Power system transient stability assessment method and related device

Abstract

The invention belongs to the technical field of power system stability discrimination, and discloses a power system transient stability assessment method and a related device. The power system transient stability evaluation method comprises the following steps: calculating to obtain transient energy of a power system; calculating to obtain critical energy based on the critical energy function of the two-machine equivalent model of the power system; correcting the critical energy by adopting the trained critical energy added value enhancement model to obtain corrected critical energy; and comparing the transient energy with the corrected critical energy, and obtaining a transient stability evaluation result of the power system based on a comparison result. According to the technical scheme, the advantages that a direct method has good interpretability and an artificial intelligence method has high precision are combined, and an accurate and mechanically interpretable stability judgment result can be quickly obtained according to the current state and possibly suffered faults of the power system.

Description

Technical Field

[0001] The present invention belongs to the technical field of power system stability determination, and in particular relates to a power system transient stability assessment method and related devices. Background Art

[0002] The power industry is a basic industry for social development and the national economy, and occupies a key position in the national economy. The safe and stable operation of the power system is a basic requirement for social progress and modern life. Once the power system loses stability, it will have a significant impact on social security and the national economy.

[0003] As renewable energy generation continues to increase, the use of large-scale power electronic equipment in the power grid has made the transient dynamic behavior of power systems increasingly complex after faults, and the accurate prediction of power system transient stability faces new challenges (the accuracy and reliability of transient stability assessment are crucial to the safe operation of the system and the best decision-making after faults).

[0004] Traditional methods for transient stability assessment of power systems mainly include time domain simulation and direct methods. The time domain simulation method formulates power system instability criteria, solves the system differential equations and determines the curves of each state variable, and incorporates the dynamic model of the electrical components into the analysis to obtain transient stability analysis results. However, the solution process is extremely time-consuming and requires high accuracy in modeling the power system, making it difficult to meet real-time decision-making needs. Direct methods (such as the extended equal area rule, transient energy function method, etc.) have fast calculation speed, intuitive and clear stability judgment principles, good interpretability, and can directly judge stability. However, for modern large-scale power systems, the grid structure is complex and modeling is difficult, making direct methods difficult to apply.

[0005] In recent years, the application of artificial intelligence in power systems has made great progress. By training a large number of system transient stability samples and mining the intrinsic relationship between electrical quantity data information and system stability, a model for evaluating the transient stability of power systems can be obtained. The main machine learning algorithms currently used in transient stability assessment include decision trees, deep belief networks, support vector machines, and graph neural networks. Although artificial intelligence methods have the advantages of faster judgment speed than time domain simulation methods and higher accuracy than direct methods, their stability judgment mechanism is usually difficult to describe, and the interpretability of their judgment results is difficult to reflect. It is difficult for power system operators to understand the judgment results, and it is difficult to take relevant measures to prevent power system instability. In summary, the currently commonly used power system transient stability judgment methods all have certain defects, and there is an urgent need for a new power system stability judgment method that is both interpretable and can achieve high accuracy. Summary of the invention

[0006] The purpose of the present invention is to provide a method and related device for evaluating transient stability of a power system to solve one or more of the above-mentioned technical problems. The technical solution disclosed in the present invention is specifically a transient stability evaluation scheme for a power system based on an intelligent enhanced transient energy function, which combines the advantages of the direct method with good interpretability and the artificial intelligence method with high accuracy, and can quickly obtain accurate and mechanistically interpretable stability judgment results based on the current state of the power system and possible faults.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions: In a first aspect, the present invention provides a method for evaluating transient stability of a power system, comprising the following steps: Using the transient energy function, the transient energy of the power system to be evaluated for transient stability is calculated; Based on the critical energy function of the two-machine equivalent model of the power system to be evaluated for transient stability, the critical energy is calculated; Using the trained critical energy added value enhancement model, the obtained critical energy is corrected to obtain a corrected critical energy; The obtained transient energy is compared with the corrected critical energy, and a transient stability evaluation result of the power system to be transiently evaluated is obtained based on the comparison result.

[0008] A further improvement of the present invention is that in the step of using the transient energy function to calculate and obtain the transient energy of the power system to be transiently stabilized, The calculation expression of transient energy is: ; In the formula, is the transient energy of the power system; is the system kinetic energy; is the system potential energy.

