing the differences from the actual historical operation data. Therefore, the errors existing in obtaining historical operation data are eliminated through the state estimation algorithm, and key components and key lines that have the greatest impact on the power grid operation are identified through accurate estimated values, improving the efficiency of subsequent clustering analysis. By determining the process operating modes of the distribution network that meet the preset conditions as the typical operating modes of the distribution network, the existing local optimum situations are avoided, and the accuracy of the obtained typical operating modes of the distribution network is improved; Through clustering analysis using the feature matrix constructed by load rates and voltages, the actual operating state of the distribution network can be more accurately reflected. Scheduling the distribution network according to the actual operating conditions can reduce the risk of faults in the distribution network, thereby improving power supply reliability and safety; It solves the technical problem of being difficult to improve the accuracy of the typical operating modes of the distribution network while ensuring the acquisition efficiency of the typical operating modes of the distribution network.

[0006] Preferably, the S1 includes: S11: Construct an initial power grid model including nodes and lines according to the power grid architecture; S12: According to the historical operation data, set electrical parameters for electrical equipment at nodes in the initial power grid model and set line parameters for lines in the initial power grid model; S13: Update the initial power grid model based on the electrical parameters and line parameters, obtain the measured electrical signals of the electrical equipment and lines based on the initial power grid model, obtain the difference between the measured electrical signals and the historical electrical signals of the electrical equipment and lines in the historical operation data. If the difference meets the preset requirements, the initial power grid model is the power grid model; otherwise, adjust the electrical parameters or line parameters according to the difference, and execute S13.

[0007] In this solution, by constructing an initial power grid model including nodes and lines, and setting electrical parameters and line parameters for the nodes and lines in the initial power grid model according to historical operation data, it can intuitively reflect the actual operation of the distribution network, thereby improving the accuracy of the initial power grid model and providing a basis for the state estimation algorithm. By comparing the difference between the measured electrical signals and the historical electrical signals of the electrical equipment and lines in the historical operation data, it can be checked whether the initial power grid model conforms to the actual operation situation. When the preset requirements are met, the initial power grid model is the power grid model; otherwise, the parameters need to be adjusted according to the difference, further improving the accuracy of the power grid model.

[0008] Preferably, the S2 includes: S21: Set the initial estimated value of the state variable according to the average value of the historical operation data; S22: Obtain the mutual relationship between the actual measured value and the initial estimated value of the state variable according to the initial estimated value and the power grid model, and then obtain the partial derivative of the state variable with respect to the mutual relationship; S23: Obtain the weight matrix based on the reliability of the actual measured value, and obtain the information matrix based on the partial derivative and the weight matrix; S24: Obtain the correction amount of the state variable according to the information matrix and the difference between the actual measured value and the initial estimated value; S25: Obtain the estimated value of the state variable based on the correction amount, and obtain the key components, the first load rate of the key components, the key lines and the first voltage of the key lines based on the estimated value of the state variable.

[0009] In this solution, by setting the initial estimated value of the state variable according to the average value of historical operation data, the rationality of the initial estimated value is ensured, and the subsequent calculation using an unreasonable initial estimated value is avoided. While saving computing resources, it also improves the acquisition efficiency of the typical operation mode of the distribution network to a certain extent. By obtaining the correlation between the actual measured value and the initial estimated value of the state variable and calculating the partial derivative, the correlation degree between the actual measured value and the estimated value can be clarified, providing a basis for subsequent state estimation. By obtaining the weight matrix based on the reliability of the actual measured value and obtaining the information matrix based on the partial derivative and the weight matrix, the reliability differences of different measured values can be fully considered. By calculating the correction amount of the state variable and updating the estimated value, the true state can be gradually approximated, improving the accuracy of state estimation. By identifying the key components and key lines that most affect the operation of the power grid with accurate estimated values, the efficiency of subsequent clustering analysis is improved, and thus the acquisition efficiency of the typical operation mode of the distribution network is improved.

