Power system secondary protection method, device and equipment and storage medium

Abstract

The invention provides an electric power system secondary protection method, device and equipment and a storage medium, and particularly relates to the technical field of electric power system secondary protection, and the method comprises the steps that a processor obtains sample state information of the t-th round of an electric power system; inputting the sample state information of the t-th round into the strategy network of the t-th round to obtain protection action execution information of the t-th round of the power system; performing reward calculation according to a protection action execution result of the t-th round to obtain an actual reward value of the t-th round; inputting the sample state information of the t-th round into the value network of the t-th round to obtain an expected accumulated reward value of the t-th round; calculating a first difference value of the t-th round according to the actual reward value of the t-th round and the expected accumulated reward value of the t-th round; and if the first difference value of the t-th round is within a preset range or t reaches a preset value, the processor determines the strategy network of the t-th round as the power system secondary protection model. The method improves the reliability and accuracy of the protection action executed by the power system.

Description

Technical Field

[0001] The present application relates to the technical field of secondary protection of power systems, and in particular to a method, device, equipment and storage medium for secondary protection of power systems. Background Art

[0002] As a key infrastructure of modern society, the safe and stable operation of power systems is of vital importance. Traditional power system secondary protection strategies mainly rely on preset rules and thresholds for fault detection and removal. However, with the increasing scale and complexity of power systems, this static protection method can no longer meet the requirements of modern power systems for flexibility and adaptability.

[0003] Traditional protection strategies often have problems with false operation and refusal to operate when facing complex and changeable power system faults. False operation refers to the protection device incorrectly issuing an action command when the power system is operating normally or without faults, resulting in unnecessary power outages and equipment damage; while refusal to operate refers to the protection device failing to operate in time when a power system fault occurs, causing the fault range to expand and may even cause serious power system accidents. In addition, traditional protection strategies also have limitations in accuracy and speed, and it is often difficult to find the best balance between rapid response and accurate judgment.

[0004] Therefore, traditional protection strategies often lead to false operation and failure of power system, which in turn leads to failure of power system operation, which is an urgent problem to be solved. Summary of the invention

[0005] In view of this, the purpose of the present application is to provide a secondary protection method, device, equipment and storage medium for an electric power system to solve the above-mentioned problems, avoid false operation and refusal to operate of the electric power system, and achieve faster, more accurate and more reliable fault elimination of the electric power system.

[0006] In a first aspect, an embodiment of the present application provides a secondary protection method for a power system, the method comprising: The processor obtains the sample state information of the tth round of the power system; wherein the sample state information of the tth round includes at least one of the following: a sample voltage amplitude, a sample current value, a sample phase angle, a sample system frequency, a sample state of a protection device, and a sample configuration parameter of the protection device; The processor inputs the sample state information of the tth round into the strategy network of the tth round to obtain the protection action execution information of the tth round of the power system; wherein the strategy network of the tth round is a mapping relationship between the reference sample state information of the tth round and the reference protection action execution information of the tth round; The processor calculates the reward according to the execution result of the protection action in the tth round to obtain the actual reward value in the tth round; The processor inputs the sample state information of the tth round into the value network of the tth round to obtain the expected cumulative reward value of the tth round; The processor calculates a first difference value of the tth round according to the actual reward value of the tth round and the expected cumulative reward value of the tth round; If the first difference value of the tth round is within a preset range or t reaches a preset value, the processor determines the strategic network of the tth round as a secondary protection model of the power system.

[0007] Preferably, the method further comprises: If the first difference value of the tth round is not within the preset range and t does not reach the preset value, the processor updates the policy network of the tth round according to the first difference value of the tth round to obtain the policy network of the t+1th round.

[0008] Preferably, the method further comprises: The processor acquires state information of the power system; wherein the state information includes at least one of the following: voltage amplitude, current value, phase angle, system frequency, state of a protection device, and configuration parameters of the protection device; If a fault occurs in the power system, the processor inputs the state information into the power system secondary protection model to obtain protection action execution information; The processor controls the power system to perform a protection action according to the protection action execution information.

[0009] Preferably, the method further comprises: The processor obtains the number of false operations according to the protection action execution information; The processor obtains the number of protection actions performed by the power system; The processor determines the ratio of the number of false operation times to the number of protection actions executed by the power system as the false operation rate of the power system.