[0009] A further improvement of the present invention is that in the step of calculating and obtaining the critical energy based on the critical energy function of the two-machine equivalent model of the power system to be transiently stabilized, The calculation expression of critical energy is: ; In the formula, represents the calculated critical energy; , All represent intermediate quantities obtained based on the two-machine equivalent model; Represents the power angle corresponding to the stable equilibrium point.

[0010] A further improvement of the present invention is that, in the step of calculating the critical energy based on the critical energy function of the two-machine equivalent model of the power system to be transiently stabilized, the step of obtaining the two-machine equivalent model includes: Obtain the topological relationship of the power system and the relevant electrical parameters of the generators and transmission lines in the power system, and establish a power system simulation model; based on the power system simulation model, perform transient stability simulation on the fault scenarios with expected fault concentration under multiple operating conditions, and obtain the historical data of the power angle of each generator under different operating conditions; Perform feature extraction and dimensionality reduction processing on the historical data of power angles of each generator under different operating conditions, and obtain the power angle data of the generator after feature extraction and dimensionality reduction processing; Based on the obtained feature extraction and dimensionality reduction data of the generator power angle, combined with the relevant electrical parameters of the generator and transmission lines in the power system, a clustering method is used to establish a two-machine equivalent model of the power system.

[0011] A further improvement of the present invention is that In the step of performing feature extraction and dimensionality reduction processing on the historical data of the power angle of each generator under different operating conditions to obtain the power angle data of the generator after feature extraction and dimensionality reduction processing, feature extraction is performed based on the change of S, and the principal component analysis method is used to perform dimensionality reduction processing on the power angle data of the generator after feature extraction; In the step of using a clustering method to establish a two-machine equivalent model of the power system based on the obtained feature extraction and dimensionality reduction processed generator power angle data, combined with relevant electrical parameters of the generator and transmission lines in the power system, a k-means clustering method is used.

[0012] A further improvement of the present invention is that the critical energy obtained is corrected by using the trained critical energy added value enhancement model, and in the step of obtaining the corrected critical energy, The calculation expression of the corrected critical energy is: ; In the formula, represents the calculated corrected critical energy; represents the calculated critical energy; represents the critical energy added value output by the critical energy added value enhancement model; The training steps of the critical energy added value enhancement model include: The power angle of the generator in the power system after the fault is used as the initial characteristic input, and the critical energy added value is used as the output of the critical energy added value enhanced model; Based on the training sample set, the convolutional neural network is trained. After reaching the preset convergence condition, the trained critical energy added value enhancement model is obtained. Each sample in the training sample set includes feature vectors at multiple moments. The feature vector at time ; In the formula, Indicates the generator number, for Moment The power angle change of the generator, They are the actual transient energy of the power system and the critical energy of the two-machine equivalent system.

[0013] A further improvement of the present invention is that in the step of comparing the obtained transient energy with the corrected critical energy and obtaining a transient stability evaluation result of the power system to be transiently stabilized based on the comparison result, If the transient energy is less than the corrected critical energy, the transient stability assessment result of the power system to be transiently stabilized is stable; If the transient energy is greater than or equal to the corrected critical energy, the transient stability assessment result of the power system to be transiently stabilized is unstable or in a critically stable state.

[0014] A second aspect of the present invention provides a power system transient stability assessment system, comprising: A first calculation module is used to calculate and obtain the transient energy of the power system to be transiently stabilized by using a transient energy function; A second calculation module is used to calculate and obtain critical energy based on a critical energy function of a two-machine equivalent model of the power system to be evaluated for transient stability; A correction module, used to correct the obtained critical energy by using the trained critical energy added value enhancement model to obtain the corrected critical energy; The comparison and evaluation module is used to compare the obtained transient energy with the corrected critical energy, and obtain a transient stability evaluation result of the power system to be transiently stabilized based on the comparison result.

[0015] According to a third aspect of the present invention, there is provided an electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for evaluating transient stability of an electric power system as described in any one of the first aspect of the present invention is implemented.

[0016] According to a fourth aspect of the present invention, there is provided a non-transitory computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the method for evaluating transient stability of a power system as described in any one of the first aspects of the present invention is implemented.

