Power system transient stability evaluation method and device based on priority strategy

Abstract

The invention relates to the technical field of power system evaluation, and discloses a power system transient stability evaluation method and device based on a priority strategy, and the method comprises the steps: forming a first evaluation sequence according to a plurality of operation sample data of a power system; and updating the first evaluation sequence according to a numerical value comparison relationship of the information entropy indexes corresponding to the multiple pieces of operation sample data to form a second evaluation sequence, and performing transient stability category prediction on the second evaluation sequence by using a pre-trained neural network cascade model, optimizing a pre-trained neural network cascade model by using the prediction result; wherein the pre-trained neural network cascade model is used for inputting the operation sample data and outputting the prediction result of the transient stability category corresponding to the operation sample data, so that transient stability evaluation is efficiently performed on the large-scale power operation data.

Description

Technical Field

[0001] The present invention relates to the technical field of power system evaluation, and in particular to a method and device for evaluating transient stability of a power system based on a priority strategy. Background Art

[0002] With the continuous development and complexity of power systems, transient stability assessment of power systems has become a key link in ensuring the safe and stable operation of power grids. Traditional transient stability assessment methods often have problems such as large computational complexity, long assessment cycle, and poor adaptability to time-varying environments.

[0003] In recent years, the application of deep learning technology in power systems has gradually increased, providing new ideas for transient stability assessment. However, it is difficult to efficiently perform transient stability assessment on large-scale power operation data. Summary of the invention

[0004] In view of this, the present invention provides a method and device for transient stability assessment of a power system based on a priority strategy, which solves the technical problem that it is difficult to efficiently perform transient stability assessment on large-scale power operation data when faced with large-scale power operation data.

[0005] A first aspect of the present invention provides a method for evaluating transient stability of a power system based on a priority strategy, comprising:

[0006] forming a first evaluation sequence according to a plurality of operation sample data of the power system;

[0007] The first evaluation sequence is updated according to the numerical comparison relationship of the information entropy indicators respectively corresponding to the plurality of the running sample data to form a second evaluation sequence;

[0008] A pre-trained neural network cascade model is used to predict the transient stability category of the second evaluation sequence, and the pre-trained neural network cascade model is optimized using the prediction result; wherein the pre-trained neural network cascade model is used to input the operating sample data and output the prediction result of the transient stability category corresponding to the operating sample data.

[0009] Preferably, before the step of updating the first evaluation sequence according to the numerical comparison relationship of the information entropy indicators respectively corresponding to the plurality of the operating sample data to form the second evaluation sequence, the step further includes: calculating the information entropy indicators respectively corresponding to the plurality of the operating sample data;

[0010] The step of calculating the information entropy indicators corresponding to the plurality of running sample data respectively comprises:

[0011] For each of the operating sample data, the operating sample data is input into a pre-trained neural network model, and the predicted probability of each transient stability category corresponding to the operating sample data is output;

[0012] The predicted probabilities of each transient stability category corresponding to the operating sample data are used to perform weighted sum calculation to obtain the information entropy index corresponding to the operating sample data.

[0013] Preferably, the step of updating the first evaluation sequence according to the numerical comparison relationship of the information entropy indicators respectively corresponding to the plurality of the operating sample data to form the second evaluation sequence includes: comparing the information entropy indicators respectively corresponding to the plurality of the operating sample data with the information entropy threshold, and screening out the plurality of operating sample data whose information entropy indicators are greater than the information entropy threshold as key operating sample data;

[0014] Arrange the plurality of key operating sample data in descending order according to the numerical comparison relationship of the information entropy index to obtain a second evaluation sequence.

[0015] Preferably, the method of using a pre-trained neural network cascade model to predict the transient stability category of the second evaluation sequence and optimizing the pre-trained neural network cascade model using the prediction result; wherein, after the step of using the pre-trained neural network cascade model to input the running sample data and output the prediction result of the transient stability category corresponding to the running sample data, the method further comprises:

[0016] After the network parameters of the pre-trained neural network model are optimized, the information entropy index corresponding to each of the running sample data is recalculated based on the pre-trained neural network model after the network parameters are optimized;

[0017] The second evaluation sequence is updated using the recalculated numerical comparison relationship of the information entropy index corresponding to the running sample data.