[0010] Preferably, the actual reward value of the tth round is calculated according to the following formula:

[0011] in, is the actual reward value of the tth round, is the number of false trips, To show the severity of refusal, For accuracy matching, To protect the delay of the action, is the stability recovery degree, is the penalty weight for misoperation, To determine the penalty weight for refusal to move, Reward weights for accuracy, is the speed penalty weight, Reward weight for stability.

[0012] Preferably, the step of the processor acquiring the sample status information of the tth round of the power system includes: The processor obtains the original sample state information of the tth round of the power system; wherein the sample state information of the tth round includes at least one of the following: original sample voltage amplitude, original sample current value, original sample phase angle, original sample system frequency, original protection device sample state and original protection device sample configuration parameter; The processor performs standardization processing on the original sample state information of the tth round to obtain the sample state information of the tth round of the power system; The sample state information of the tth round of the power system is calculated according to the following formula:

[0013] in, is the sample state information of the tth round of the power system, is the original sample state information of the tth round of the power system, is the average value of the original sample state information of the power system in the tth round, is the standard deviation of the original sample state information of the power system in the tth round.

[0014] The power system secondary protection method provided by this application brings the following beneficial effects: The present application provides a method for secondary protection of an electric power system, which calculates the first difference of the tth round according to the actual reward value of the tth round and the expected cumulative reward value of the tth round; if the first difference of the tth round is within a preset range or t reaches a preset value, the strategy network of the tth round is determined as the secondary protection model of the electric power system. The method regards the protection process of the electric power system as a sequential decision problem, comprehensively considers the false operation rate, the rejection rate, the accuracy and speed of the protection action, and obtains the secondary protection model of the electric power system through interactive learning and updating of the strategy network and the value network, which effectively reduces the false operation rate and the rejection rate, and improves the reliability and accuracy of the protection actions performed by the electric power system, so that when a fault occurs, the protection device can respond and perform the protection action more quickly and accurately, effectively narrow the scope of the fault, and reduce the impact of the fault on the power grid.

[0015] In a second aspect, the present application further provides a secondary protection device for a power system, the device comprising: An acquisition module, used for acquiring the sample state information of the tth round of the power system; wherein the sample state information of the tth round includes at least one of the following: a sample voltage amplitude, a sample current value, a sample phase angle, a sample system frequency, a sample state of a protection device, and a sample configuration parameter of the protection device; A data processing module, used for inputting the sample state information of the tth round into the strategy network of the tth round to obtain the protection action execution information of the tth round of the power system; wherein the strategy network of the tth round is a mapping relationship between the reference sample state information of the tth round and the reference protection action execution information of the tth round; The data processing module is also used to calculate rewards according to the execution results of the protection actions in the tth round to obtain the actual reward value of the tth round; input the sample state information of the tth round into the value network of the tth round to obtain the expected cumulative reward value of the tth round; calculate the first difference of the tth round according to the actual reward value of the tth round and the expected cumulative reward value of the tth round; if the first difference of the tth round is within a preset range or t reaches a preset value, the strategy network of the tth round is determined as the secondary protection model of the power system.

[0016] The power system secondary protection device provided in the embodiment of the present application has the same technical features as the power system secondary protection method provided in the above-mentioned embodiment, so it can also solve the same technical problems and achieve the same technical effects.

[0017] In a third aspect, the present application provides a computing device, including a memory and a processor; One or more computer programs are stored in the memory, and the one or more computer programs include instructions; when the instructions are executed by the processor, the computing device executes the method as described in any one of the first aspects.

[0018] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium is used to store a computer program, and the computer program is used to execute the method as described in any one of the first aspects.

[0019] In a fifth aspect, the present application provides a computer program product, wherein the computer program product comprises one or more computer instructions, and when the computer instructions are executed by a computer, the computer executes the method as described in any one of the first aspects.

[0020] Other features and advantages of the present application will be described in the following description, and partly become apparent from the description, or understood by practicing the present application. The purpose and other advantages of the present application are realized and obtained by the structures particularly pointed out in the description and the drawings.

[0021] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the specific implementation methods of the present application or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0023] Figure 1 A schematic diagram of a secondary protection method for a power system provided in an embodiment of the present application; Figure 2 A schematic diagram of the structure of a secondary protection device for a power system provided in an embodiment of the present application; Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution of the present application will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application.

[0025] To facilitate understanding of this embodiment, the embodiments of the present application are described in detail below.