[0017] Compared with the prior art, the present invention has the following beneficial effects: In the transient stability assessment method of the power system disclosed in the present invention, the transient energy function is used to calculate the transient energy of the power system, which can accurately reflect the energy change of the power system after being disturbed, and provide accurate basic data for the subsequent transient stability assessment; the critical energy function of the two-machine equivalent model is introduced to further refine the assessment standard, so that the assessment result is closer to the actual operating state of the power system; the critical energy is corrected using the trained critical energy added value enhancement model, which effectively considers the influence of various complex factors in the system on the critical energy, and the assessment results are compared after correction, which significantly improves the accuracy of the transient stability assessment of the power system. It should be emphasized that by correcting the trained critical energy added value enhancement model, it is possible to continuously learn and adapt to the new characteristics and changes of the power system, so that the assessment method can keep pace with the times and maintain the accuracy and effectiveness of the assessment results. In summary, the present invention gradually approaches the final evaluation result by calculating transient energy, critical energy, correcting critical energy, comparative evaluation and other steps, thereby improving the evaluation efficiency; in addition, the trained model is used for correction to avoid complex real-time calculations, making the evaluation process faster and more efficient; the evaluation steps are clear and easy to understand and implement. In practical applications, accurate evaluation results can be obtained by simply following the steps; the evaluation results are presented in an intuitive manner, which is convenient for decision makers to quickly understand the transient stability status of the power system and formulate corresponding control strategies and measures accordingly, which can provide strong technical support for the safe and stable operation of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below; obviously, the drawings described below are some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 It is a flow chart of a method for evaluating transient stability of a power system in an embodiment of the present invention.

[0020] Figure 2 It is a flow chart of a method for evaluating transient stability of a power system based on an intelligent enhanced transient energy function in a specific embodiment of the present invention.

[0021] Figure 3 It is a schematic diagram of a two-machine equivalent model of an IEEE39 node test system selected in an embodiment of the present invention.

[0022] Figure 4 It is a schematic diagram of a comparison curve between the transient energy of the generator calculated by the method of the embodiment of the present invention and the corrected critical energy.

[0023] Figure 5 It is a schematic diagram of a power system transient stability assessment system in an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention; it is obvious that the described embodiments and technical solutions are only part of the embodiments of the present invention, not all of the embodiments.

[0025] All other embodiments obtained by those of ordinary skill in the art without creative work based on the technical solutions disclosed in the embodiments of the present invention belong to the scope of protection of the present invention. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device including a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] See also Figure 1 , a method for evaluating transient stability of a power system provided by an embodiment of the present invention comprises the following steps: Step 1, using a transient energy function, to calculate and obtain the transient energy of the power system to be transiently stabilized; Step 2, based on the critical energy function of the two-machine equivalent model of the power system to be transient stability evaluated, calculate and obtain the critical energy; Step 3, using the trained critical energy added value enhancement model to correct the critical energy obtained in step 2 to obtain the corrected critical energy; Step 4: compare the transient energy obtained in step 1 with the corrected critical energy obtained in step 3, and obtain a transient stability assessment result of the power system based on the comparison result.

[0027] In the technical solution provided by the embodiment of the present invention, by using the transient energy function to calculate the transient energy of the power system, the energy change of the power system after being disturbed can be accurately reflected, and accurate basic data can be provided for transient stability evaluation; the critical energy function of the two-machine equivalent model is introduced to further refine the evaluation criteria, so that the evaluation results are closer to the actual operating state of the power system, and the critical energy is corrected by using the trained critical energy value-added enhancement model, which effectively considers the influence of various complex factors (such as nonlinear elements, load changes, network structure changes, etc.) on the critical energy in the system, thereby significantly improving the accuracy of the evaluation. The method of the embodiment of the present invention does not rely on a specific power system model or assumptions, and has strong versatility and adaptability. Whether it is a large power system or a small power system, whether it is a simple network or a complex network, the method of the present invention can be used for transient stability evaluation. It should be emphasized that by training the critical energy value-added enhancement model, the new characteristics and changes of the power system can be continuously learned and adapted, so that the evaluation method can keep pace with the times and maintain the accuracy and effectiveness of the evaluation results.

[0028] In summary, the method provided by the embodiment of the present invention decomposes the complex problem of transient stability assessment of the power system into a plurality of relatively simple steps, and gradually approaches the final assessment result by calculating transient energy, critical energy, and correcting critical energy, thereby improving the efficiency of the assessment; in addition, the use of a trained model for correction avoids complex real-time calculations, making the assessment process faster and more efficient. The assessment steps provided by the method of the present invention are clear and unambiguous, easy to understand and implement. In practical applications, accurate assessment results can be obtained by simply following the steps; the assessment results are presented in an intuitive manner, which facilitates decision makers to quickly understand the transient stability status of the power system and formulate corresponding control strategies and measures accordingly. The technical solution of the embodiment of the present invention significantly improves the accuracy and efficiency of transient stability assessment of the power system by accurately calculating transient energy and critical energy and introducing a trained model for correction, and can provide strong technical support for the safe and stable operation of the power system.

