Protection device on-site rapid test method based on improved whale algorithm
By improving the whale algorithm and cloud-optimizing the test sequence, the design and debugging problems in the on-site test of relay protection devices are solved, and a fast and accurate test process is achieved, ensuring the normal operation of the distribution system.
Patent Information
- Application Number
- CN202411033395.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-30
- Publication Date
- 2024-12-03
AI Technical Summary
In the field testing of relay protection devices, there are unreasonable design drawings, low professional level of construction personnel, and uncertain factors during on-site debugging, which leads to difficulties in power distribution construction, which may cause damage and failure of equipment operation, and lack of effectiveness. test method.
Using the fast on-site testing method of protection devices based on improved whale algorithm, the rapid testing architecture with optimized test sequence in the cloud is used to extract common features of multi-configuration protection tests, and the rapid testing model of protection devices is built, and the test sequence is optimized based on improved whale algorithm is simplified. Testing process, shorten the test time.
It realizes rapid and simplified on-site testing of relay protection devices, improves testing efficiency, reduces testing complexity, ensures the accuracy of test results and the safe operation of equipment.
Smart Images

Figure CN119064688A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of on-site testing of relay protection devices in power systems, and in particular relates to an on-site rapid testing method for protection devices based on an improved whale algorithm. Background Art
[0002] With the progress and development of science and technology in my country, the emergence of relay protection devices can effectively protect the normal operation of the distribution system. Therefore, power personnel must correctly set the relay protection devices and other related values to isolate faults in a timely and accurate manner and maintain the normal operation of the distribution system.
[0003] There are many problems that may occur during the on-site testing of relay protection devices, such as unreasonable design drawings, low professional level of construction personnel, and many uncertain factors during on-site commissioning. These problems may be very unfavorable to the development of power distribution construction, or may have a serious impact on the overall power distribution system, or may cause damage to the equipment during operation and cause equipment failure. Therefore, it is very important to propose a new method for testing protection devices.
[0004] At present, the research focus of protection device testing is on the on-site debugging method of microcomputer protection devices (Jiang Baorong, Pan Xilin. Research on on-site debugging method of microcomputer protection devices in 10kV substations [J]. China Equipment Engineering, 2019 (24): 112-114.); on the other hand, it is on the analysis of on-site debugging problems of relay protection in 10kV distribution systems (Fu Li. Analysis of common problems in on-site debugging of relay protection in 10kV distribution systems [J]. Modern Industrial Economy and Informatization, 2018, 8 (15): 111-112. DOI: 10.16525 / j.cnki.14-1362 / n.2018.15.48.). However, only the on-site testing of microcomputer protection devices or only the analysis of the problems of on-site testing of relay protection in distribution systems have not put forward effective suggestions. Summary of the invention
[0005] The purpose of the present invention is to provide a method for on-site rapid testing of protection devices based on an improved whale algorithm to address the problems existing in the prior art. The method is based on a cloud-based rapid testing architecture for protection devices that optimizes the test sequence, extracts common features of multi-configuration protection tests, and optimizes the protection test sequence based on the improved whale algorithm on the basis of model construction for rapid testing of protection devices.
[0006] The technical solution of the present invention is:
[0007] The on-site rapid testing method of the protection device based on the improved whale algorithm includes the following steps:
[0008] S1. Build a fast on-site testing architecture for protection devices based on cloud-based optimized test sequence, and test the protection devices to be tested;
[0009] S2. Extract common features of multi-configuration protection tests by comparing test conditions of different submodules;
[0010] S3. Complete the model construction of the rapid test of the protection device and improve the constraint conditions;
[0011] S4. Protection test sequence optimization based on improved whale algorithm.
[0012] Specifically, the on-site rapid test architecture in step S1 includes a cloud server, a test terminal, an on-site protection device to be tested, and related debugging wiring.
[0013] Specifically, the testing of the protection device to be detected in step S1 includes the following steps:
[0014] ① Establish the communication connection between the test terminal and the cloud server, and the voltage / current, switch position and communication debugging connection between the test terminal and the protection device to be tested;
[0015] ②Test input voltage / current electrical quantity time T 1 , Test terminal startup time T 2 , Test terminal stop time T 3 , Protect hard / soft pressure plate throw-in and throw-out time T 4 , Device reset time T 5 , Single test report generation time T 6 ;
[0016] ③ Send the protection device to be tested, the protection interval to be tested, and all the time information of step ② to the cloud server;
[0017] ④Cloud server based on the protection configuration to be detected and its action time TP i Analyze the common characteristics of protection tests, complete the division of protection test submodules, optimize the test sequence based on all the time information in step ②, and send the results to the test terminal;
[0018] ⑤The test terminal executes the test task and records the test results, generating a test report and synchronizing it to the cloud server.