[0010] Preferably, in S25, obtaining the estimated value of the state variable based on the correction amount includes: Comparing the absolute value of the correction amount with a preset absolute value. If the absolute value of the correction amount is less than the preset absolute value, the estimated value of the state variable is obtained according to the correction amount; otherwise, the initial estimated value is updated based on the correction amount, and S22 is executed.

[0011] In this solution, by comparing the absolute value of the correction amount with the preset absolute value, when the absolute value of the correction amount is small, it indicates that the initial estimated value is close to the actual value. Therefore, the estimated value of the state variable is obtained according to the correction amount. When the absolute value of the correction amount is large, it indicates that there is a large deviation between the initial estimated value and the actual value. Therefore, the initial estimated value is updated based on the correction amount, and S22 is re-executed. By means of iteration, the actual value is gradually approximated, improving the accuracy of the estimated value of the state variable.

[0012] Preferably, in S25, obtaining the key components, the first load rate of the key components, the key lines, and the first voltage of the key lines based on the estimated value of the state variable includes: Obtaining the second load rate of the electrical equipment based on the first estimated value in the estimated value, and obtaining the second voltage of the line based on the second estimated value in the estimated value; Taking the electrical equipment with a second load rate greater than the preset load rate as the key component and obtaining the first load rate of the key component. Comparing the second voltage with the rated voltage, and taking the line for which the comparison fails as the key line and obtaining the first voltage of the key line.

[0013] Preferably, S3 includes: S31: Normalize the first voltage based on the highest voltage and the lowest voltage in the historical operation data, and normalize the first load rate based on the highest load rate and the lowest load rate in the historical operation data; S32: Obtain a feature matrix based on the critical line name and the clustering analysis features.

[0014] Preferably, the S4 includes: S41: Randomly select several data from the feature matrix as the initial clustering centers; S42: Perform sample point assignment based on the initial clustering centers, thereby updating the initial clustering centers and obtaining the number of updates; S43: If the initial clustering centers meet the convergence condition or the number of updates is greater than the preset number of updates, then execute S44; otherwise, execute S42. S44: Obtain the operation mode of the distribution network process based on the initial clustering centers, set preset conditions based on the full coverage principle, and determine the operation mode of the distribution network process that meets the preset conditions as the typical operation mode of the distribution network.

[0015] Preferably, the S4 further includes: If there is no operation mode of the distribution network process that meets the preset conditions in the operation mode of the distribution network process, then clean the historical operation data, process outliers and missing values, and execute S2.

[0016] Beneficial effects of this solution: Considering that in the process of obtaining historical operation data, due to the influence of internal and external factors, the obtained historical operation data may be different from the actual historical operation data. The state estimation algorithm can more accurately estimate the state of the system by combining the mathematical model of the system and the historical operation data, thereby reducing the difference from the actual historical operation data. Therefore, the error existing in the acquisition of historical operation data is eliminated through the state estimation algorithm. The key components and key lines that most affect the operation of the power grid are identified through accurate estimated values, improving the efficiency of subsequent clustering analysis. By determining the operation mode of the distribution network process that meets the preset conditions as the typical operation mode of the distribution network, the existing local optimal situation is avoided, and the accuracy of the obtained typical operation mode of the distribution network is improved; Through clustering analysis using the feature matrix constructed by the load rate and voltage, the actual operation state of the distribution network can be more accurately reflected. Scheduling the distribution network according to the actual operation situation can reduce the risk of the distribution network failing, thereby improving the power supply reliability and security; By comparing the absolute value of the correction amount with a preset absolute value, when the absolute value of the correction amount is small, it indicates that the initial estimate value is close to the actual value. Therefore, the estimated value of the state variable is obtained according to the correction amount. When the absolute value of the correction amount is large, it indicates that there is a large deviation between the initial estimate value and the actual value. Therefore, the initial estimate value is updated based on the correction amount, and S22 is executed again. By means of iteration, the actual value is gradually approximated, the accuracy of the estimated value of the state variable is improved, and further the accuracy of the obtained typical distribution network operation mode is improved; It solves the technical problem that it is difficult to improve the accuracy of the typical distribution network operation mode while ensuring the acquisition efficiency of the typical distribution network operation mode.

[0017] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the method for obtaining the typical distribution network operation mode considering voltage characteristics are implemented.