[0029] In one embodiment of the present invention, in step 1, in the process of calculating and obtaining the transient energy of the power system to be transiently stabilized by using the transient energy function, the calculation expression of the transient energy is: ; In the formula, is the transient energy of the power system; is the system kinetic energy; is the system potential energy.

[0030] Explanatory, the kinetic energy and potential energy of the power system are calculated separately, and then summed to obtain the transient energy of the actual power system.

[0031] In the preferred technical solution provided by the embodiment of the present invention, a specific function for calculating the transient energy of the power system is given, which can provide accurate basic data for subsequent transient stability evaluation.

[0032] In one embodiment of the present invention, in step 2, in the process of calculating and obtaining the critical energy based on the critical energy function of the two-machine equivalent model of the power system to be transiently stabilized, the calculation expression of the critical energy function is: ; In the formula, represents the calculated critical energy; , All represent intermediate quantities obtained based on the two-machine equivalent model; Represents the power angle corresponding to the stable equilibrium point.

[0033] In one embodiment of the present invention, in step 3, the critical energy obtained in step 2 is corrected using the trained critical energy added value enhancement model. In the process of obtaining the corrected critical energy, the calculation expression of the corrected critical energy is: ; In the formula, represents the calculated corrected critical energy; It represents the critical energy added value output by the critical energy added value enhancement model.

[0034] Explanatory, in the preferred technical solution of the embodiment of the present invention, using Correcting the critical energy of the two-machine equivalent model avoids complex real-time calculations and makes the evaluation process faster. The stability judgment result obtained by comparing the corrected critical energy with the transient energy value meets the actual stability situation and can significantly improve the accuracy and efficiency of the transient stability evaluation of the power system.

[0035] In one embodiment of the present invention, the step of obtaining a two-machine equivalent model of a power system to be evaluated for transient stability includes: Step 1) Obtain the topological relationship of the power system, the relevant electrical parameters of the generators and transmission lines in the power system, and establish a power system simulation model; Based on the power system simulation model, perform transient stability simulation on the fault scenarios of the expected fault concentration under multiple operating conditions, and obtain the historical data of the power angle of each generator under different operating conditions; Step 2), based on the change of S, feature extraction is performed on the historical data of the power angle of each generator obtained in step 1), and on this basis, the power angle data after feature extraction is reduced in dimension using the principal component analysis method to obtain the power angle data of the generator after feature extraction and dimension reduction; Step 3), based on the generator power angle data obtained in step 2) after feature extraction and dimensionality reduction, combined with the relevant electrical parameters of the generators and transmission lines in the power system, a two-machine equivalent model of the original power system is established using the k-means clustering method.

[0036] In a further exemplary preferred technical solution, in step 1), the relevant electrical parameters of the generator and the transmission line in the power system include: generator inertia time constant, generator transient reactance, generator transient reactance, generator transient potential, transmission line impedance and node voltage at both ends of the transmission line; in addition, the historical data of the power angle of each generator under different operating conditions include: generator power angle data under normal operating conditions, generator power angle data after fault; fault conditions include three-phase short circuit, two-phase short circuit to ground, two-phase short circuit, single-phase short circuit to ground, broken line and other line faults; lines where faults may occur include AC lines with higher voltage levels in the power system grid (in the exemplary technical solution, a comparison threshold can be set to distinguish) and all DC lines; possible fault locations on each line include random positions of the transmission line.

[0037] In a further exemplary preferred technical solution, in step 2), in the process of extracting features from the historical data of the power angles of each generator obtained in step 1) based on the S change, a Gaussian window function with an adjustable width is used to perform an S transformation on the historical data of the power angles of each generator to extract its feature data, and the extracted feature data is transformed with the time-frequency spectrum matrix obtained by the S transformation Indicates that the expression is: ; ; In the formula, Represents the time-frequency spectrum matrix obtained by transformation; Indicates the initial power angle data of the generator; is the Gaussian window function, in the Gaussian window function definition, is the shape parameter of the Gaussian window function; , Respectively Time and frequency in the domain, Used to control the time position of the window function. Used to control the width of the window function.