[0026] The present application embodiment provides a secondary protection method for a power system, such as Figure 1 As shown, Figure 1 A schematic diagram of a secondary protection method for a power system provided in an embodiment of the present application. The method comprises the following steps: S101, the processor obtains the tth round of sample status information of the power system.

[0027] The sample status information of the tth round includes at least one of the following: a sample voltage amplitude, a sample current value, a sample phase angle, a sample system frequency, a sample status of a protection device, and a sample configuration parameter of the protection device.

[0028] Specifically, the sample voltage amplitude, sample current value, sample phase angle, sample system frequency, sample state of the protection device (enabling or disabling the protection device, in test mode, and in standby mode), and sample configuration parameters of the protection device (action threshold and delay setting) are stored in the state space.

[0029] In addition, the processor is also provided with an experience replay buffer for storing a fixed number of state transition samples. The state transition samples include sample state information of each iteration cycle, the protection action performed by the power system corresponding to the sample state information, the actual reward value, the expected cumulative reward value, and information on whether to terminate the iteration. At the end of each iteration cycle, the new state transition samples are stored in the experience replay buffer, and the previously stored state transition samples are eliminated (deleted) to keep the memory size of the experience replay buffer unchanged. In order to improve the learning and updating efficiency of the subsequent policy network and value network, the state transition samples of one iteration cycle can be randomly extracted from the experience replay buffer for the subsequent learning and updating process of the policy network and value network.

[0030] More specifically, the processor obtains the original sample state information of the tth round of the power system; wherein the sample state information of the tth round includes at least one of the following: original sample voltage amplitude, original sample current value, original sample phase angle, original sample system frequency, sample state of original protection device and sample configuration parameter of original protection device; the original sample state information of the tth round is standardized to obtain the sample state information of the tth round of the power system; wherein the sample state information of the tth round of the power system is calculated according to the following formula:

[0031] in, is the sample state information of the power system in the tth round, is the original sample state information of the power system in the tth round, is the average value of the original sample state information of the power system in the tth round, is the standard deviation of the original sample state information of the power system in the tth round.

[0032] Since the original sample state information of the power system obtained by the sensors set up in the power system may be interfered by data or noise, it is necessary to use a sliding average filter algorithm to denoise the original sample state information, and fill in the missing values ​​in the original sample state information through a linear interpolation algorithm. In order to obtain the sample state information, the original sample state information also needs to be standardized to eliminate the dimensional difference, so that the power system secondary protection model can be trained through the sample state information in the future.

[0033] Such a setting improves the integrity and purity of the sample status information, standardizes the original sample status information, improves the accuracy and reliability of the sample status information, and thus improves the reliability of the secondary protection model of the power system obtained by subsequent training.

[0034] S102, the processor inputs the sample state information of the tth round into the strategy network of the tth round to obtain the protection action execution information of the tth round of the power system.

[0035] Among them, the policy network of the tth round is the mapping relationship between the reference sample state information of the tth round and the reference protection action execution information of the tth round.

[0036] Specifically, the above-mentioned strategy network adopts an MLP (Multilayer Perceptron, multilayer perceptron or feedforward neural network) structure or a CNN (Convolutional Neural Network, convolutional neural network) structure, and the specific selection depends on the representation form of the sample state information of the power system. If the sample state information is represented in the form of a vector, the MLP structure is selected; if the sample state information is represented in the form of a matrix or an image, the CNN structure is selected. The above-mentioned processor inputs the sample state information in the form of a vector or a matrix into the strategy network, so that the strategy network outputs the probability distribution of the protection action performed by the power system for the sample state information, that is, the protection action execution information.

[0037] With this setting, the processor can increase the nonlinear expression ability of the policy network through the ReLU (Rectified Linear Unit) activation function and use L2 regularization technology to prevent overfitting of the policy network.

[0038] With such a configuration, the processor can adjust the protection action execution information of the tth round through the sample status information of the tth round to optimize the protection effect on the power system.

[0039] S103, the processor calculates the reward according to the execution result of the protection action in the tth round to obtain the actual reward value of the tth round.

[0040] Specifically, the actual reward value of round t is calculated according to the following formula:

[0041] in, is the actual reward value of the tth round, is the number of false trips, To show the severity of refusal, For accuracy matching, To protect the delay of the action, is the stability recovery degree, is the penalty weight for misoperation, To determine the penalty weight for refusal to move, Reward weights for accuracy, is the speed penalty weight, Reward weight for stability.