[0019] Specifically, the common feature extraction of the multi-configuration protection test in step S2 includes the following steps:
[0020] (I): For different interval protection configurations of the protection device, the starting elements and action elements of different protections are divided into n test sub-modules according to their phases, and the sub-module test conditions M are recorded. i , as shown in formula (1):
[0021]
[0022] in, They are the voltage / current information applied for module testing, YB i For the hard or soft pressure plate information required for the test, W QF The circuit breaker position information required for the protection function test, Circuit breaker action information during protection function test;
[0023] (II): Compare the test conditions of different sub-modules and define the common test characteristics of sub-modules:
[0024] (III) Submodules with common test features are packaged into a test package, achieving dimensionality reduction from n test submodules to p test packages.
[0025] Specifically, the model construction of the protection device rapid test in step S3 includes the following steps:
[0026] Step 1: Construct a test model Min.T with the goal of minimizing the test time, as shown in formula (2):
[0027]
[0028] Among them, n i For all protection configuration test sequences for the interval to be detected, the corresponding T i The number of times the delay occurs, i = 1, 2, ..., 6;
[0029] Step 2: Construct constraints.
[0030] The constraints include the following:
[0031] ① Parallel test constraints of protection configuration, as shown in formula (3)
[0032] 1≤TMS P ≤C (3)
[0033] Among them, TMS P is the number of devices protected in parallel testing, C is the threshold, based on information extracted from the test terminal, 3-4 is recommended;
[0034] ② Circuit breaker operation times constraint, as shown in formula (4)
[0035]
[0036] Among them, TMS QF Configure the number of circuit breaker actions in the test sequence for all protections in the bay to be tested;
[0037] ③ Incremental application amount constraint, as shown in formula (5)
[0038] ΔU>0,ΔI>0 (5)
[0039] Among them, ΔU and ΔI are the incremental changes of the voltage and current applied during the test, respectively.
[0040] Specifically, the protection test sequence optimization based on the improved whale algorithm in step S4 includes the following steps:
[0041] Step 1: parse the protection configuration, take the number of test packets p as the population number, set the maximum number of iterations and encode p test packets;
[0042] Step 2: Initialize the population using a combination of global search and random search to obtain the initial solution T min ;
[0043] Step 3: Calculate the fitness value of the optimal individual, and perform mutation operation on the optimal order according to a certain probability. The mutation operation formula is as follows:
[0044]
[0045] in, is the current position vector of the humpback whale, is the position vector of the humpback whale after iterative update, and ε is 0.01. is a vector of random numbers in [0,1], is a vector of 0 and 1;
[0046] Step 4: Update individual position information, as shown in formula (7). By calculating the fitness value, determine whether the fitness requirement is met. When the maximum number of iterations is reached, output the optimal order. Otherwise, continue to update the information and return to step 3.
[0047]
[0048] in, is the coefficient vector, is the distance between other whales and the target prey;
[0049] Step 5: After the iteration is completed, the optimal sequence will be sent to the test terminal and the tests will be executed in sequence. The test results will be tracked and the test report will be generated synchronously. After the test is completed, the test report will be sent to the cloud server.
[0050] The beneficial effects of the present invention are as follows: a fast on-site testing architecture for protection devices based on cloud-based optimized test sequence is proposed, and common feature extraction of multi-configuration protection tests and model construction of fast testing of protection devices are performed, and common feature extraction of multi-configuration protection tests simplifies the on-site testing process of protection devices; protection test sequence optimization based on the improved whale algorithm shortens the on-site testing time of protection devices and makes testing fast. BRIEF DESCRIPTION OF THE DRAWINGS
[0051] Figure 1 The protection device test architecture diagram provided by the present invention
[0052] Figure 2 It is a structural schematic diagram of the on-site rapid testing process of the protection device provided by the present invention;
[0053] Figure 3 It is to improve the solution process of the whale algorithm;
[0054] Figure 4 It is the Gantt chart of the test case (SWOA);
[0055] Figure 5 It is the Gantt chart of the test case (WOA);
[0056] Figure 6 is the test case convergence curve. DETAILED DESCRIPTION
[0057] The technical solution of the present invention is described in detail below in conjunction with the accompanying drawings and specific implementation methods.