[0038] In a further exemplary preferred technical solution, in step 2), in the process of reducing the dimension of the power angle data after feature extraction by using the principal component analysis method, the data is processed according to the principal component analysis method to remove unimportant features and noise in the power angle data and improve the efficiency of feature extraction; wherein, Power angle characteristic data Perform principal component analysis to make it Dimensionality reduction The steps of dimensioning include: firstly, decentralizing the power angle data and subtracting the mean value from each eigenvalue; then calculating the covariance. dimensional data set, whose covariance is a symmetric matrix called The covariance matrix of The scatter matrix representing the degree of sample dispersion can be obtained; select the first The largest eigenvalue, its corresponding The eigenvectors are each used as the eigenvector matrix composed of row vectors ; In summary, the power angle characteristic data from Dimensionality reduction The expression of dimension is: ; To explain more specifically, the principal component analysis method can be used to process the data according to the following formula, including: ; Explanatory, this formula represents the decentralized processing of power angle data, Indicates Dimensional data; ; Explanatoryally, this formula means The covariance matrix of ;Explanatory, this formula represents the scatter matrix that measures the dispersion of the sample; ; Explanatoryally, this formula means from Dimensionality reduction dimension, Represents the covariance matrix The largest eigenvalue, its corresponding The eigenvectors are respectively used as the eigenvector matrix composed of row vectors.

[0039] In a further preferred technical solution, in step 3), in the process of establishing the two-machine equivalent model of the original power system using the k-means clustering method, the clustering process of the k-means clustering method specifically includes: (1) Select two initial cluster centers; (2) Calculate the distance between the remaining data samples and the cluster center: (3) Update cluster center: (4) Repeat steps (2) and (3) until the updated cluster center is consistent with the previous cluster center or has only a slight movement (less than the set threshold); Among them, the distance between the remaining data samples and the cluster center can be calculated according to the following formula, expressed as: ; In the formula, Indicates the data dimension, Indicates data samples, Indicates Cluster centers; The cluster center is updated according to the following formula, expressed as: ; In the formula, Indicates that it is assigned to All data samples of group clustering The set composed of Indicates that there are a total of data samples, Represents data samples No. The value of the dimension data.

[0040] In one embodiment of the present invention, in step 3), the node admittance matrix of the two-machine equivalent model is expressed as:

[0041] in, , Respectively represent nodes Self-admittance and nodes The mutual admittance between express The admittance angle.

[0042] Among them, calculating the equivalent parameters and constructing the two-machine equivalent model of the power system include using the following formula for calculation: ; ; ; ;

[0043]

[0044] in, , represents the equivalent two-machine inertia time constant; , Represents the equivalent transient potential of two machines; , Represents the equivalent transient reactance of two machines; , Indicates equivalent load.

[0045] In one embodiment of the present invention, a unit coherence criterion is used as an effective indicator for measuring the similarity between units, including using the following formula for calculation: ; ; in, Generator and generator The power angle change; For observation time; Generator and generator exist The power angle of the moment, is the proportionality coefficient. In one embodiment of the present invention, in step 3, the critical energy obtained in step 2 is corrected using a trained critical energy added value enhancement model, and in the process of obtaining the corrected critical energy, the derived critical energy is enhanced based on a convolutional neural network, and the derived critical energy is enhanced based on a convolutional neural network; wherein, based on the expected data set, a training sample set is obtained, and a critical energy added value enhancement model is constructed; In a specific exemplary technical solution, the step of obtaining a trained critical energy added value enhancement model includes: Firstly, transient stability simulation is carried out in the simulation environment to obtain the state observations and related data sets of the power system; Calculate the transient energy of the actual power system under different faults according to the calculation formula; The transient energy obtained by simulation calculation, the critical energy calculated by the two-machine equivalent system and the generator power angle change curve after the fault are input into the neural network. The local features in the input data are extracted through the convolution operation of the convolution layer. The pooling layer samples the generator power angle feature data to reduce the feature dimension, retain its important features, and reduce the training time. Finally, the fully connected layer returns the enhanced critical energy added value, so that the enhanced critical energy can meet the actual stability of the system.

[0046] In one embodiment of the present invention, the transient energy of the actual power system is obtained according to the transient energy calculation formula, the critical energy additional value is output by the critical energy additional value enhancement model and the critical energy of the two-machine equivalent system is corrected, and the transient energy is compared with the corrected critical energy to obtain the stability judgment result.

