Power distribution network inverse time limit overcurrent constant value setting method based on improved mayfly naiad algorithm

Abstract

A power distribution network inverse time limit overcurrent constant value setting method based on an improved mayfly naiad algorithm comprises the following steps: S1, taking action time and protection coordination degree of main protection and backup protection of a power distribution network as objective functions, and establishing a protection optimization setting mathematical model in combination with constraint conditions considering protection selectivity and sensitivity requirements; s2, on the basis of the mayfly naiad algorithm, generating a super-uniformly distributed initial mayfly naiad population by using a Halton sequence, and introducing a variable step size strategy rationalization search mechanism and adding random disturbance to obtain an improved mayfly naiad algorithm; and S3, solving the protection optimized and set mathematical model by using an improved mayfly naiad algorithm to obtain a coordination strategy between the main protection and the backup protection. The defect that protection is easily influenced by fault types is overcome, the setting work of the distribution line can be automatically realized, and the efficiency is greatly improved.

Description

Technical Field

[0001] The invention relates to the technical field of electric power systems, and in particular to a method for setting relay protection values ​​when emergency power supply is restored after a fault. Technical Background

[0002] After the distributed generation DG (Distributed Generation) is connected to the distribution network, it has a significant impact on the operation mode and topology of the distribution network. The traditional three-stage current protection method can no longer adapt to this change, so the protection method needs to be adjusted to ensure the reliable operation of the DG distribution network. In this regard, the setting optimization method of inverse time overcurrent protection provides reliable protection for the distribution network containing DG, thereby ensuring its normal operation.

[0003] Compared with the traditional three-stage current protection, the inverse time overcurrent protection has a shorter action time and higher reliability, and has been widely used in distribution networks. It is usually solved by simple linear programming, genetic algorithm, improved particle swarm algorithm and other methods. However, the traditional optimization algorithm still has some defects for complex fault scenarios with many constraints. In order to avoid the defects of the traditional optimization algorithm, some scholars later used intelligent algorithms to optimize the inverse time overcurrent protection, but the algorithm itself has poor robustness and is prone to fall into local optimality. In addition, this method linearizes the nonlinear problems in the constraints and cannot guarantee the reliability of the protection. Moreover, most of the research on distribution networks containing DG focuses on communication-based protection. Compared with traditional protection, communication-based protection has certain advantages, but it has too high reliability requirements for the communication system. Once the communication collapses, the protection will fail.

[0004] Based on the above, there are few universal protection setting methods for different fault types of distribution networks without DG, with a single type of DG, or with various types of DG. How to adjust the protection setting in real time according to changes in the distribution network operating status and fault type is a major difficulty. Summary of the invention

[0005] In view of the above problems, the present invention provides a method for setting a distribution network inverse time overcurrent setting value based on an improved mayfly algorithm.

[0006] In order to achieve the above object, the technical solution of the present invention is: a method for setting the inverse time overcurrent setting value of a distribution network based on an improved mayfly algorithm, comprising the following steps:

[0007] S1. Taking the action time and protection coordination of the main protection and backup protection of the distribution network as the objective function, combined with the constraints of protection selectivity and sensitivity requirements, a mathematical model for protection optimization setting is established;

[0008] S2. Based on the mayfly algorithm, the Halton sequence is used to generate an initial mayfly population with super uniform distribution, and a variable step size strategy is introduced to rationalize the search mechanism and add random disturbances to obtain an improved mayfly algorithm.

[0009] S3. The improved mayfly algorithm is used to solve the mathematical model of protection optimization setting to obtain the coordination strategy between the main protection and the backup protection.

[0010] In step S1, the objective function is:

[0011]

[0012] In the formula: A is the number of main protection devices; C is the number of backup protection devices; L is the total number of fault lines; When a short circuit fault occurs on the current line j, protection i acts as the main protection action time; is the action time of protection k as backup protection when short circuit occurs in the current line j; R CO For the purpose of protection coordination, if the main protection device fails to operate in time, the backup protection device should be able to intervene quickly in a short time to avoid false operation and refusal to operate; λ1, λ2, and λ3 are weight coefficient values.