[0018] Preferably, the pre-trained neural network cascade model includes a plurality of neural network models connected in series;

[0019] The step of using a pre-trained neural network cascade model to predict the transient stability category of the second evaluation sequence, and optimizing the pre-trained neural network cascade model using the prediction result, comprises:

[0020] Inputting the second evaluation sequence into the first neural network model in the multiple pre-trained neural network cascade models for prediction, and obtaining a transient stability category prediction result corresponding to the second evaluation sequence;

[0021] Determining the confidence of the second evaluation sequence by using the transient stability category prediction result and the transient stability category simulation result corresponding to the second evaluation sequence;

[0022] When the confidence of the second evaluation sequence does not reach a preset confidence threshold, a plurality of running sample data after expanding the current running simulation time window are updated in the second evaluation sequence;

[0023] Inputting the updated second evaluation sequence into a second neural network model adjacent to the first neural network model for prediction, to obtain a transient stability category prediction result corresponding to the updated second evaluation sequence;

[0024] Determining the confidence of the updated second evaluation sequence using the transient stability category prediction result and the transient stability category simulation result corresponding to the updated second evaluation sequence;

[0025] When the confidence of the updated second evaluation sequence does not reach the preset confidence threshold, the multiple running sample data after the current running simulation time window continue to be expanded to update the second evaluation sequence, and so on, until the confidence of the second evaluation sequence reaches the preset confidence threshold or reaches the preset maximum running time window, the iteration stops, and the neural network cascade model after the iteration stops is output.

[0026] Preferably, the step of updating the second evaluation sequence with a plurality of running sample data after expanding the current running simulation time window comprises:

[0027] Obtain multiple running sample data after expanding the current running simulation time window;

[0028] The second evaluation sequence is updated according to the obtained numerical comparison relationship of the information entropy indicators respectively corresponding to the multiple running sample data after the current running simulation time window is expanded.

[0029] Preferably, when the iteration of the neural network cascade model stops, the network parameters of the neural network model in the neural network cascade model are optimized using the input and output results of the last iteration of the neural network cascade model.

[0030] In a second aspect, the present invention further provides a power system transient stability assessment device based on a priority strategy, comprising:

[0031] A sequence forming module, used for forming a first evaluation sequence according to a plurality of operation sample data of the power system;

[0032] A sequence updating module, used for updating the first evaluation sequence according to the numerical comparison relationship of the information entropy indicators respectively corresponding to the plurality of the running sample data, to form a second evaluation sequence;

[0033] A model evaluation module is used to use a pre-trained neural network cascade model to predict the transient stability category of the second evaluation sequence, and to optimize the pre-trained neural network cascade model using the prediction result; wherein the pre-trained neural network cascade model is used to input the operating sample data and output the prediction result of the transient stability category corresponding to the operating sample data.

[0034] In a third aspect, the present invention further provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the power system transient stability assessment method based on the priority strategy as described in the first aspect.

[0035] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed, the steps of the method for transient stability assessment of a power system based on a priority strategy as described in the first aspect are implemented.

[0036] It can be seen from the above technical scheme that the present invention updates the sorting of the operating sample data by utilizing the numerical comparison relationship of the information entropy indicators corresponding to multiple operating sample data, thereby giving priority to the identified operating sample data with higher complexity into the evaluation sequence, ensuring priority processing of the operating sample data that has a more significant impact on the model performance improvement, and then using the pre-trained neural network cascade model to predict the transient stability category of the updated evaluation sequence, and using the prediction results to optimize the pre-trained neural network cascade model, thereby improving the accuracy and efficiency of the model evaluation, and at the same time enhancing the model's adaptability to the time-varying operating mode and topology of the power system, thereby efficiently performing transient stability evaluation on large-scale power operation data. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 A schematic diagram of an application environment of a power simulation system provided by an embodiment of the present invention;

[0038] Figure 2 A flowchart of a method for evaluating transient stability of a power system based on a priority strategy provided by an embodiment of the present invention;

[0039] Figure 3 A flowchart for updating a first evaluation sequence provided by an embodiment of the present invention;

[0040] Figure 4 A schematic diagram of the neural network cascade model evaluation principle provided by an embodiment of the present invention;

[0041] Figure 5A schematic diagram of the principle of batch temporary stability evaluation of a neural network cascade model provided by an embodiment of the present invention;

[0042] Figure 6 A schematic diagram of the structure of a power system transient stability assessment device based on a priority strategy provided by an embodiment of the present invention;