[0047] Explanatory, transient energy reflects the energy exchange of the power system after a disturbance or fault; the transient energy of the actual power system is compared with the corrected critical energy: if the transient energy of the actual power system is less than the corrected critical energy, then the system can be considered stable because the system has sufficient energy margin to withstand external disturbances or faults; if the transient energy of the actual power system is greater than or equal to the corrected critical energy, then the system can be considered unstable or in a critical stability state because the system may not be able to withstand further disturbances or faults.

[0048] In summary, the embodiment of the present invention discloses a method for evaluating transient stability of a power system based on an intelligent enhanced transient energy function, including the following steps: obtaining relevant electrical parameters within the scope of the power system to be judged for stability and historical data related to the operation of the power grid; performing S transformation on the power angle data of the generator set and extracting characteristic data using the principal component analysis method; clustering the generators into two groups using the k-means method; deriving the critical energy function of the two-machine equivalent model according to the grouping results; correcting the critical energy based on a convolutional neural network, establishing a critical energy added value enhancement model, and performing stability judgment based on the enhanced model. The technical solution provided by the embodiment of the present invention overcomes the problem of poor interpretability of judging system stability using a pure artificial intelligence method, and has a high accuracy rate in judging stability.

[0049] See also Figures 2 to 4 In a specific embodiment of the present invention, a method for evaluating transient stability of a power system based on an intelligent enhanced transient energy function is provided, comprising the following steps: Step S1, obtaining the topological relationship of the power system and the electrical parameters related to the generators and transmission lines in the system, establishing a power system simulation model, generating an expected fault set, performing transient stability simulation on the fault scenarios in the expected fault set under multiple working conditions, and obtaining the power angle data of each generator; Specifically, the generator power angle history data includes the generator power angle data after the fault; the fault conditions should include line faults such as three-phase short circuit to ground, two-phase short circuit to ground, two-phase short circuit, single-phase short circuit to ground, etc.; the lines where the fault may occur include AC lines with higher voltage levels in the power system grid and all DC lines; the possible fault locations on each line include random locations of the transmission line; the expected fault set is generated according to the above faults; among them, in the IEEE39-node system, there are 10 generators, 39 nodes, and 46 AC lines, and on all 46 lines The most serious three-phase short-circuit fault was set. The fault occurred in 0.1s and lasted from 0.03s to 0.20s with a step size of 0.01s. The occurrence position was 10% away from the left node of each line. The node system load fluctuated randomly from 80% to 120% under the benchmark load, with a total of 11 operating modes. The simulation time was 10s with a step size of 0.01s. The stability of the sample was marked according to whether the maximum power angle difference of the generator exceeded 360°. Based on the above faults, a total of 9108 samples were generated, including 2666 unstable samples and 6442 stable samples.

[0050] Step S2, obtaining the historical data of the power angle of each generator under the different operating conditions, extracting features of the power angle data based on the change of S, and on this basis, performing dimensionality reduction processing on the feature data using the principal component analysis method; Explanatory, S transform is a time-frequency analysis tool for signal analysis and processing, combining the advantages of Fourier transform and wavelet transform. It is mainly used to analyze non-stationary signals, that is, signals whose frequency characteristics change over time. The basic idea of ​​S transform is to reveal the frequency components of the signal and its changes over time by converting the signal between the time domain and the frequency domain; principal component analysis is a commonly used statistical analysis technique, mainly used for data dimensionality reduction and feature extraction. Its basic idea is to project high-dimensional data into a new low-dimensional space through linear transformation to retain the main information and structural characteristics in the data.

[0051] Step S3, based on the generator power angle data after feature extraction and dimensionality reduction processing, and combined with the relevant electrical parameters of the generator and the transmission line, a two-machine equivalent model of the original system is established using the k-means clustering method; Explanatory, k-means clustering is a widely used unsupervised learning algorithm for dividing a data set into k different clusters, so that the similarity of data points within the same cluster is high, while the similarity of data points between different clusters is low. The main goal of k-means clustering is to achieve data partitioning by minimizing the square error within the cluster. Since the present invention is to establish a two-machine equivalence model, k can be set to 2.