[0042] Above is the false operation penalty item. When the protection device erroneously operates in the normal operation of the power system or without faults, the protection device is punished according to the false operation times of the power system protection action. Give negative rewards. The more, the greater the negative reward to curb unnecessary power outages and equipment damage.

[0043] Above This is a refusal penalty item. If the protection device fails to operate in time when a fault occurs in the power system, the severity of the refusal is evaluated based on the serious consequences that the fault may cause (such as equipment damage, large-scale power outages, etc.). , and give corresponding negative rewards. The higher it is, the larger the negative reward.

[0044] Above The accuracy reward item is given according to the matching degree between the protection action and the fault condition. The accuracy matching degree is determined by comparing the overlap degree (overlap ratio) between the protection action range and the fault range. , the higher the overlap, the higher the accuracy of the match. The higher it is, the more accurate the protection action is, and the greater the positive reward given.

[0045] Above The time delay of the protection action is determined by the time from the detection of the power system fault to the execution of the protection action. , according to the delay of protection action Give rewards or punishments. Protect the delay of actions The shorter the delay, the greater the reward; the delay of the protection action The longer it is, the greater the penalty.

[0046] Above The stability bonus item is the stability recovery of the power system after the protection action is executed, which is also an important indicator for evaluating the quality of the protection strategy. The stability recovery degree is evaluated by the recovery of the sample voltage amplitude and the sample system frequency. , and give corresponding rewards. Stability recovery The higher it is, the bigger the reward.

[0047] The above-mentioned penalty weight for misoperation , Refusal to act penalty weight , Accuracy reward weight , speed penalty weight , Stability reward weight , is set according to the priority and actual needs of the power system to execute protection actions, so as to ensure that the actual reward value in the tth round can accurately reflect the overall performance of the protection strategy.

[0048] With this setting, through continuous trial and error and feedback mechanism, the reinforcement learning algorithm gradually learns and finds the optimal protection strategy (strategy network) during the training process, so that in practical applications, the status of the power system can be monitored in real time, and when a fault occurs in the power system, a rapid response can be made and the optimal protection action can be executed, thereby ensuring the safe and stable operation of the power system.

[0049] S104, the processor inputs the sample state information of the tth round into the value network of the tth round to obtain the expected cumulative reward value of the tth round.

[0050] Specifically, the above-mentioned value network also adopts an MLP structure or a CNN structure. If the sample state information is represented in the form of a vector, the MLP structure is selected; if the sample state information is represented in the form of a matrix or an image, the CNN structure is selected. The above-mentioned processor inputs the sample state information in the form of a vector or a matrix into the value network so that the value network outputs an expected cumulative reward value; wherein the expected cumulative reward value is used to characterize the estimated value of the reward that can be obtained by executing the protection action in the state of the power system corresponding to the sample state information of the tth round, and is the output result of the value network when only the state of the power system corresponding to the sample state information of the tth round is considered. The processor can update the value network of the tth round according to the expected cumulative reward value of the tth round, obtain the value network of the t+1th round, and then obtain the expected cumulative reward value of the t+1th round through the value network of the t+1th round.

[0051] With this setting, the processor can increase the nonlinear expression ability of the value network through the ReLU activation function and use L2 regularization technology to prevent overfitting of the policy network.

[0052] S105: The processor calculates a first difference value of the tth round according to the actual reward value of the tth round and the expected cumulative reward value of the tth round.

[0053] Specifically, the first difference (advantage function) reflects the superiority of the protection action performed by the power system relative to the average level, and is a key signal for subsequent guiding strategy network updates.

[0054] S106, the processor determines whether the first difference value of the tth round is within a preset range or t reaches a preset value, if so, executes step S107, if not, executes S108.

[0055] Specifically, the processor continuously updates the weights of the strategy network and the value network to gradually optimize the protection strategy of the power system, and determines whether to stop updating the strategy network by judging the first difference or t (number of iterations).

[0056] S107: The processor determines the strategy network of the tth round as a secondary protection model of the power system.