[0018] The present invention also provides a storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are loaded and executed by a processor, the steps of the method for obtaining the typical distribution network operation mode considering voltage characteristics are implemented. Description of the Drawings

[0019] By reading the detailed description of the non-restrictive embodiments with reference to the following drawings, other features, purposes, and advantages of the present invention will become more obvious. The drawings are only for the purpose of showing the preferred embodiments and are not considered as limiting the present invention. Moreover, throughout the drawings, the same reference numerals are used to represent the same components.

[0020] Figure 1 It is a flowchart of the method for obtaining the typical distribution network operation mode considering voltage characteristics of the present invention. Detailed Embodiments

[0021] To make the purpose, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described here are only the best embodiments of the present invention, only used to explain the present invention, and do not limit the protection scope of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present invention.

[0022] Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts depict the operations (or steps) as sequential processes, many of the operations (or steps) can be implemented in parallel, concurrently, or simultaneously. In addition, the order of the operations can be rearranged. The process can be terminated when its operations are completed, but it can also have additional steps not included in the drawings; the process can correspond to a method, function, procedure, subroutine, subprogram, and so on.

[0023] Embodiment 1: As Figure 1 shown, a method for obtaining a typical operation mode of a distribution network considering voltage characteristics includes the following steps: S1: Establish a power grid model based on the power grid architecture in the distribution network.

[0024] The S1 includes: S11: Construct an initial power grid model including nodes and lines according to the power grid architecture; S12: According to the historical operation data, set electrical parameters for the electrical equipment of the nodes in the initial power grid model and set line parameters for the lines in the initial power grid model; S13: Update the initial power grid model based on the electrical parameters and line parameters, obtain the measured electrical signals of the electrical equipment and lines based on the initial power grid model, obtain the difference between the measured electrical signals and the historical electrical signals of the electrical equipment and lines in the historical operation data, if the difference meets the preset requirements, the initial power grid model is the power grid model, otherwise adjust the electrical parameters or line parameters according to the difference, and execute S13.

[0025] In this embodiment, the initial power grid model constructed according to the actual power grid architecture includes all nodes (power plants, substations, transformers, etc.), lines (transmission lines, distribution lines, etc.) and the connection modes between the lines. The historical operation data is obtained from the SCADA system, and the acquisition unit of the historical operation data is years. The historical operation data is exported from the distribution network SCADA system, that is, the cime and DT data from January 1, 2023 to December 31, 2023, including 365 days of cime data and 8,760 points of DT data, a total of 8,760 mode samples. The electrical parameters and line parameters of the initial power grid model are set through the historical electrical parameters of electrical equipment and historical line parameters in the historical operation data, so that the constructed initial power grid model can more accurately reflect the operation state of the actual power grid, providing a basis for the state estimation algorithm, facilitating the state estimation algorithm to obtain accurate estimation values, and then ensuring that key components and key lines that have a greater impact on the actual operation of the power grid can be obtained, improving the accuracy of the typical influence modes of the distribution network. By simulating the actual operation of the power grid through the initial power grid model, the measured electrical signals are obtained, and the measured electrical signals are compared with the historical electrical signals. The electrical signals at least include voltage, current, phase, frequency and amplitude, which can verify whether the initial power grid model conforms to the actual operation situation. When the preset requirements are met, the initial power grid model is the power grid model; otherwise, the parameters need to be adjusted according to the difference, further improving the accuracy of the power grid model.

[0026] The preset requirements are flexibly set according to user needs. If it is necessary to maximize the accuracy, when the difference is less than the preset value, the preset requirements are met. At this time, the preset value is set relatively small, thus meeting the requirement of maximizing the accuracy. If both accuracy and efficiency are required, the preset value can be set according to user experience. At this time, the preset value is relatively larger than the above preset value. By flexibly setting the preset requirements, the flexibility of obtaining the typical operation modes of the distribution network is improved. The electrical parameters or line parameters are adjusted according to the difference and experience. For example, if the measured electrical signal of an electrical equipment at a certain node, that is, the simulated voltage, is significantly lower than the historical electrical signal, and the electrical equipment at this node is connected to a long-distance transmission line. According to experience, the reactance parameter of the long-distance transmission line has a significant impact on the voltage drop. Therefore, the reactance parameter of this transmission line can be considered to be increased. At the same time, the voltage setting of the generator or transformer connected to the electrical equipment at this node can also be checked. When the voltage setting is too low, the voltage is adjusted. When the preset requirements are not met, the electrical parameters or line parameters are adjusted, and S13 is continued to be executed, so that the power grid model can accurately reflect the actual operation of the power grid, and then the accuracy and adaptability of the obtained power grid model are improved.