[0013] In step S1, the constraints of protection selectivity and sensitivity are considered, specifically:

[0014] In order to ensure the selectivity requirements between the main protection and the backup protection, the time difference between the main and backup protection meets the following requirements:

[0015]

[0016] Where: ΔT is the minimum value of the time difference between the main protection and the backup protection. In order to ensure the reliability of the protection, ΔT cannot be too small;

[0017] In order to ensure that the relay can operate without error at the maximum starting load current and can operate accurately at the minimum fault current, the relay should have a certain sensitivity. Its sensitivity S j The specific constraints are:

[0018]

[0019] Where: I op is the starting current of the relay; I l·min is the minimum starting load current of the line; I l·max is the maximum starting load current;

[0020] The relay action time limit is:

[0021] T i,min ≤T≤Ti,max

[0022] Where: T i,min is the lower limit of the relay action time; T i,max is the upper limit of the relay action time; T is the relay action time value;

[0023] The range of relay time setting coefficient is: TDS i,min ≤TDS≤TDS i,max

[0024] Where: TDS is the relay time setting coefficient; TDS i,min It is the lower limit of the relay time setting coefficient; TDS i,max It is the upper limit value of the relay time setting coefficient;

[0025] The protection device should meet the selectivity requirement, that is, when a fault occurs, it can accurately select the fault section and avoid false operation and refusal to operate. The selectivity can be expressed by constraining the false operation rate and refusal to operate rate:

[0026] P fault (I)≥η1(false operation rate constraint)

[0027] P reject (I)≤η2(rejection rate constraint)

[0028] Where: P fault (I) is the probability of malfunction of the protection device; P reject (I) is the probability of the protection device failing to operate; η1 and η2 are the upper limits of the allowable false operation rate and failure rate.

[0029] In step S3, the mayfly algorithm is specifically as follows:

[0030] S31 Initialization

[0031] Initialize the positions and velocities of male and female mayfly populations,

[0032] Randomly generate n m Male mayflies f Position x of female mayflies m and x f ;

[0033] The search space of mayfly i is expressed as: i ~U(lb,ub)

[0034] Where: lb, ub are the lower and upper limits of the search space respectively;

[0035] Initialize the speed v of male and female mayflies m and v f , usually initialized to zero or a small random value;

[0036] S32. Activities of male mayflies

[0037] Assumptions is the position space of male mayfly m at the tth iteration, by moving to the current position Add Speed Arriving at the new position, that is:

[0038]

[0039] Where: z1 and z2 are the weight coefficients of individual learning and social learning respectively; pbest mj is the optimal position of m mayflies in dimension j; gbest j is the optimal position of the entire mayfly population in dimension j; c1 and c2 are random numbers in the interval [0,1]; ω1 is the inertia weight of male mayflies,

[0040] In order to enhance the population diversity of the mayfly algorithm, an additional local search is performed on the male mayfly with the best fitness through a weighted neighborhood exploration strategy. The formula is as follows:

[0041]

[0042] Where: The current position of the best male mayfly; δ search radius, usually a small positive number, dynamic adjustment can enhance the search ability; is a uniformly distributed random number in the interval [-1,1]; is the position of a randomly selected individual in the neighborhood; η is the neighborhood learning factor, which controls the influence weight of the optimal individual and the random neighborhood individual;

[0043] S33. Activities of female mayflies

[0044] Female mayflies are ranked according to their fitness function. The best females will be attracted to the best males, and the second-best females will be attracted to the second-best males. Assume is the position space of the female mayfly f at the tth iteration, by transforming the velocity Add to the current position to change the position, that is:

[0045]

[0046]

[0047] Where: ω2 is the inertia weight of the female mayfly; ∈ is the random flight coefficient; r is a random value in the range of [-1,1]; N(0,1) is a standard normal distribution random variable; a1 is the female learning factor; c3 is a random number in the interval [0,1]; is the speed of female mayfly f at the tth iteration; is the position space of male mayfly m at the tth iteration; is the position space of the female mayfly f at the tth iteration; is the speed of the female mayfly f after the t+1th iteration; is the objective function value of the strategy corresponding to the position of male mayfly m after t iterations; is the objective function value of the strategy corresponding to the position of the female mayfly f after t iterations;