[0058] Example 1
[0059] like Figure 1 The present invention provides a novel method for rapid on-site testing of a protection device based on an improved whale algorithm, which specifically includes the following steps:
[0060] S1: On-site rapid testing architecture of protection devices based on cloud-based optimized test sequence. This testing architecture consists of a cloud server, a protection test terminal, on-site protection cabinets to be tested, and related debugging wiring, etc. Figure 1 As shown. The protection test terminal sends test commands and data through the debugging connection with the protection panel cabinet; it connects intelligently with the cloud server to receive the protection test sequence and send the test report after the test is completed. The cloud server is used to store the information of the substation and the tested object, receive the execution command of the protection test terminal and send the protection test sequence, and receive the test report after the test is completed.
[0061] The test process of the protection cabinet for the interval includes ① establishing the communication connection between the test terminal and the cloud server, testing the voltage / current between the test terminal and the protection cabinet, the switch position of the circuit breaker, and the debugging connection of the communication; ② testing the input voltage / current electrical quantity time T 1 , Test terminal startup time T 2 , Test terminal stop time T 3 , Protect hard / soft pressure plate throw-in and throw-out time T 4 , Device reset time T 5 , Single test report generation time T 6 ; ③ The maintenance personnel send the maintenance substation, the protection interval to be detected, and all the time information of step ② to the cloud server; ④ The cloud server is based on the configuration of the protection to be detected and its action time TP i Analyze the common characteristics of protection testing, complete the division of protection testing sub-modules, and optimize the test sequence based on all the time information in step ②, and send its results to the test terminal; ⑤ The test terminal executes the test task and records the test results, and generates a test report that is synchronized to the cloud server.
[0062] S2: Common feature extraction for multi-configuration protection testing.
[0063] For different interval protection configurations of protection devices, the starting elements and action elements of different protections are divided into n test sub-modules according to phase, such as longitudinal differential protection, ground distance protection, phase distance protection, zero-sequence overcurrent protection, etc., which are divided into single sub-modules according to phase. For n test sub-modules, use the test terminal to test each sub-module and record the test conditions M of each sub-module i , as shown in formula (1):
[0064]
[0065] in, They are the voltage / current information applied for module testing, YB i For the hard or soft pressure plate information required for the test, W QF The circuit breaker position information required for the protection function test, It is the circuit breaker action information during the protection function test.
[0066] Compare the test conditions of different sub-modules and define the common test characteristics of sub-modules: ① Application amount Same or nearly the same; ② Submodule M i With M j Test switching time interval Δt<T s1 ,in The same amount of application in feature ① mainly reflects that the debugging wiring does not need to be changed, and the test terminal does not need to switch the test items. Under the condition of the same debugging wiring, sub-modules with similar test conditions can be tested at the same time. The test switching time in feature ② mainly reflects that the test conditions are similar. Even if the test items need to be switched, the time interval for adjusting the test terminal is still very short. Therefore, this feature can improve the test speed of the sub-module.
[0067] After the test terminal completes the test on all test sub-modules, it packages the sub-modules with common test features into a test package, thereby achieving dimensionality reduction from n test sub-modules to p test packages, thereby reducing the complexity of building a rapid test model for the protection device.
[0068] S3: Model construction for rapid testing of protective devices.
[0069] The mathematical model of the protection device rapid test can be described as: P test packages {P 1 , P 2 , …P n}Perform tests on the object being tested. Each test package P i Contains i The goal of the protection device quick test is to test each protection device so that the test completion time is the shortest.
[0070] Considering the identifiability of action information during the protection test and the limit on the number of device actions during the test operation, the protection device rapid test should meet the following constraints:
[0071] (1) The parallel test information of each protection device should not be too much;
[0072] (2) The number of circuit breaker operations in the protection configuration test sequence should not be too many;
[0073] (3) The voltage and current applied during the protection test must be in an increasing relationship.
[0074] Considering the situation of on-site testing of protection devices, the shortest completion time of on-site testing of the tested object is used as the evaluation index. The objective function can be expressed by a mathematical model:
[0075]
[0076] In order to ensure the identifiability of action information during the protection test and the limit on the number of device actions during the test operation, the following constraints are constructed.
[0077] (1) Parallel test constraints of protection configuration, as shown in formula (3)
[0078] 1≤TMS P ≤C (3)
[0079] Among them, TMS P is the number of devices protected in parallel testing, C is the threshold, and 3-4 is recommended based on information extracted from the test terminal.