[0027] S2: Based on the historical operation data of the distribution network and the power grid model, use the state estimation algorithm to obtain the estimated values of the state variables, and based on the estimated values of the state variables, obtain the key components, the first load rates of the key components, the key lines, and the first voltages of the key lines.

[0028] The S2 includes: S21: Set the initial estimated values of the state variables according to the average value of the historical operation data; S22: Obtain the correlation between the actual measured values and the initial estimated values of the state variables according to the initial estimated values and the power grid model, and then obtain the partial derivatives of the state variables with respect to the correlation; S23: Obtain the weight matrix based on the reliability of the actual measured values, and obtain the information matrix based on the partial derivatives and the weight matrix; S24: Obtain the correction amount of the state variables according to the information matrix and the difference between the actual measured values and the initial estimated values; S25: Obtain the estimated values of the state variables based on the correction amount, and based on the estimated values of the state variables, obtain the key components, the first load rates of the key components, the key lines, and the first voltages of the key lines.

[0029] In S25, the obtaining the estimated values of the state variables based on the correction amount includes: Compare the absolute value of the correction amount with a preset absolute value. If the absolute value of the correction amount is less than the preset absolute value, obtain the estimated values of the state variables according to the correction amount. Otherwise, update the initial estimated values based on the correction amount and execute S22.

[0030] In S25, the obtaining the key components, the first load rates of the key components, the key lines, and the first voltages of the key lines based on the estimated values of the state variables includes: Obtain the second load rates of the electrical equipment based on the first estimated values in the estimated values, and obtain the second voltages of the lines based on the second estimated values in the estimated values; Take the electrical equipment with the second load rate greater than the preset load rate as the key components, and obtain the first load rates of the key components. Compare the second voltages with the rated voltages, and take the lines that fail to be compared as the key lines, and obtain the first voltages of the key lines.

[0031] In this embodiment, the state variables include lines and electrical equipment. The initial estimated values are the voltage estimated value of the line and the load rate estimated value of the electrical equipment. The actual measured values are the actual voltage value of the line and the actual load rate value of the electrical equipment. At the same time, when obtaining the load rate estimated value, the parameters for calculating the load rate estimated value change according to the type of electrical equipment to meet the requirements of different electrical equipment. The initial estimated value of the state variable is set according to the average value of the historical operation data, ensuring the rationality of the initial estimated value and avoiding subsequent calculations using unreasonable initial estimated values. While saving computing resources, it also improves the acquisition efficiency of the typical operation mode of the distribution network to a certain extent. The specific method for obtaining the weight matrix based on the reliability of the actual measured value is as follows: The user evaluates the reliability, i.e., accuracy, of the obtained actual measured value according to actual experience. After evaluation, the higher the accuracy of the actual measured value, the greater its weight relative to the actual measured value with lower accuracy. By obtaining the weight matrix, the calculation formula for the information matrix based on the partial derivative and the weight matrix is G = W T HW, where G is the information matrix, W is the weight matrix, and H represents the product of the inverse of the Jacobian matrix composed of partial derivatives and the Jacobian matrix. The calculation formula for the correction amount is Δx = G -1 W T r, where Δx represents the correction amount and r represents the difference between the actual measured value and the initial estimated value. By obtaining the mutual relationship between the actual measured value and the initial estimated value of the state variable and calculating the partial derivative, the correlation degree between the actual measured value and the estimated value can be clarified, providing a basis for subsequent state estimation. By obtaining the weight matrix based on the reliability of the actual measured value and obtaining the information matrix based on the partial derivative and the weight matrix, the reliability differences of different measured values can be fully considered. According to the information matrix and the difference, the correction amount is obtained, improving the accuracy of the correction amount and thus the accuracy of the obtained estimated value. By accurately identifying the key components and key lines that most affect the operation of the power grid through the estimated value, the efficiency of subsequent clustering analysis is improved, and thus the acquisition efficiency of the typical operation mode of the distribution network is improved.