[0048] S34, Mayfly speed limit

[0049]

[0050] Where: V max v is the maximum speed limited by the mayfly; i represents the set of speeds of male and female mayflies;

[0051] S35, Mayfly mating

[0052] Female mayflies are ranked according to their fitness values. The best female mates with the best male, and the second-ranked female mates with the second-ranked male. After mating, two offspring are produced. The calculation formula is:

[0053]

[0054] Where: x offspring1 、x offspring2 is the specific position of the two offspring; x male is the position corresponding to the male mayfly; x female is the position of the corresponding female mayfly; α is the mating weight coefficient,

[0055]

[0056] Where: f male and f female is the fitness of male and female mayflies.

[0057] In step S3, the Halton sequence is used to initialize the male and female mayfly populations, specifically:

[0058] For the jth dimension, use a prime number p d Generate the nth point in the Halton sequence as a cardinality:

[0059]

[0060] Where: a i for The i-th digit in the base expansion;

[0061] Initialize the mayfly population:

[0062] x n =[h1(n),h2(n),...,h j (n)]

[0063] h1(n),h2(n),…,h j (n) is the Halton sequence value corresponding to each dimension;

[0064] Mapping to the search space:

[0065] x i,j =lb j +h j (n)·(ub j -lb j )

[0066] Where: ub j is the upper bound of the j-th dimension of the search space; lb j is the lower bound of the j-th dimension of the search space; x i,j is the actual position of the i-th mayfly individual (including females and males) in the j-th dimension.

[0067] In step S3, a variable step size strategy is introduced to rationalize the search mechanism, specifically:

[0068] In the early stages of the search, a larger step size is used, and in the later stages of the search, the step size is gradually reduced. At this time, the speed calculation formula of the male mayfly of the mayfly algorithm is changed to:

[0069]

[0070] The speed calculation formula of the female mayfly in the mayfly algorithm is changed to:

[0071]

[0072] Where: χ(t) is the dynamic step size factor.

[0073] The dynamic step size factor χ(t) is adapted by the following formula:

[0074]

[0075] Where: g max , g min is the upper and lower limits of the step factor; it is the current number of iterations; it max is the maximum number of iterations.

[0076] In step S3, random disturbance is added, namely:

[0077] After the mayflies mate in step S35, a normally distributed random disturbance is added to the selected offspring. The specific formula is:

[0078]

[0079] Where: β is the random disturbance intensity; ρ(-1.1) is the standard normal distribution with mean 1 and variance -1.

[0080] In step S4, the improved mayfly algorithm is used to solve the mathematical model of protection optimization setting to obtain the coordination strategy between the main protection and the backup protection, which is specifically:

[0081] S41, initialize the mayfly population size, variable dimension, and maximum number of iterations it _max ;

[0082] S42, initializing the position of the mayfly population, and bringing the position of each mayfly in the mayfly population into the objective function, calculating the fitness value of each mayfly in the initial population, and determining the current optimal mayfly position and its fitness value according to the calculation result;

[0083] S43, update the position and speed of the mayfly population, and the male mayfly with the best fitness continues the wedding dance.

[0084] And female mayflies that are not attracted to male mayflies fly randomly;

[0085] S44, sorting by fitness value, mating between male and female mayflies to produce the same number of offspring;

[0086] S45, after each iteration, check whether the maximum number of iterations has been reached, if not, return to step S43 to perform the next iteration; otherwise, go to step S46;

[0087] S46, output the optimal mayfly position, obtain the optimal solution, and the program ends.

[0088] The present invention proposes an improved inverse time overcurrent protection optimization setting, establishes a mathematical model for the optimization setting of the inverse time overcurrent protection of the distribution network, and adopts the improved mayfly algorithm (IMA) to solve the model; the present invention has the ability to adjust the protection setting value in real time, and can make instant adjustments according to the operating status of the distribution network and the changes in the fault type, thereby improving the sensitivity of the protection system. When the inverse time overcurrent protection setting optimization problem has a solution, through the optimization setting of the inverse time overcurrent protection parameters, while ensuring selectivity, the speed of protection can be effectively improved, which is conducive to the realization of emergency power restoration in the fast fault area, and is more in line with actual needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0089] Figure 1 It is a schematic diagram of the process of the present invention;