[0080] (2) Circuit breaker operation times constraint, as shown in formula (4)
[0081]
[0082] Among them, TMS QF The number of circuit breaker operations in the test sequence for all protection configurations for the bay to be tested.
[0083] (3) Incremental application amount constraint, as shown in formula (5)
[0084] ΔU>0,ΔI>0 (5)
[0085] Among them, ΔU and ΔI are the incremental changes of the voltage and current applied during the test, respectively.
[0086] S4: Protection test sequence optimization based on improved whale algorithm.
[0087] Basic Whale Algorithm: The Whale Algorithm (WOA) imitates the hunting process of humpback whales, including the behavior of spiraling up and spitting out bubble nets to attack prey. WOA first gives a set of random solutions according to the problem, and then continuously optimizes and updates according to certain rules until the optimal solution is found. It can be divided into three processes: random search, bubble net attack, and spiral rise.
[0088] ① Prey encirclement process: The first step of the whale algorithm is to encircle the prey. At this time, the whales are surrounding the prey. Assuming that the position of the captured prey is the position of the optimal solution candidate, the closer the position is to the prey, the more likely it is a local optimal solution. Other whales update their positions based on the position of the target prey, that is, the position of the optimal candidate solution. The specific mathematical model is as follows:
[0089]
[0090]
[0091]
[0092]
[0093]
[0094] in, is the distance between other whales and the target prey, t is the current iteration number, is the coefficient vector, is the position vector of the current optimal solution, is the current position vector of the humpback whale, is the position vector of the humpback whale after iterative update. is a random vector between (0, 1). max is the maximum number of iterations, and the value of a decreases from 2 to 0 as the number of iterations increases.
[0095] ② Bubble attack process: When attacking prey, whales will first spit out bubbles to surround the prey in the bubbles, and then approach the food by spiraling upward and shrinking the encirclement at the same time.
[0096] (1) Narrowing the encirclement
[0097] In the mathematical model, approaching the prey is done by reducing the value of a, so that The fluctuation range of also decreases as a decreases.
[0098] (2) Spiral Ascent
[0099] By spiraling upward, whales can capture prey to a greater extent and enhance the local search ability of the algorithm. The specific mathematical model is as follows:
[0100]
[0101]
[0102] In order to achieve the synchronous update of the position of the spiral rise and the shrinking encirclement, the algorithm introduces the probability r to determine the swimming mode of the whale.
[0103]
[0104] ③ Random search
[0105] In order to ensure the global search capability of WOA, When , other whale individuals can be randomly selected as the best search agent, and the whale updates its position information towards the randomly selected whale.
[0106]
[0107]
[0108] In the algorithm's search process, parameter A has the following effects: When , WOA adopts a random search method, away from the reference humpback whale, and randomly selects humpback whales as the best proxy within the feasible domain to expand the global search range. When the whale's bubble net attack strategy is to randomly choose to spiral up or shrink and surround with a probability of one in two to update the individual's position.
[0109] Improved whale algorithm:
[0110] ①Coding rules
[0111] Since the whale algorithm is generally used for continuous optimization problems, the on-site rapid test of protection devices belongs to a discrete combinatorial optimization problem, and the whale algorithm needs to be discretized in an appropriate way to solve the problem. The coding rule adopts a two-stage coding rule based on random keys to sort the test packages and tests respectively, based on the sorting of test packages and tests.
[0112] (1) Test package sorting
[0113] The integer codes of this part are {P 11 , P 12 , P 21 , P 22 , …P nj}, P ij Represents the number of tests for each test package.
[0114] (2) Test sorting
[0115] The integer at each position represents the test package number of this test package. The number of times this integer appears indicates the test number of this test package. This arrangement effectively clarifies the order of tests and the priority between different tests. For the h-th appearance of test package number P, it indicates the h-th test of test package P, and the total number of tests of all test packages after all the tests are arranged is equal to the total number of tests of this test package. This encoding is more flexible and can adapt to the situation where the number of tests of a test package is uncertain.
[0116] ② Population initialization
[0117] After the encoding and decoding are completed, the population is initialized. In order to overcome the problem that the quality of the randomly generated initial solution may not be high and it takes a lot of time to iterate to generate the optimal solution, the population is initialized by combining global search and local search in a certain proportion. For example, there are 60% global search, 30% local search, and 10% random initial solution in the population, which can effectively improve the quality of the initial solution.