[0032] By comparing the absolute value of the correction amount with a preset absolute value, when the absolute value of the correction amount is small, it indicates that the initial estimate value is close to the actual value. Therefore, the estimated value of the state variable is obtained based on the correction amount. Specifically, the estimated value of the state variable obtained based on the correction amount is: the sum of the initial estimate value and the correction amount is the estimated value. When the absolute value of the correction amount is large, it indicates that there is a large deviation between the initial estimate value and the actual value. Therefore, the initial estimate value is updated based on the correction amount, and S22 is executed again. The initial estimate value updated based on the correction amount is specifically: the sum of the initial estimate value and the correction amount is used as the initial estimate value. The preset absolute value is flexibly set according to user requirements, and the actual value is gradually approximated through an iterative method, improving the accuracy of the estimated value of the state variable. The first estimated value in the estimated value is the load rate, and the second estimated value is the voltage. By obtaining the load rates of all electrical devices and the voltages of all lines, and then identifying key components and key lines based on the load rate and voltage, the typical operation method of the distribution network obtained can more truly reflect the state of the distribution network. By obtaining and identifying key components and key lines, the operation modes with significant changes in voltage and load rate caused by source-load fluctuations, power grid power flow, and voltage variability are specifically covered. Then, it is convenient to adjust the power grid according to the operation mode, achieve load balancing, reduce losses, and thus improve the regulation efficiency of the power system.

[0033] S3: Using the normalized first load rate and the first voltage as clustering analysis features to obtain a feature matrix

[0034] S3 includes: S31: Normalize the first voltage based on the highest voltage and the lowest voltage in the historical operation data, and normalize the first load rate based on the highest load rate and the lowest load rate in the historical operation data; S32: Obtain the feature matrix based on the key line name and the clustering analysis features.

[0035] In this embodiment, the calculation formula for normalizing the first voltage based on the highest voltage and the lowest voltage in the historical operation data is In the formula, X is the normalized first voltage, X i is the unnormalized first voltage, X max is the highest voltage, X min is the lowest voltage. The calculation formula for normalizing the first load rate based on the highest load rate and the lowest load rate in the historical operation data is In the formula, Y is the normalized first load rate, Y i is the unnormalized first load rate, Y max is the highest load rate, Y minis the lowest load rate. Through normalization, the influence of different dimensions and data ranges on data analysis is eliminated, making the data at the same order of magnitude, facilitating comparison and analysis, and improving the comparability of the data. With the key line name, i.e., the feeder name, on the abscissa and the clustering analysis feature, i.e., the eigenvalue, on the ordinate, each row of data represents an operation mode and is assigned a number as a unique identifier to construct data in matrix form, providing analysis data for the clustering analysis of the clustering algorithm.

[0036] S4: Use the feature matrix as the input of the clustering algorithm to obtain the operation modes of the distribution network process, and determine the typical operation modes of the distribution network for those operation modes of the distribution network process that meet the preset conditions.

[0037] The said S4 includes: S41: Randomly select several data from the feature matrix as the initial clustering centers; S42: Based on the initial clustering centers, perform sample point assignment, thereby updating the initial clustering centers and obtaining the number of updates; S43: If the initial clustering centers meet the convergence condition or the number of updates is greater than the preset number of updates, then execute S44; otherwise, execute S42; S44: Obtain the operation modes of the distribution network process based on the initial clustering centers, set the preset conditions based on the full coverage principle, and determine the typical operation modes of the distribution network for those operation modes of the distribution network process that meet the preset conditions.

[0038] The said S4 further includes: If there is no operation mode of the distribution network process that meets the preset conditions, then clean the historical operation data, process outliers and missing values, and execute S2.