[0043] Figure 7 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0044] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] The present application provides a method for evaluating transient stability of a power system based on a priority strategy, which can be applied to Figure 1 In the application environment of the power simulation system shown. Among them, the simulator 101 is communicatively connected with the server 102, and the data storage system 103 can store the data that the server 102 needs to process. The simulator 101 can simulate the operating data of the power equipment, such as the time series data of the generator power angle. The server 102 obtains the operating data simulated by the simulator 101, and uses the simulated operating data to execute the transient stability assessment method of the power system based on the priority strategy. The data storage system 103 can be integrated on the server 102, or it can be placed on the cloud or other network servers. The server 102 can be an independent physical server, or it can be a server cluster or distributed system composed of multiple physical servers, or it can be a cloud server that provides cloud computing services.

[0046] like Figure 2 As shown, in Figure 1 The server 102 shown executes a method for evaluating transient stability of a power system based on a priority strategy provided in an embodiment of the present application, including: steps S1 to S3. Among them:

[0047] Step S1: forming a first evaluation sequence according to a plurality of operation sample data of a power system.

[0048] The operation sample data may be time series data of the generator power angle of the power system in the current simulation time window, and a plurality of operation sample data may form a first evaluation sequence according to a time series relationship.

[0049] Step S2: updating the first evaluation sequence according to the numerical comparison relationship of the information entropy indicators corresponding to the plurality of running sample data to form a second evaluation sequence.

[0050] Among them, the information entropy index can quantify the amount of information contained in each running sample data and its critical stability. The higher the information entropy index of the running sample data, the higher the amount of information contained in the running sample data and its critical stability, and the higher it should be arranged. Therefore, the first evaluation sequence is updated according to the numerical comparison relationship of the information entropy indicators corresponding to multiple running sample data to form a second evaluation sequence sorted according to the size of the information entropy index, so that the identified running sample data with higher complexity are given priority in the evaluation sequence to ensure that the running sample data that can significantly improve the model performance is given priority.

[0051] Step S3, using a pre-trained neural network cascade model to predict the transient stability category of the second evaluation sequence, and using the prediction result to optimize the pre-trained neural network cascade model; wherein the pre-trained neural network cascade model is used to input the running sample data and output the prediction result of the transient stability category corresponding to the running sample data.

[0052] Among them, the pre-trained neural network cascade model is a pre-trained neural network cascade model, whose input is labeled operating sample data, and whose output is the label of the transient stability category corresponding to the operating sample data. The transient stability categories include transient stability, critical stability and instability. The last layer structure of the neural network cascade model is a softmax layer, and finally the probability distribution of each transient stability category is output. The neural network cascade model selects the largest prediction probability of these three transient stability categories as its final prediction result for the transient stability category of the power system.

[0053] It should be noted that the embodiment of the present invention updates the sorting of the operating sample data by utilizing the numerical comparison relationship of the information entropy indicators corresponding to multiple operating sample data, thereby giving priority to the identified operating sample data with a higher degree of complexity into the evaluation sequence, ensuring that the operating sample data that has a more significant impact on the model performance is processed first, and then the pre-trained neural network cascade model is used to predict the transient stability category of the updated evaluation sequence, and the pre-trained neural network cascade model is optimized using the prediction results, thereby improving the accuracy and efficiency of the model evaluation, and at the same time enhancing the model's adaptability to the time-varying operating mode and topology of the power system, thereby efficiently performing transient stability evaluation on large-scale power operation data.

[0054] In some embodiments, the higher the information entropy index of the running sample data, the greater the uncertainty of its prediction result, and it may contain more complex or critical information, so it is more necessary to give priority to the evaluation. In order to accurately calculate the information entropy index of each running sample data, before step S2, it also includes: the step of calculating the information entropy index corresponding to the multiple running sample data.

[0055] Among them, the steps of calculating the information entropy indicators corresponding to multiple running sample data respectively include: steps S201~S202.

[0056] Step S201: for each running sample data, input the running sample data into a pre-trained neural network model, and output the predicted probability of each transient stability category corresponding to the running sample data.

[0057] The pre-trained neural network model is a pre-trained neural network model, whose input is labeled running sample data, and whose output is the predicted probability of the label of the transient stability category corresponding to the running sample data.