[0057] Specifically, if the first difference is within the preset range, it means that the expected cumulative reward value of the tth round is closest to the actual reward value of the tth round. At this time, the update iteration of the strategy network of this round is stopped, and the strategy network of this round is used as the secondary protection model of the power system and applied to the power system; if t reaches the preset value, it means that the number of iterations of the strategy network of this round reaches the preset number. At this time, the update iteration of the strategy network of this round can also be stopped, and the strategy network of this round is used as the secondary protection model of the power system and applied to the power system.

[0058] S108: The processor updates the policy network of the tth round according to the first difference of the tth round to obtain the policy network of the t+1th round.

[0059] Specifically, if the first difference is not within the preset range and t does not reach the preset value, it means that there is still a difference between the expected cumulative reward value of the tth round and the actual reward value of the tth round, and the number of iterations of the policy network of this round has not reached the preset number. At this time, the policy network of this round continues to be updated to obtain the policy network of the next round (t+1th round).

[0060] The embodiment of the present application provides a method for secondary protection of a power system, which calculates the first difference of the tth round according to the actual reward value of the tth round and the expected cumulative reward value of the tth round; if the first difference of the tth round is within a preset range or t reaches a preset value, the strategy network of the tth round is determined as the secondary protection model of the power system. The method obtains the secondary protection model of the power system by collecting sample state information and training it using the PPO (Proximal Policy Optimization) algorithm. This enables the secondary protection model of the power system to be dynamically adjusted according to the state of the real-time power system, effectively responding to the increase in the scale and complexity of the electronic system, and improving the flexibility and adaptability of the secondary protection model of the power system. In the prior art, since the protection strategy of the electronic system is fixed, it often causes unnecessary power outages and equipment damage in the system due to false operation and refusal to operate problems, and even causes serious accidents.

[0061] Compared with the existing technology, this method regards the power system protection process as a sequential decision-making problem, comprehensively considers the false operation rate, refusal rate, accuracy and speed of protection action, optimizes the protection strategy by guiding the continuous trial and error and feedback mechanism of the PPO algorithm, and obtains the secondary protection model of the power system through interactive learning and updating of the strategy network and the value network, which effectively reduces the false operation rate and refusal rate, improves the reliability and accuracy of the protection actions performed by the power system, and enables the protection device to respond and execute protection actions more quickly and accurately when a fault occurs, effectively narrowing the scope of the fault and reducing the impact of the fault on the power grid.

[0062] In addition, after the training of the power system secondary protection model is completed, the power system secondary protection model needs to be tested. The specific testing process is as follows: (1) Construct a simulated power system environment that can simulate various operating states and fault conditions of a real power system. Use historical fault data or artificially synthesized fault scenarios as test inputs, and deploy the trained power system secondary protection model in the simulated environment for testing. By observing the model's response when a fault occurs and simulating the protection actions performed by the power system, the effectiveness and accuracy of the power system secondary protection model are preliminarily evaluated.

[0063] (2) Record the protection actions performed by the simulated power system and compare them with the actual optimal protection actions to obtain the comparison results. The actual optimal protection actions are obtained based on expert experience, historical data or simulation analysis. By comparing the results, the false action rate (the ratio of the number of times the power system secondary protection model mistakenly triggers the protection actions performed by the simulated power system to the number of all protection actions performed by the simulated power system), the rejection rate (the ratio of the number of times the power system secondary protection model fails to trigger the protection actions performed by the simulated power system when it should), the accuracy (the ability of the power system secondary protection model to correctly identify and respond to faults) and the speed (the time from the power system secondary protection model detecting the fault to the simulated power system performing the protection action) can be calculated.

[0064] (3) In order to further evaluate the performance of the power system secondary protection model, sensitivity analysis and robustness testing are required. Sensitivity analysis is to study the response of the power system secondary protection model to different simulated power system states, such as load changes, line parameter changes and fault types, such as short circuit and ground fault, in order to verify the adaptability and flexibility of the power system secondary protection model; robustness testing is to evaluate the performance of the power system secondary protection model under harsh conditions by introducing adverse factors such as noise interference and data deviation, so as to detect the stability and reliability of the power system secondary protection model.

[0065] (4) If the performance of the power system secondary protection model does not meet the requirements and standards of actual application, it needs to be tuned or retrained. Tuning includes adjusting the structure of the power system secondary protection model, optimizing hyperparameters, and increasing training data. Retraining the power system secondary protection model involves changing the training strategy, adopting more advanced algorithms, or introducing more feature information. Through continuous iteration and optimization, the performance of the power system secondary protection model meets the actual application requirements.