[0039] In this embodiment, sample points are allocated based on the initial clustering centers, and then the initial clustering centers are re-obtained specifically as follows: calculate the distances between the data in the adjustment matrix, i.e., other samples and the initial clustering centers. Samples are allocated to the corresponding initial clustering centers according to the distances from the initial clustering centers. After all samples are allocated, calculate the center points of each category again, and select the one closest to the center position as the new clustering center. Repeat the above steps until the initial clustering centers no longer change significantly and stop. No longer changing significantly means meeting the convergence condition or the number of updates being greater than the preset number of updates. The convergence condition means that the moving distance of the initial clustering centers is less than 1e-4, and the preset number of updates is 300 times. Set the preset conditions based on the full coverage principle, and determine the typical operation mode of the distribution network that meets the preset conditions specifically as follows: If there are a preset number of similar distribution network process operation modes among several obtained distribution network process operation modes, then this similar distribution network process operation mode is the typical operation mode of the distribution network. The specific standard for similarity is as follows: For example, if the distribution network process operation mode obtained for the first time includes A, B, and C, the distribution network process operation mode obtained for the second time includes H, F, G, and M, and the distribution network process operation mode obtained for the third time includes P, Q, and W. At this time, the number of distribution network process operation modes in the first and third distribution network process operation modes is both 3. So, the first and third obtained distribution network process operation modes are initially similar. If A, B, and C obtained for the first time correspond to P, Q, and W obtained for the third time respectively, then determine whether the number of the same data in A and P, B and Q, and C and W is greater than the set number respectively. If all are greater, it means that the distribution network process operation mode obtained for the first time is similar to the distribution network process operation mode obtained for the third time. At this time, take the distribution network process operation mode obtained for the first time or the distribution network process operation mode obtained for the third time as the typical operation mode of the distribution network, or take the set of the distribution network process operation mode obtained for the first time and the distribution network process operation mode obtained for the third time as the typical operation mode of the distribution network. Obtain the typical operation mode of the distribution network through several random selections, avoiding the existing local optimal situation and improving the accuracy of the obtained typical operation mode of the distribution network. When all the data in the feature matrix are traversed by random selection and still do not meet the preset conditions, it indicates that there are missing or overly miscellaneous historical operation data, which in turn leads to the difficulty of the obtained key components and key lines in reflecting the actual operation of the power grid. Therefore, after processing the historical operation data, re-execute S2, thereby improving the flexibility of the present invention.

[0040] Embodiment 2: This embodiment also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the method for obtaining the typical operation mode of the distribution network considering voltage characteristics.

[0041] Embodiment 3: This embodiment also provides a storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are loaded and executed by a processor, the steps of the method for obtaining a typical operation mode of a distribution network considering voltage characteristics are implemented.

[0042] The above specific implementation manners are the preferred implementation manners of the method for obtaining a typical operation mode of a distribution network considering voltage characteristics of the present invention, and do not limit the specific implementation scope of the present invention. The scope of the present invention includes but is not limited to this specific implementation manner. All equivalent changes made according to the shape and structure of the present invention are within the protection scope of the present invention.

Claims

1. A method for obtaining a typical operation mode of a distribution network considering voltage characteristics, characterized in that: The following steps are involved: S1: Establish a power grid model based on the power grid architecture in the distribution network; S2: Based on the historical operation data of the distribution network and the power grid model, a state estimation algorithm is used to obtain an estimated value of a state variable, and based on the estimated value of the state variable, a key component, a first load rate of the key component, a key line, and a first voltage of the key line are obtained; S3: using the normalized first load rate and the first voltage as cluster analysis features to obtain a feature matrix; S4: Using the characteristic matrix as the input of the clustering algorithm to obtain the distribution network process operation mode, and determining the distribution network process operation mode that meets the preset conditions as the typical operation mode of the distribution network.