[0058] It should be noted that the neural network model can be a single neural network, a neural network model in a neural network cascade model, or a neural network model independent of the neural network cascade model, and there is no limitation here.

[0059] Step S202: Perform weighted sum calculation using the predicted probabilities of each transient stability category corresponding to the running sample data to obtain an information entropy index corresponding to the running sample data.

[0060] Among them, the calculation method of the information entropy index of the running sample data is:

[0061]

[0062] In the formula, is the information entropy index of running sample data i, It represents the output vector of the i-th running sample data evaluated by the neural network model. Each element represents the predicted probability of the sample being stable, critically stable, and unstable. is the index of the transient stability category, L is the number of transient stability categories, and is set to 3 in the present invention.

[0063] In some embodiments, Figure 3 As shown, in step S2, the first evaluation sequence is updated according to the numerical comparison relationship of the information entropy indicators corresponding to the plurality of running sample data to form the second evaluation sequence, including steps S211 to S212. Among them:

[0064] Step S211: compare the information entropy indicators corresponding to the multiple operation sample data with the information entropy threshold, and select the multiple operation sample data whose information entropy indicators are greater than the information entropy threshold as the key operation sample data.

[0065] The information entropy threshold is a preset information entropy experience value.

[0066] Step S212: Arrange the plurality of key operating sample data in descending order according to the numerical comparison relationship of the information entropy index to obtain a second evaluation sequence.

[0067] Among them, the identified key operating sample data will be given priority in the evaluation sequence, and arranged in order from high to low according to the information entropy index to form the second evaluation sequence, thereby ensuring the priority setting of the model evaluation process, so that samples with higher information value and potential influence can be evaluated earlier.

[0068] In some embodiments, the embodiments of the present invention set a dynamic update mechanism for the evaluation queue to enhance the adaptability of the model to the complex power system environment and ensure the real-time and reliable evaluation. After step S3 in the present application, steps S301 to S202 are also included. Among them:

[0069] Step S301: After the network parameters of the pre-trained neural network model are optimized, the information entropy index corresponding to each running sample data is recalculated based on the pre-trained neural network model after the network parameters are optimized.

[0070] It should be noted that when the neural network cascade model is unable to derive a certain number of samples of the transient stability category due to insufficient credibility, for example, the neural network cascade model has not concluded that the transient stability category is not stable after reaching the preset maximum number of iterations, it is necessary to optimize the network parameters of the pre-trained neural network model, and recalculate the information entropy index corresponding to each running sample data based on the pre-trained neural network model after the network parameter optimization, so that the second evaluation sequence is updated synchronously, thereby further optimizing the evaluation process and improving the evaluation efficiency and accuracy.

[0071] Step S302: Update the second evaluation sequence using the numerical comparison relationship of the information entropy index corresponding to the recalculated running sample data.

[0072] In some embodiments, the pre-trained neural network cascade model includes multiple neural network models connected in series. The neural network model can adopt a recurrent neural network (RNN). The multiple neural network models connected in series can be set to include a first neural network model, a second neural network model, a third neural network model...the nth neural network model.

[0073] In the embodiment of the present application, the process of using the pre-trained neural network cascade model to predict the transient stability category of the second evaluation sequence in step S3, and optimizing the pre-trained neural network cascade model using the prediction result, includes steps S311 to S316. Among them:

[0074] Step S311: input the second evaluation sequence into the first neural network model in the pre-trained neural network cascade model for prediction, and obtain a transient stability category prediction result corresponding to the second evaluation sequence.

[0075] The first neural network model may be the first neural network model in the neural network cascade model, or may be any other neural network model other than the last one.

[0076] Step S312: Determine the confidence of the second evaluation sequence by using the transient stability category prediction result and the transient stability category simulation result corresponding to the second evaluation sequence.

[0077] The transient stability category simulation result of the second evaluation sequence is determined by the transient stability time domain simulation of the simulator, and whether the final state is transient stability or instability, or critical instability. The transient stability category simulation result can be understood as the actual transient stability category. The confidence of the second evaluation sequence is determined by comparing the transient stability category prediction result corresponding to the second evaluation sequence and the transient stability category simulation result.

[0078] Step S313: When the confidence of the second evaluation sequence does not reach a preset confidence threshold, a plurality of running sample data after the current running simulation time window is expanded are updated in the second evaluation sequence.