[0052] S4. Based on the two-machine equivalent model of the original system, the transient energy function and critical energy function calculation formula under this model are derived; where:

[0053]

[0054]

[0055]

[0056]

[0057]

[0058]

[0059] In the formula, is the system transient energy, is the system kinetic energy, is the system potential energy; is the system generator potential energy, is the system AC load potential energy, is the potential energy related to the transmission power between the generator bus and the load bus and the transmission power between the load buses, is the potential energy associated with the power transferred between the AC bus and the commutation bus, , , , are the voltage amplitude and phase angle of the bus at the stable equilibrium point of the system after the fault is cleared, , , , are the voltage amplitude and phase angle of the busbar at the end of the fault, ; Calculate the kinetic energy and potential energy of the power system separately, and sum them up to get the actual power system transient energy; The critical energy of the two-machine equivalent model is calculated according to the following formula, expressed as: ; ; in, represents the power angle corresponding to the stable equilibrium point; It represents the critical energy added value output by the critical energy added value enhancement model.

[0060] The transient energy and critical energy calculation formula are calculated according to the following formula: , , , :

[0061]

[0062] in, It represents the transient potential of two machines of equal value; Represents the equivalent two-machine inertia time constant.

[0063] Step S5, enhancing the derived critical energy based on a convolutional neural network to obtain a critical energy added value enhancement model; wherein, The training process of the convolutional neural network is as follows: Considering that the initial features should be able to effectively reflect the transient process of the system, and real-time data collection is feasible, the power angle of the system generator after the fault is selected as the initial feature input. First, the constructed neural network is trained offline: a training sample set is established. For a sample size of The sample set, each sample , The feature vector at time , where Number the generator, For the The power angle change of each generator is used to characterize the power angle stability of the system. The network structure with three convolutional layers and pooling layers, which respectively represent the actual transient energy of the system and the critical energy of the two-machine system, can fully extract the characteristics of stable and unstable conditions without making the network structure redundant.

[0064] The actual stability of the system is reflected by the power angle data in the input eigenvector; the output of the critical energy additional value enhancement model is the critical energy additional value, and the modified critical energy should be added with this additional value on the basis of the critical energy calculated by the two-machine equivalent model.

[0065] The actual power system transient energy is obtained according to the transient energy calculation formula, and the critical energy added value is output by the critical energy added value enhancement model. , and the modified critical energy is obtained , compare the two values ​​to get the stability judgment result. Figure 4 It can be seen from the comparison that after the critical energy is corrected by the convolutional neural network, the stability judgment accuracy of the transient energy function method has been greatly improved.

[0066] The following are device embodiments of the present invention, which can be used to implement the method embodiments of the present invention. For details not disclosed in the device embodiments, please refer to the method embodiments of the present invention.

[0067] See also Figure 5 In an embodiment of the present invention, a power system transient stability assessment system is provided, comprising: A first calculation module is used to calculate and obtain the transient energy of the power system to be transiently stabilized by using a transient energy function; A second calculation module is used to calculate and obtain critical energy based on a critical energy function of a two-machine equivalent model of the power system to be evaluated for transient stability; A correction module, used to correct the obtained critical energy by using the trained critical energy added value enhancement model to obtain the corrected critical energy; The comparison and evaluation module is used to compare the obtained transient energy with the corrected critical energy, and obtain a transient stability evaluation result of the power system to be transiently stabilized based on the comparison result.

[0068] In one embodiment of the present invention, a computer device is provided, the computer device comprising a processor and a memory, the memory being used to store a computer program, the computer program comprising program instructions, and the processor being used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, and is specifically suitable for loading and executing one or more instructions in a computer storage medium to implement a corresponding method flow or a corresponding function; the processor described in the embodiment of the present invention can be used to perform the operation of the transient stability assessment method of the power system.

[0069] In one embodiment of the present invention, a storage medium is provided, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device for storing programs and data. It is understandable that the computer-readable storage medium here can include both built-in storage media in the computer device and, of course, extended storage media supported by the computer device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM (Random Access Memory) memory, or a non-volatile memory (non-volatile memory), such as at least one disk memory. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the transient stability assessment method of the power system in the above embodiment.

[0070] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, optical storage, etc.) containing computer-usable program codes.

[0071] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0072] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0073] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0074] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for evaluating transient stability of a power system, characterized in that: The following steps are involved: The transient energy function is used to calculate the transient energy of the power system to be evaluated for transient stability; Based on the critical energy function of the two-machine equivalent model of the power system to be evaluated for transient stability, the critical energy is calculated; Using the trained critical energy added value enhancement model, the obtained critical energy is corrected to obtain a corrected critical energy; The obtained transient energy is compared with the corrected critical energy, and a transient stability evaluation result of the power system to be transiently evaluated is obtained based on the comparison result.