2. The method for obtaining a typical operation mode of a distribution network considering voltage characteristics according to claim 1, characterized in that: The S1 includes: S11: constructing an initial power grid model including nodes and lines according to the power grid architecture; S12: setting electrical parameters for electrical equipment at nodes in the initial power grid model and setting line parameters for lines in the initial power grid model according to the historical operation data; S13: Update the initial power grid model based on the electrical parameters and line parameters, obtain the measured electrical signals of the electrical equipment and the lines based on the initial power grid model, obtain the difference between the measured electrical signals and the historical electrical signals of the electrical equipment and the lines in the historical operation data, if the difference meets the preset requirements, the initial power grid model is the power grid model, otherwise adjust the electrical parameters or line parameters according to the difference, and execute S13.

3. The method for obtaining a typical operation mode of a distribution network considering voltage characteristics according to claim 2, characterized in that: The S2 includes: S21: setting an initial estimated value of a state variable according to an average value of the historical operation data; S22: obtaining a relationship between an actual measured value of a state variable and an initial estimated value according to the initial estimated value and a power grid model, and then obtaining a partial derivative of the state variable with respect to the relationship; S23: Obtain a weight matrix based on the reliability of the actual measurement value, and obtain an information matrix based on the partial derivative and the weight matrix; S24: obtaining a correction amount of the state variable according to the information matrix and the difference between the actual measured value and the initial estimated value; S25: obtaining an estimated value of the state variable based on the correction amount, and obtaining a key element, a first load rate of the key element, a key line, and a first voltage of the key line based on the estimated value of the state variable.

4. The method for obtaining a typical operation mode of a distribution network considering voltage characteristics according to claim 3, characterized in that: In S25, obtaining an estimated value of the state variable based on the correction amount includes: The absolute value of the correction is compared with the preset absolute value. If the absolute value of the correction is smaller than the preset absolute value, an estimated value of the state variable is obtained according to the correction. Otherwise, the initial estimated value is updated based on the correction and S22 is executed.

5. The method for obtaining a typical operation mode of a distribution network considering voltage characteristics according to claim 3, characterized in that: In S25, the step of acquiring the key element, the first load rate of the key element, the key circuit and the first voltage of the key circuit based on the estimated value of the state variable includes: Acquire a second load rate of the electrical device based on a first estimated value among the estimated values, and acquire a second voltage of the line based on a second estimated value among the estimated values; An electrical device whose second load rate is greater than a preset load rate is taken as a key component, and the first load rate of the key component is obtained, the second voltage is compared with the rated voltage, the line for which the comparison is unsuccessful is taken as a key line, and the first voltage of the key line is obtained.

6. The method for obtaining a typical operation mode of a distribution network considering voltage characteristics according to claim 1, characterized in that: The S3 includes: S31: normalizing the first voltage based on the highest voltage and the lowest voltage in the historical operation data, and normalizing the first load rate based on the highest load rate and the lowest load rate in the historical operation data; S32: Acquire a feature matrix based on the key line names and cluster analysis features.

7. The method for obtaining a typical operation mode of a distribution network considering voltage characteristics according to claim 1, characterized in that: The S4 includes: S41: randomly selecting a number of data from the feature matrix as initial clustering centers; S42: Allocate sample points based on the initial cluster center, thereby updating the initial cluster center and obtaining the number of updates; S43: If the initial cluster center meets the convergence condition or the update number is greater than the preset update number, execute S44, otherwise execute S42; S44: Obtaining the distribution network process operation mode based on the initial cluster center, setting preset conditions based on the full coverage principle, and determining the distribution network process operation mode that meets the preset conditions as the typical operation mode of the distribution network.

8. The method for obtaining a typical operation mode of a distribution network considering voltage characteristics according to claim 7, characterized in that: The S4 further comprises: If there is no distribution network process operation mode that meets the preset conditions in the distribution network process operation mode, the historical operation data is cleaned, abnormal values ​​and missing values ​​are processed, and S2 is executed.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method for acquiring a typical operation mode of a distribution network considering voltage characteristics are implemented as described in any one of claims 1 to 8.

10. A storage medium, characterized in that: The storage medium stores computer executable instructions, and when the computer executable instructions are loaded and executed by the processor, the steps of the method for obtaining a typical operating mode of a distribution network considering voltage characteristics as described in any one of claims 1 to 8 are implemented.

Images