[0079] When the confidence of the second evaluation sequence does not reach the preset confidence threshold, it means that the data in the current simulation time window is insufficient to make an accurate judgment. At this time, the model will request the simulator to continue running to expand the time window to obtain more data. As the simulation progresses, the new, longer time window data is passed to the next neural network, which performs more in-depth feature extraction and stability evaluation based on richer information.

[0080] For example, Figure 4As shown, the evaluator adopts a neural network cascade model, which includes multiple neural network models connected in series. The neural network model can adopt an RNN recurrent neural network. Due to the different updates and lengths of the simulation time window, it is for evaluating a single sample of the cascade neural network structure. When evaluating a sample, the simulator is used to simulate the sample, and the simulated data of 0~0.5s is given to the first neural network model CNN-1 under the evaluator as input; the simulation continues to 0~1s, and the simulated data of 0~1s is given to the first neural network model CNN-2 as input, ..., as the simulation time increases, the longer the simulation time window of the neural network receiving the input data, the more accurate the evaluation, until the confidence level of judging that the transient stability category is stable is 99.9, the result that the transient stability category is stable is accepted, and the data set of the transient stability category is updated.

[0081] In one example, in order to ensure the dynamic and real-time nature of the evaluation process, the information entropy index of the samples newly added to the evaluation queue will be continuously calculated. By continuously calculating the information entropy index of the newly added running sample data, the newly added samples that have a significant impact on the model optimization can be quickly identified, and the ranking of the second evaluation queue can be adjusted accordingly to ensure the efficient use of evaluation resources.

[0082] To this end, the steps in the second evaluation sequence are updated by expanding the multiple running sample data after the current running simulation time window, including steps S3131 to S3132. Among them:

[0083] Step S3131, obtaining multiple running sample data after expanding the current running simulation time window.

[0084] Step S3132: update the second evaluation sequence according to the obtained numerical comparison relationship of the information entropy indicators corresponding to the multiple running sample data after the current running simulation time window is expanded.

[0085] It can be understood that by dynamically adjusting the evaluation queue according to the addition of new samples and model updates in real time, the accuracy and efficiency of the power system transient stability assessment model based on deep learning are significantly improved, while the model's ability to quickly adapt to the time-varying operating environment of the power system is enhanced.

[0086] Step S314: input the updated second evaluation sequence into a second neural network model adjacent to the first neural network model for prediction, and obtain a transient stability category prediction result corresponding to the updated second evaluation sequence.

[0087] Step S315: Determine the confidence of the updated second evaluation sequence by using the transient stability category prediction result and the transient stability category simulation result corresponding to the updated second evaluation sequence.

[0088] Step S316: When the confidence of the updated second evaluation sequence does not reach the preset confidence threshold, continue to expand the multiple running sample data after the current running simulation time window to update the second evaluation sequence, and iterate in this way until the confidence of the second evaluation sequence reaches the preset confidence threshold or reaches the preset maximum running time window. The iteration stops and the neural network cascade model after the iteration stops is output.

[0089] It can be understood that the cascade mechanism of the neural network cascade model ensures that the model can dynamically adjust the simulation time window according to actual evaluation needs, which not only avoids unnecessary long simulation time, but also ensures the accuracy and reliability of the evaluation results.

[0090] Since the cascade architecture of the neural network cascade model is composed of multiple layers of neural networks, its output refers to the transient stability category of the output of the last layer of neural network. At the same time, the input data time window of the last layer of the cascade architecture is the longest and the received data is the richest, so the output result is more accurate than that of the upstream network. Based on this, the final output of the cascade architecture can be used to correct and optimize the upstream neural network.

[0091] To this end, in some embodiments, when the iteration of the neural network cascade model stops, the network parameters of the neural network model in the neural network cascade model are optimized using the input and output results of the last iteration of the neural network cascade model.

[0092] For example, Figure 5 As shown, the second evaluation sequence is recorded as "head end-task queue-tail end", and 1, 2, ..., i+1, i+2 in the second evaluation sequence represent different running sample data. The running sample data in the second evaluation sequence is simulated by a time domain simulator, and the second evaluation sequence is predicted by a neural network cascade model.