[0118] ③Add mutation operator
[0119] Since high-quality genes have good adaptability during the evolution process, they will be preserved for evolution, which makes the whale algorithm prone to fall into the local optimal solution in the later stage of evolution. In order to overcome the problem that the whale algorithm is easily trapped in the local optimal solution, the mutation operator is introduced and improved on the basis of the traditional whale algorithm. Here, the mutation operation in the genetic algorithm is introduced to the optimal individual. The basic method is to perform a mutation operation on each dimensional element in the optimal individual with a certain probability. Here, the mutation probability is set to, and the mutation operation formula is as follows:
[0120]
[0121] Among them, ε is taken as 0.01, is a vector of random numbers in [0,1], is a vector of 0s and 1s.
[0122] ④SWOA algorithm steps, flow chart as follows Figure 3 As shown,
[0123] 1) Parse the protection configuration, take the number of test packets p as the population number, set the maximum number of iterations and encode p test packets.
[0124] 2) Use the combined method of global search and random search to initialize the population and improve the quality of the initial solution, thereby obtaining the initial solution T min .
[0125] 3) Calculate the fitness value of the optimal individual and perform mutation operations on the optimal sequence with a certain probability.
[0126] 4) Update individual position information and determine whether the fitness requirements are met by calculating the fitness value. When the maximum number of iterations is reached, output the optimal order. Otherwise, continue to update the information and return to step three.
[0127] 5) After the iteration is completed, the optimal sequence will be sent to the test terminal and the tests will be executed in sequence. The test results will be tracked and the test report will be generated synchronously. After the test is completed, the test report will be sent to the cloud server.
[0128] Step 5: The protection device on-site rapid test process is as follows Figure 2 shown.
[0129] Example 2
[0130] This implementation verifies the field testing method for the protection device proposed in Example 1 of the present invention. The WOA algorithm is discretized to solve the problem of rapid field testing, and a mutation operator is added. The improved whale algorithm is abbreviated as SWOA. Figure 4 , Figure 5 as well as Figure 6 Taking the protection device PCS-955-G as an example, the optimal solution T is given.min Gantt chart and convergence curve.
[0131] Figure 4 The Gantt chart of the PCS-955-G protection device test using the improved whale algorithm is shown. Figure 5 The whale algorithm Gantt chart is used for the protection device PCS-955-G test. The vertical axis of the Gantt chart is the applied amount, the horizontal axis is the test time t / min, and the protection configuration types are longitudinal differential protection, zero-sequence differential protection, ground distance protection, zero-sequence overcurrent protection, and phase distance protection. The protection codes corresponding to each protection configuration type submodule are shown in Table 1. O / R represents protection action and return respectively. For example, longitudinal differential protection B phase: 87L-Z A -O / R.
[0132] Table 1 Protection code table
[0133]
[0134] exist Figure 4 and Figure 5 It can be seen that the ground distance protection submodule and the zero-sequence overcurrent protection submodule partially overlap, which means that the debugging wiring does not need to be changed during the on-site debugging of this function, and the test terminal does not need to switch test items. Under the same debugging wiring, submodules with similar test conditions can be tested at the same time, thereby achieving the purpose of on-site rapid testing. Figure 4 and Figure 5 It can be seen that the improved whale algorithm has better results than the whale algorithm. This reflects that the whale algorithm is prone to fall into the local optimal solution in the later stage of evolution, and only the local optimal order is sent. Compared with the whale algorithm, the improved whale algorithm has better global search ability due to the addition of mutation operators.
[0135] Figure 6 The figure shows the convergence curve of the protection device PCS-943-G test. The corresponding relationships in the figure are: SWOA-improved whale algorithm; WOA-whale algorithm; PSO-particle swarm algorithm; ABC-artificial bee colony algorithm. Figure 6 The convergence results of SWOA and WOA and the test completion time are Figure 4 and Figure 5 The Gantt chart in the figure corresponds to that in the figure, and it can be seen that SWOA has a relatively fast convergence speed and the shortest test completion time compared to the other three algorithms.
[0136] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or some technical features can be replaced by equivalents without departing from the spirit of the technical solution of the present invention, which should be included in the scope of the technical solution for protection of the present invention.
Claims
1. A rapid on-site testing method for protection devices based on an improved whale algorithm, characterized in that: The steps include: S1. Build a fast on-site testing architecture for protection devices based on cloud-based optimized test sequence, and test the protection devices to be tested; S2. Extract common features of multi-configuration protection tests by comparing test conditions of different submodules; S3. Complete the model construction of the rapid test of the protection device and improve the constraint conditions; S4. Protection test sequence optimization based on improved whale algorithm.