2. A method for evaluating transient stability of a power system according to claim 1, characterized in that: In the step of using the transient energy function to calculate and obtain the transient energy of the power system to be transiently stabilized, The calculation expression of transient energy is: ； In the formula, is the transient energy of the power system; is the system kinetic energy; is the system potential energy.

3. A method for evaluating transient stability of a power system according to claim 1, characterized in that: In the step of calculating and obtaining the critical energy based on the critical energy function of the two-machine equivalent model of the power system to be transiently stabilized, The calculation expression of critical energy is: ； In the formula, represents the calculated critical energy; , All represent intermediate quantities obtained based on the two-machine equivalent model; Represents the power angle corresponding to the stable equilibrium point.

4. A method for evaluating transient stability of a power system according to claim 1, characterized in that: In the step of calculating and obtaining the critical energy based on the critical energy function of the two-machine equivalent model of the power system to be transiently stabilized, the step of obtaining the two-machine equivalent model includes: Obtain the topological relationship of the power system and the relevant electrical parameters of the generators and transmission lines in the power system, and establish a power system simulation model; based on the power system simulation model, perform transient stability simulation on the fault scenarios with expected fault concentration under multiple operating conditions, and obtain the historical data of the power angle of each generator under different operating conditions; Perform feature extraction and dimensionality reduction processing on the historical data of power angles of each generator under different operating conditions, and obtain the power angle data of the generator after feature extraction and dimensionality reduction processing; Based on the obtained feature extraction and dimensionality reduction data of the generator power angle, combined with the relevant electrical parameters of the generator and transmission lines in the power system, a clustering method is used to establish a two-machine equivalent model of the power system.

5. A method for evaluating transient stability of a power system according to claim 4, characterized in that: In the step of performing feature extraction and dimensionality reduction processing on the historical data of the power angle of each generator under different operating conditions to obtain the power angle data of the generator after feature extraction and dimensionality reduction processing, feature extraction is performed based on the change of S, and the principal component analysis method is used to perform dimensionality reduction processing on the power angle data of the generator after feature extraction; In the step of using a clustering method to establish a two-machine equivalent model of the power system based on the obtained feature extraction and dimensionality reduction processed generator power angle data, combined with relevant electrical parameters of the generator and transmission lines in the power system, a k-means clustering method is used.

6. A method for evaluating transient stability of a power system according to claim 1, characterized in that: In the step of using the trained critical energy added value enhancement model to correct the obtained critical energy and obtain the corrected critical energy, The calculation expression of the corrected critical energy is: ； In the formula, represents the calculated corrected critical energy; represents the calculated critical energy; represents the critical energy added value output by the critical energy added value enhancement model; The training steps of the critical energy added value enhancement model include: The power angle of the generator in the power system after the fault is used as the initial characteristic input, and the critical energy added value is used as the output of the critical energy added value enhanced model; Based on the training sample set, the convolutional neural network is trained. After reaching the preset convergence condition, the trained critical energy added value enhancement model is obtained. Each sample in the training sample set includes feature vectors at multiple moments. The feature vector at time ; In the formula, Indicates the generator number, for Moment The power angle change of the generator, They are the actual transient energy of the power system and the critical energy of the two-machine equivalent system.

7. A method for evaluating transient stability of a power system according to claim 1, characterized in that: In the step of comparing the obtained transient energy with the corrected critical energy and obtaining a transient stability evaluation result of the power system to be transiently stabilized based on the comparison result, If the transient energy is less than the corrected critical energy, the transient stability assessment result of the power system to be transiently stabilized is stable; If the transient energy is greater than or equal to the corrected critical energy, the transient stability assessment result of the power system to be transiently stabilized is unstable or in a critically stable state.

8. A power system transient stability assessment system, characterized in that: include: A first calculation module is used to calculate and obtain the transient energy of the power system to be transiently stabilized by using a transient energy function; A second calculation module is used to calculate and obtain critical energy based on a critical energy function of a two-machine equivalent model of the power system to be evaluated for transient stability; A correction module, used to correct the obtained critical energy by using the trained critical energy added value enhancement model to obtain the corrected critical energy; The comparison and evaluation module is used to compare the obtained transient energy with the corrected critical energy, and obtain a transient stability evaluation result of the power system to be transiently stabilized based on the comparison result.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the method for evaluating transient stability of a power system according to any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for evaluating transient stability of a power system according to any one of claims 1 to 7 is implemented.

Images