[0093] The area of ​​the horizontal dotted line in the time domain simulator represents the length of the time domain simulation window. For example, the first horizontal dotted line represents the simulation to 0.5s. At this point, the running sample data obtained from the simulation of 0~0.5s is used for preliminary evaluation of the first neural network model; the second and third horizontal dotted lines represent the simulation to 1s and 2s respectively. Similarly, the running sample data obtained from the simulation of 0~1s and 0~2s are used as input to the second and third neural network models for evaluation. When the confidence of a neural network model meets the requirements, the subsequent simulation is stopped, and there is no need to use the next level of neural network prediction.

[0094] For the i-th running sample data, the probability of the first neural network model CNN1- (input is 0~0.5s simulation data) being evaluated as stable is 69.9%, which does not meet the requirements, and the prediction results are not retained. Therefore, the simulation is continued, and a longer simulation time window (0~1s simulation data) is used to input the second neural network model CNN-2. The probability of being evaluated as stable is 85.2%, which still does not meet the requirements, and the prediction results are also not retained. Further simulation is performed, and a longer simulation time window (0~2s simulation data) is used to input the third neural network model CNN-3. At this time, the probability of being evaluated as stable is 99.9%, which meets the requirements, so this result is used as the final evaluation result. At this time, the transient stability category is stable and is fed back to CNN-1 and CNN-2, that is, the two neural network models CNN-1 and CNN-2 are notified that the labels corresponding to the 0~0.5s and 0~1s data they input are stable, which is equivalent to obtaining two sub-samples determined by the input time window short-label. When the sub-samples accumulate to a certain number, the network parameters of the two neural network models CNN-1 and CNN-2 can be updated.

[0095] In addition, the information entropy index proposed in the present invention is calculated by the first neural network CNN-1 in the cascade architecture. After simulating the i-th sample, CNN-1 is used to make a preliminary prediction of all running sample data. According to the three stable state probabilities obtained by the prediction, the information entropy index of each sample is calculated, and the subsequent samples to be evaluated are sorted according to the size of the index, and the task queue is updated. For example, the original sample evaluation order is i, i+1, i+2..., and after sorting by information entropy value, the new order becomes i, i+25, i+16..., thereby ensuring that the samples that have the most significant impact on model performance are always given priority.

[0096] Based on the same inventive concept, an embodiment of the present application further provides a priority strategy-based power system transient stability assessment device for implementing the priority strategy-based power system transient stability assessment method involved above.

[0097] The implementation solution to the problem provided by the device is similar to the implementation solution recorded in the above method. Therefore, the specific limitations in one or more embodiments of the power system transient stability assessment device based on the priority strategy provided below can be referred to the limitations on the power system transient stability assessment method based on the priority strategy above, and will not be repeated here.

[0098] like Figure 6 As shown, an embodiment of the present invention provides a power system transient stability assessment device based on a priority strategy, comprising:

[0099] A sequence forming module 100 is used to form a first evaluation sequence according to a plurality of operation sample data of the power system;

[0100] A sequence updating module 200 is used to update the first evaluation sequence according to the numerical comparison relationship of the information entropy indicators corresponding to the plurality of running sample data to form a second evaluation sequence;

[0101] The model evaluation module 300 is used to use a pre-trained neural network cascade model to predict the transient stability category of the second evaluation sequence, and to optimize the pre-trained neural network cascade model using the prediction results; wherein the pre-trained neural network cascade model is used to input running sample data and output the prediction results of the transient stability category corresponding to the running sample data.

[0102] In some embodiments, the device further includes an information entropy calculation module, which is used to calculate information entropy indicators corresponding to a plurality of running sample data, including:

[0103] For each running sample data, the running sample data is input into the pre-trained neural network model, and the predicted probability of each transient stability category corresponding to the running sample data is output;

[0104] The predicted probabilities of each transient stability category corresponding to the running sample data are used for weighted summation calculation to obtain the information entropy index corresponding to the running sample data.

[0105] In some embodiments, the sequence update module 200 is specifically used to compare the information entropy indicators corresponding to multiple operating sample data with the information entropy threshold, and screen out multiple operating sample data whose information entropy indicators are greater than the information entropy threshold as key operating sample data; arrange the multiple key operating sample data in descending order according to the numerical comparison relationship of the information entropy indicators to obtain a second evaluation sequence.

[0106] In some embodiments, the device also includes: a first network optimization module, which is used to recalculate the information entropy index corresponding to each running sample data based on the pre-trained neural network model after the network parameters of the pre-trained neural network model are optimized; and update the second evaluation sequence using the numerical comparison relationship of the information entropy index corresponding to the recalculated running sample data.