[0090] Figure 2 It is a flow chart of solving the tuning model by the improved mayfly algorithm in the present invention;

[0091] Figure 3 It is the IEEE39 node topology diagram;

[0092] Figure 4 This is a comparison chart of the convergence degree of the algorithm proposed in the present invention and the traditional algorithm when solving the problem. Specific implementation method 1

[0094] The present invention will be further explained below in conjunction with the accompanying drawings and specific implementations. It should be understood that the following specific implementations are only used to illustrate the present invention and are not used to limit the scope of the present invention.

[0095] A method for setting the inverse time overcurrent setting value of distribution network based on improved mayfly algorithm, such as Figure 1-2 As shown, the following steps are included:

[0096] S1. Give full play to the performance of relay protection devices, take the minimum total action time of the main protection and backup protection of the distribution network as the objective function, and establish a mathematical model for protection optimization and setting in combination with the constraints of protection selectivity and sensitivity requirements; the input of the mathematical model for protection optimization and setting is: characteristics of relay protection equipment, load changes, line load, power generation capacity, system frequency and other information; the output is: output the optimal action time of the main protection and backup protection and the coordination strategy between the main protection and backup protection.

[0097] Compared with the time-limited overcurrent protection, the inverse time-limited overcurrent protection has improved sensitivity and speed, and is widely used in distribution networks. The inverse time-limited overcurrent protection includes the time setting coefficient T P and starting current I P Two major parameters, inverse time overcurrent protection action characteristic equation:

[0098]

[0099] Where: i is the number of the protection device; j is the number of the line where the fault occurs; T pi is the time setting coefficient of protection device i; t ij To protect the action time; I ij is the current flowing through the protection; T ref To protect the starting current of equipment i; N T Inverse time characteristic index.

[0100] To give full play to the performance of relay protection devices, the objective function of the operation time and protection coordination of the main protection and backup protection of the distribution network is:

[0101]

[0102] In the formula: A is the number of main protection devices; C is the number of backup protection devices; L is the total number of fault lines; When a short circuit fault occurs on the current line j, protection i acts as the main protection action time; is the action time of protection k as backup protection when short circuit occurs in the current line j; R CO To ensure the coordination of protection, if the main protection device fails to operate in time, the backup protection device should be able to intervene quickly in a short time to avoid false operation and refusal to operate. λ1, λ2, and λ3 are weight coefficient values.

[0103] Consider the protection selectivity, sensitivity requirements and other constraints in combination with the actual operation conditions, including:

[0104] S21. In order to ensure the selectivity requirements between the main protection and the backup protection, it is necessary to design the time difference between the main and backup protection, that is:

[0105]

[0106] Where: ΔT is the minimum value of the time difference between the main protection and the backup protection. In order to ensure the reliability of the protection, ΔT cannot be too small;

[0107] S22. In order to ensure that the relay can operate without error at the maximum starting load current and can operate accurately at the minimum fault current, the relay should have a certain sensitivity. Its sensitivity S j The specific constraints are:

[0108]

[0109] Where: I op is the starting current of the relay; I l·min is the minimum starting load current of the line; I l·max is the maximum starting load current;

[0110] S23, relay action time limit, that is:

[0111] T i,min ≤T≤T i,max

[0112] Where: T i,min is the lower limit of the relay action time; T i,max is the upper limit of the relay action time; T is the relay action time value.

[0113] S24, the range of the relay time setting coefficient is: TDS i,min≤TDS≤TDS i,max

[0114] Where: TDS is the relay time setting coefficient; TDS i,min It is the lower limit of the relay time setting coefficient; TDS i,max It is the upper limit of the relay time setting coefficient.

[0115] S25. The protection device should meet the selectivity requirement, that is, when a fault occurs, it can accurately select the fault section and avoid false operation and refusal to operate. Selectivity can be expressed by constraining the false operation rate and refusal to operate rate:

[0116] P fault (I)≥η1(false operation rate constraint)

[0117] P reject (I)≤η2(rejection rate constraint)

[0118] Where: P fault (I) is the probability of malfunction of the protection device; P reject (I) is the probability of the protection device failing to operate; η1 and η2 are the upper limits of the allowable false operation rate and failure rate.