[0066] Taking the power system in region A as an example, we first collect historical fault data and power system parameters in region A, deploy the trained power system secondary protection model in the power system in region A, and test the power system secondary protection model using historical fault data. By comparing the protection actions performed by the simulated power system with the actual optimal protection actions, the comparison results show that the power system secondary protection model has false operation and refusal to operate under fault type B. Sensitivity analysis and robustness testing are carried out for the false operation and refusal to operate under fault type B, and it is found that the power system secondary protection model is highly sensitive to state changes and noise interference of the power system. Therefore, the power system secondary protection model is tuned (adjusting the structure of the power system secondary protection model and adding regularization terms). After multiple iterations and optimizations, the performance of the power system secondary protection model is significantly improved, the false operation rate and refusal rate are significantly reduced, and the accuracy and speed are improved.

[0067] In one embodiment, the processor obtains status information of the power system; wherein the status information includes at least one of the following: voltage amplitude, current value, phase angle, system frequency, status of the protection device and configuration parameters of the protection device; if a fault occurs in the power system, the status information is input into the secondary protection model of the power system to obtain protection action execution information; and the power system is controlled to execute protection action according to the protection action execution information.

[0068] Specifically, when the power system secondary protection model is formally applied in the power system, it is necessary to obtain the status information of the power system in real time. If the power system fails at this time, the power system secondary protection model will adjust the protection strategy for the power system in real time according to the status information of the power system obtained by the processor in real time, so that the power system can timely and accurately perform protection actions for the fault. For example, at a certain moment, the voltage amplitude of the power system suddenly decreases or the phase angle and system frequency deviate from the standard phase angle and standard system frequency. At this time, the power system secondary protection model will respond in time, generate information about the increase of the voltage amplitude or adjust the phase angle and system frequency at this time according to the standard phase angle and standard system frequency, and send it to the power system, so that the power system increases the voltage amplitude or adjusts the phase angle and system frequency according to the standard phase angle and standard system frequency.

[0069] In this method, the trained power system secondary protection model is deployed to the actual power system to realize real-time monitoring of the power system status information and rapid response to power system faults, which effectively improves the safety and stability of the power system, reduces the fault losses of the power system, and provides strong support for the intelligent protection of the power system.

[0070] In one embodiment, the processor obtains the number of false operations based on the protection action execution information; obtains the number of protection actions executed by the power system; and determines the ratio of the number of false operations to the number of protection actions executed by the power system as the false operation rate of the power system.

[0071] Specifically, in order to further verify the performance of the secondary protection model of the power system, it is necessary to conduct real-time monitoring of the power system where the secondary protection model of the power system is deployed, record the number of protection actions and the number of false operations when the power system fails, and then obtain the false operation rate of the power system to verify the resistance of the current system to false operations.

[0072] In this way, the stability and accuracy of the secondary protection model of the power system can be measured by the false operation rate of the power system, so that subsequent staff can optimize the secondary protection model of the power system.

[0073] Based on the above method embodiment, the present application embodiment also provides a secondary protection device for a power system, such as Figure 2 As shown, Figure 2 A schematic diagram of the structure of a secondary protection device for a power system provided in an embodiment of the present application. The device includes: an acquisition module 21 and a data processing module 22. The functions of each module are as follows: An acquisition module 21 is used to acquire the sample state information of the tth round of the power system; wherein the sample state information of the tth round includes at least one of the following: a sample voltage amplitude, a sample current value, a sample phase angle, a sample system frequency, a sample state of a protection device, and a sample configuration parameter of the protection device; A data processing module, used for inputting the sample state information of the tth round into the strategy network of the tth round to obtain the protection action execution information of the tth round of the power system; wherein the strategy network of the tth round is a mapping relationship between the reference sample state information of the tth round and the reference protection action execution information of the tth round; The data processing module 22 is also used to calculate rewards based on the execution results of the protection actions in the tth round to obtain the actual reward value of the tth round; input the sample status information of the tth round into the value network of the tth round to obtain the expected cumulative reward value of the tth round; calculate the first difference of the tth round based on the actual reward value of the tth round and the expected cumulative reward value of the tth round; if the first difference of the tth round is within a preset range or t reaches a preset value, the strategy network of the tth round is determined as the secondary protection model of the power system.