2. The on-site rapid testing method for a protection device based on the improved whale algorithm according to claim 1 is characterized in that: The on-site rapid test architecture in step S1 includes a cloud server, a test terminal, an on-site protection device to be tested, and related debugging wiring.
3. The on-site rapid testing method for protection devices based on the improved whale algorithm according to claim 2 is characterized in that: The step S1 of testing the protection device to be detected comprises the following steps: ① Establish the communication connection between the test terminal and the cloud server, and the voltage / current, switch position and communication debugging connection between the test terminal and the protection device to be tested; ② Test input voltage / current electrical quantity time T1, test terminal start time T2, test terminal stop time T3, protection hard / soft pressure plate throw-in / out time T4, device reset time T5, single test report generation time T6; ③ Send the protection device to be tested, the protection interval to be tested, and all the time information of step ② to the cloud server; ④Cloud server based on the protection configuration to be detected and its action time TP i Analyze the common characteristics of protection tests, complete the division of protection test submodules, optimize the test sequence based on all the time information in step ②, and send the results to the test terminal; ⑤The test terminal executes the test task and records the test results, generating a test report and synchronizing it to the cloud server.
4. The on-site rapid testing method for protection devices based on the improved whale algorithm according to claim 1 is characterized in that: The common feature extraction of the multi-configuration protection test in step S2 includes the following steps: (I): For different interval protection configurations of the protection device, the starting elements and action elements of different protections are divided into n test sub-modules according to their phases, and the sub-module test conditions M are recorded. i , as shown in formula (1): in, They are the voltage / current information applied for module testing, YB i For the information of hard or soft pressure plate required for testing, W QF The circuit breaker position information required for the protection function test, Circuit breaker action information during protection function test; (II): Compare the test conditions of different sub-modules and define the common test characteristics of sub-modules: (III) Submodules with common test features are packaged into a test package, achieving dimensionality reduction from n test submodules to p test packages.
5. The on-site rapid testing method for protection devices based on the improved whale algorithm according to claim 1 is characterized in that: The model construction of the protection device rapid test in step S3 includes the following steps: Step 1: Construct a test model Min.T with the goal of minimizing the test time, as shown in formula (2): Among them, n i For all protection configuration test sequences for the interval to be detected, the corresponding T i The number of times the delay occurs, i = 1, 2, ..., 6; Step 2: Construct constraints.
6. The on-site rapid testing method for protection devices based on the improved whale algorithm according to claim 5 is characterized in that: The constraints include the following: ① Parallel test constraints of protection configuration, as shown in formula (3) 1≤TMS P ≤C (3) Among them, TMS P is the number of devices protected in parallel testing, C is the threshold, based on information extracted from the test terminal, 3-4 is recommended; ② Circuit breaker operation times constraint, as shown in formula (4) Among them, TMS QF Configure the number of circuit breaker actions in the test sequence for all protections in the bay to be tested; ③ Incremental application amount constraint, as shown in formula (5) ΔU>0,ΔI>0 (5) Among them, ΔU and ΔI are the incremental changes of the voltage and current applied during the test, respectively.
7. The on-site rapid testing method for protection devices based on the improved whale algorithm according to claim 1 is characterized in that: The protection test sequence optimization based on the improved whale algorithm in step S4 includes the following steps: Step 1: parse the protection configuration, take the number of test packets p as the population number, set the maximum number of iterations and encode p test packets; Step 2: Initialize the population using a combination of global search and random search to obtain the initial solution T min ; Step 3: Calculate the fitness value of the optimal individual, and perform mutation operation on the optimal order according to a certain probability. The mutation operation formula is as follows: in, is the current position vector of the humpback whale, is the position vector of the humpback whale after iterative update, and ε is 0.
01. is a vector of random numbers in [0,1], is a vector of 0 and 1; Step 4: Update individual position information, as shown in formula (7). By calculating the fitness value, determine whether the fitness requirement is met. When the maximum number of iterations is reached, output the optimal order. Otherwise, continue to update the information and return to step 3. in, is the coefficient vector, is the distance between other whales and the target prey; Step 5: After the iteration is completed, the optimal sequence will be sent to the test terminal and the tests will be executed in sequence. The test results will be tracked and the test report will be generated synchronously. After the test is completed, the test report will be sent to the cloud server.
Citation Information
Cited By
Intelligent substation relay protection device detection method and system, and medium
CN121831302A
Relay protection automatic test sequence optimization method and system based on dynamic priority
CN122264358A