[0107] In some embodiments, the pre-trained neural network cascade model includes a plurality of neural network models connected in series;

[0108] The model evaluation module 300 is specifically used to input the second evaluation sequence into the first neural network model in the pre-trained neural network cascade model for prediction, and obtain the transient stability category prediction result corresponding to the second evaluation sequence; determine the confidence of the second evaluation sequence using the transient stability category prediction result corresponding to the second evaluation sequence and the transient stability category simulation result; when the confidence of the second evaluation sequence does not reach the preset confidence threshold, the multiple running sample data after the current running simulation time window is expanded are updated in the second evaluation sequence; the updated second evaluation sequence is input into the second neural network model adjacent to the first neural network model for prediction. Prediction is performed to obtain a transient stability category prediction result corresponding to the updated second evaluation sequence; the confidence of the updated second evaluation sequence is determined using the transient stability category prediction result corresponding to the updated second evaluation sequence and the transient stability category simulation result; when the confidence of the updated second evaluation sequence does not reach a preset confidence threshold, multiple running sample data after the current running simulation time window are continued to be expanded to update the second evaluation sequence, and so on and so forth, until the confidence of the second evaluation sequence reaches a preset confidence threshold or reaches a preset maximum running time window, the iteration stops, and the neural network cascade model after the iteration stops is output.

[0109] In some embodiments, multiple running sample data after expanding the current running simulation time window are updated in the second evaluation sequence, including: obtaining multiple running sample data after expanding the current running simulation time window; updating the second evaluation sequence according to the numerical comparison relationship of the information entropy indicators corresponding to the multiple running sample data after expanding the current running simulation time window.

[0110] In some embodiments, when the iteration of the neural network cascade model stops, the network parameters of the neural network model in the neural network cascade model are optimized using the input and output results of the last iteration of the neural network cascade model.

[0111] like Figure 7 As shown, an embodiment of the present invention further provides an electronic device, the electronic device 10 includes a memory 20 and a processor 30, the memory 20 stores a computer program, and when the computer program is executed by the processor 30, the processor 30 executes the steps of the power system transient stability assessment method based on the priority strategy in any of the above embodiments.

[0112] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed, the steps of the method for transient stability assessment of a power system based on a priority strategy as described in any of the above embodiments are implemented.

[0113] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, electronic device and computer storage medium can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0114] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0115] In several embodiments provided by the present invention, it is understood that each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and a part of a module, a program segment or a code includes one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved.

[0116] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, electronic devices, computer storage media and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0117] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0118] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0119] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for executing all or part of the steps of the method described in each embodiment of the present invention through a computer device (which can be a personal computer, server, or network device, etc.). The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (full name in English: Read-Only Memory, English abbreviation: ROM), random access memory (full name in English: Random Access Memory, English abbreviation: RAM), disk or optical disk and other media that can store program codes.

[0120] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for evaluating transient stability of a power system based on a priority strategy, characterized in that: include: forming a first evaluation sequence according to a plurality of operation sample data of the power system; The first evaluation sequence is updated according to the numerical comparison relationship of the information entropy indicators respectively corresponding to the plurality of the running sample data to form a second evaluation sequence; A pre-trained neural network cascade model is used to predict the transient stability category of the second evaluation sequence, and the pre-trained neural network cascade model is optimized using the prediction result; wherein the pre-trained neural network cascade model is used to input the operating sample data and output the prediction result of the transient stability category corresponding to the operating sample data.

2. The method for evaluating transient stability of a power system based on a priority strategy according to claim 1, characterized in that: Before the step of updating the first evaluation sequence according to the numerical comparison relationship of the information entropy indicators respectively corresponding to the plurality of the operating sample data to form the second evaluation sequence, the step further includes: calculating the information entropy indicators respectively corresponding to the plurality of the operating sample data; The step of calculating the information entropy indicators corresponding to the plurality of running sample data respectively comprises: For each of the operating sample data, the operating sample data is input into a pre-trained neural network model, and the predicted probability of each transient stability category corresponding to the operating sample data is output; The predicted probabilities of each transient stability category corresponding to the operating sample data are used to perform weighted sum calculation to obtain the information entropy index corresponding to the operating sample data.