[0074] Preferably, the data processing module 22 is further configured to update the policy network of the tth round according to the first difference of the tth round to obtain the policy network of the t+1th round if the first difference of the tth round is not within the preset range and t does not reach the preset value.

[0075] Preferably, the acquisition module 22 is further used to acquire state information of the power system; wherein the state information includes at least one of the following: voltage amplitude, current value, phase angle, system frequency, state of the protection device and configuration parameters of the protection device; The data processing module 22 is also used to input the status information into the secondary protection model of the power system to obtain protection action execution information if a fault occurs in the power system; and control the power system to execute the protection action according to the protection action execution information.

[0076] Preferably, the data processing module 22 is further used to obtain the number of false operations according to the protection action execution information; The acquisition module 21 is further used to acquire the number of protection actions performed by the power system; The data processing module 22 is further configured to determine the ratio of the number of false operation times to the number of protection actions executed by the power system as the false operation rate of the power system.

[0077] Preferably, the actual reward value of the tth round is calculated according to the following formula:

[0078] in, is the actual reward value of the tth round, is the number of false trips, To show the severity of refusal, For accuracy matching, To protect the delay of the action, is the stability recovery degree, is the penalty weight for misoperation, To determine the penalty weight for refusal to move, Reward weights for accuracy, is the speed penalty weight, Reward weight for stability.

[0079] Preferably, the acquisition module 21 is specifically used to acquire the original sample state information of the tth round of the power system; wherein the sample state information of the tth round includes at least one of the following: original sample voltage amplitude, original sample current value, original sample phase angle, original sample system frequency, sample state of original protection device and sample configuration parameter of original protection device; perform standardization processing on the original sample state information of the tth round to obtain the sample state information of the tth round of the power system; The sample state information of the tth round of the power system is calculated according to the following formula:

[0080] in, is the sample state information of the power system in the tth round, is the original sample state information of the power system in the tth round, is the average value of the original sample state information of the power system in the tth round, is the standard deviation of the original sample state information of the power system in the tth round.

[0081] The power system secondary protection device provided in the embodiment of the present application has the same technical features as the power system secondary protection method provided in the above-mentioned embodiment, so it can also solve the same technical problems and achieve the same technical effects.

[0082] The present application also provides a computing device. Figure 3 As shown, this figure is a schematic diagram of a computing device provided in an embodiment of the present application, and the computing device 400 includes a bus 401, a processor 402, a communication interface 403 and a memory 404. The processor 402, the memory 404 and the communication interface 403 communicate with each other through the bus 401.

[0083] The bus 401 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0084] The processor 402 may be any one or more of a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).

[0085] The communication interface 403 is used for communicating with the outside. The memory 404 may include a volatile memory, such as a random access memory (RAM). The memory 404 may also include a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD) or a solid state drive (SSD).

[0086] The memory 404 stores executable codes, and the processor 402 executes the executable codes to perform the aforementioned power system secondary protection method.

[0087] The embodiment of the present application also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium that can be stored by a computing device or a data storage device such as a data center containing one or more available media. The available medium can be a magnetic medium (such as a floppy disk, a hard disk, a tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid-state hard disk). The computer-readable storage medium includes instructions that instruct the computing device to execute the above method.

[0088] The embodiment of the present application further provides a computer program product, which includes one or more computer instructions. When the computer instructions are loaded and executed on a computing device, the process or function described in the embodiment of the present application is generated in whole or in part.

[0089] The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer or data center to another website, computer or data center via wired (e.g., coaxial cable, optical fiber) or wireless (e.g., infrared, wireless, microwave, etc.) means.

[0090] When the computer program product is executed by a computer, the computer executes any of the aforementioned methods for secondary protection of a power system. The computer program product may be a software installation package, and when any of the aforementioned methods for secondary protection of a power system is required, the computer program product may be downloaded and executed on a computer.

[0091] The descriptions of the processes or structures corresponding to the above-mentioned figures have different emphases. For parts that are not described in detail in a certain process or structure, please refer to the relevant descriptions of other processes or structures.

[0092] The above description is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present application should be included in the protection scope of the present application.

Claims

1. A secondary protection method for a power system, characterized in that: The method comprises: The processor obtains the sample state information of the tth round of the power system; wherein the sample state information of the tth round includes at least one of the following: a sample voltage amplitude, a sample current value, a sample phase angle, a sample system frequency, a sample state of a protection device, and a sample configuration parameter of the protection device; The processor inputs the sample state information of the tth round into the strategy network of the tth round to obtain the protection action execution information of the tth round of the power system; wherein the strategy network of the tth round is a mapping relationship between the reference sample state information of the tth round and the reference protection action execution information of the tth round; The processor calculates the reward according to the execution result of the protection action in the tth round to obtain the actual reward value in the tth round; The processor inputs the sample state information of the tth round into the value network of the tth round to obtain the expected cumulative reward value of the tth round; The processor calculates a first difference value of the tth round according to the actual reward value of the tth round and the expected cumulative reward value of the tth round; If the first difference value of the tth round is within a preset range or t reaches a preset value, the processor determines the strategic network of the tth round as a secondary protection model of the power system.

2. The power system secondary protection method according to claim 1, characterized in that: The method further comprises: If the first difference value of the tth round is not within the preset range and t does not reach the preset value, the processor updates the policy network of the tth round according to the first difference value of the tth round to obtain the policy network of the t+1th round.

3. The power system secondary protection method according to claim 1, characterized in that: The method further comprises: The processor acquires state information of the power system; wherein the state information includes at least one of the following: voltage amplitude, current value, phase angle, system frequency, state of a protection device, and configuration parameters of the protection device; If a fault occurs in the power system, the processor inputs the state information into the power system secondary protection model to obtain protection action execution information; The processor controls the power system to perform a protection action according to the protection action execution information.

4. The power system secondary protection method according to claim 3, characterized in that: The method further comprises: The processor obtains the number of false operations according to the protection action execution information; The processor obtains the number of protection actions performed by the power system; The processor determines the ratio of the number of false operation times to the number of protection actions executed by the power system as the false operation rate of the power system.

5. The power system secondary protection method according to claim 1, characterized in that: The actual reward value of round t is calculated according to the following formula: in, is the actual reward value of the tth round, is the number of false trips, To show the severity of refusal, For accuracy matching, To protect the delay of the action, is the stability recovery degree, is the penalty weight for misoperation, To determine the penalty weight for refusal to move, Reward weights for accuracy, is the speed penalty weight, Reward weight for stability.

6. The power system secondary protection method according to claim 1, characterized in that: The step of the processor acquiring the sample state information of the tth round of the power system includes: The processor obtains the original sample state information of the tth round of the power system; wherein the sample state information of the tth round includes at least one of the following: original sample voltage amplitude, original sample current value, original sample phase angle, original sample system frequency, original protection device sample state and original protection device sample configuration parameter; The processor performs standardization processing on the original sample state information of the tth round to obtain the sample state information of the tth round of the power system; The sample state information of the tth round of the power system is calculated according to the following formula: in, is the sample state information of the tth round of the power system, is the original sample state information of the tth round of the power system, is the average value of the original sample state information of the power system in the tth round, is the standard deviation of the original sample state information of the tth round of the power system.

7. A secondary protection device for a power system, characterized in that: The device comprises: An acquisition module, used for acquiring the sample state information of the tth round of the power system; wherein the sample state information of the tth round includes at least one of the following: a sample voltage amplitude, a sample current value, a sample phase angle, a sample system frequency, a sample state of a protection device, and a sample configuration parameter of the protection device; A data processing module, used for inputting the sample state information of the tth round into the strategy network of the tth round to obtain the protection action execution information of the tth round of the power system; wherein the strategy network of the tth round is a mapping relationship between the reference sample state information of the tth round and the reference protection action execution information of the tth round; The data processing module is also used to calculate rewards according to the execution results of the protection actions in the tth round to obtain the actual reward value of the tth round; input the sample state information of the tth round into the value network of the tth round to obtain the expected cumulative reward value of the tth round; calculate the first difference of the tth round according to the actual reward value of the tth round and the expected cumulative reward value of the tth round; if the first difference of the tth round is within a preset range or t reaches a preset value, the strategy network of the tth round is determined as the secondary protection model of the power system.

8. A computing device, characterized in that including memory and processor; One or more computer programs are stored in the memory, and the one or more computer programs include instructions; when the instructions are executed by the processor, the computing device executes the method as claimed in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store a computer program, and the computer program is used to execute the method according to any one of claims 1 to 6.

10. A computer program product, characterized in that The computer program product comprises one or more computer instructions. When the computer instructions are executed by a computer, the computer performs the method according to any one of claims 1 to 6.

Images