3. The method for evaluating transient stability of a power system based on a priority strategy according to claim 1, characterized in that: The step of updating the first evaluation sequence according to the numerical comparison relationship of the information entropy indicators respectively corresponding to the plurality of the operating sample data to form the second evaluation sequence includes: comparing the information entropy indicators respectively corresponding to the plurality of the operating sample data with the information entropy threshold, and selecting the plurality of operating sample data whose information entropy indicators are greater than the information entropy threshold as key operating sample data; Arrange the plurality of key operating sample data in descending order according to the numerical comparison relationship of the information entropy index to obtain a second evaluation sequence.

4. The method for evaluating transient stability of a power system based on a priority strategy according to claim 2, characterized in that: The method further comprises: using a pre-trained neural network cascade model to predict the transient stability category of the second evaluation sequence, and optimizing the pre-trained neural network cascade model using the prediction result; wherein the pre-trained neural network cascade model is used to input the running sample data and output the prediction result of the transient stability category corresponding to the running sample data, and further comprises: After the network parameters of the pre-trained neural network model are optimized, the information entropy index corresponding to each of the running sample data is recalculated based on the pre-trained neural network model after the network parameters are optimized; The second evaluation sequence is updated using the recalculated numerical comparison relationship of the information entropy index corresponding to the running sample data.

5. The method for evaluating transient stability of a power system based on a priority strategy according to claim 1, characterized in that: The pre-trained neural network cascade model includes a plurality of neural network models connected in series; The step of using a pre-trained neural network cascade model to predict the transient stability category of the second evaluation sequence, and optimizing the pre-trained neural network cascade model using the prediction result, comprises: Inputting the second evaluation sequence into the first neural network model in the multiple pre-trained neural network cascade models for prediction, and obtaining a transient stability category prediction result corresponding to the second evaluation sequence; Determining the confidence of the second evaluation sequence by using the transient stability category prediction result and the transient stability category simulation result corresponding to the second evaluation sequence; When the confidence of the second evaluation sequence does not reach a preset confidence threshold, a plurality of running sample data after expanding the current running simulation time window are updated in the second evaluation sequence; Inputting the updated second evaluation sequence into a second neural network model adjacent to the first neural network model for prediction, to obtain a transient stability category prediction result corresponding to the updated second evaluation sequence; Determining the confidence of the updated second evaluation sequence using the transient stability category prediction result and the transient stability category simulation result corresponding to the updated second evaluation sequence; When the confidence of the updated second evaluation sequence does not reach the preset confidence threshold, the multiple running sample data after the current running simulation time window continue to be expanded to update the second evaluation sequence, and so on, until the confidence of the second evaluation sequence reaches the preset confidence threshold or reaches the preset maximum running time window, the iteration stops, and the neural network cascade model after the iteration stops is output.

6. The method for evaluating transient stability of a power system based on a priority strategy according to claim 5, characterized in that: The steps of updating the second evaluation sequence with multiple running sample data after expanding the current running simulation time window include: Obtain multiple running sample data after expanding the current running simulation time window; The second evaluation sequence is updated according to the obtained numerical comparison relationship of the information entropy indicators respectively corresponding to the multiple running sample data after the current running simulation time window is expanded.

7. The method for evaluating transient stability of a power system based on a priority strategy according to claim 5, characterized in that: When the iteration of the neural network cascade model stops, the network parameters of the neural network model in the neural network cascade model are optimized using the input and output results of the last iteration of the neural network cascade model.

8. A power system transient stability assessment device based on priority strategy, characterized in that: include: A sequence forming module, used for forming a first evaluation sequence according to a plurality of operation sample data of the power system; A sequence updating module, used for updating the first evaluation sequence according to the numerical comparison relationship of the information entropy indicators respectively corresponding to the plurality of the running sample data, to form a second evaluation sequence; A model evaluation module is used to use a pre-trained neural network cascade model to predict the transient stability category of the second evaluation sequence, and to optimize the pre-trained neural network cascade model using the prediction result; wherein the pre-trained neural network cascade model is used to input the operating sample data and output the prediction result of the transient stability category corresponding to the operating sample data.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the processor executes the steps of the power system transient stability assessment method based on the priority strategy as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the steps of the power system transient stability assessment method based on the priority strategy as described in any one of claims 1 to 7 are implemented.

Images