[0119] S3. Based on the mayfly algorithm, the Halton sequence is used to generate an initial mayfly population with super uniform distribution, and a variable step size strategy is introduced to rationalize the search mechanism and gene mutation;

[0120] Using the Halton sequence as the initialization of the male and female mayfly population, each mayfly is randomly placed in the d-dimensional problem search space, with a d-dimensional vector X = [x1,…,x d ] represents the potential solution represented by the mayfly, and the quality of the solution is evaluated according to the fitness function f(X) of each mayfly. The speed of each mayfly is a d-dimensional vector X = [v1,…,v d ], defined as the change in the mayfly position in each iteration, and each mayfly will adjust its flight direction to its individual optimal position so far and the optimal position obtained by the entire population so far.

[0121] The basic ephemera algorithm is as follows:

[0122] S31 Initialization

[0123] Initialize the positions and velocities of male and female mayfly populations,

[0124] Randomly generate n m Male mayflies f Position x of female mayflies m and x f :

[0125] xi ~U(lb,ub)

[0126] Where lb and ub are the lower and upper limits of the search space respectively.

[0127] Initialization speed v m and v f , usually initialized to zero or a small random value.

[0128] S32. Activities of male mayflies

[0129] Assumptions is the position space of male mayfly m at the tth iteration, by moving to the current position Add Speed Arriving at the new position, that is:

[0130]

[0131] Where: z1 and z2 are the weight coefficients of individual learning and social learning respectively; pbest mj is the optimal position of m mayflies in dimension j; gbest j is the optimal position of the entire mayfly population in dimension j; c1 and c2 are random numbers in the interval [0,1]; ω1 is the inertia weight of male mayflies,

[0132] In order to enhance the population diversity of the mayfly algorithm, an additional local search can be performed on the male mayfly with the best fitness through a weighted neighborhood exploration strategy. The formula is as follows:

[0133]

[0134] Where: The current position of the best male mayfly; δ search radius (disturbance intensity), usually a small positive number, dynamic adjustment can enhance the search ability; is a uniformly distributed random number in the interval [-1,1]; is the position of an individual randomly selected in the neighborhood; η is the neighborhood learning factor, which controls the influence weight of the optimal individual and the random neighborhood individual.

[0135] S33. Activities of female mayflies

[0136] Female mayflies are ranked according to their fitness function, the best females are attracted to the best males, the second-best females are attracted to the second-best males, and so on. is the position space of the female mayfly f at the tth iteration, by transforming the velocity Add to the current position to change the position, that is:

[0137]

[0138]

[0139] Where: ω2 is the inertia weight of the female mayfly; ∈ is the random flight coefficient; r is a random value in the range of [-1,1]; is a standard normally distributed random variable; a1 is the female learning factor; c3 is a random number in the interval [0,1].

[0140] S34, Mayfly speed limit

[0141]

[0142] Where: V max The maximum speed limited to mayflies;

[0143] S35, Mayfly mating

[0144] Female mayflies are ranked according to their fitness function. The best female mates with the best male, the second-ranked female mates with the second-ranked male, and so on. After mating, two offspring are produced. The calculation formula is:

[0145]

[0146] Where: x offspring1 、x offspring2 is the specific position of the two offspring; x male is the position corresponding to the male mayfly;

[0147] x female is the position of the corresponding female mayfly; α is the mating weight coefficient, which includes:

[0148]

[0149] Where: f male and f female is the fitness of male and female mayflies.

[0150] In step S31, the Halton sequence is used as the initialization of the male and female mayfly population to prevent the population from gathering near the local optimal solution. By covering different areas of the search space, the Halton sequence can provide better global search capabilities, thereby improving the convergence and performance of the optimization algorithm. Specifically:

[0151] For the jth dimension, use a prime number p d Generate the nth point in the Halton sequence as a cardinality:

[0152]

[0153] Where: ai For p j- The i-th digit in the base expansion;

[0154] Initialize the mayfly population:

[0155] x n =[h1(n),h2(n),...,h j (n)]

[0156] h1(n),h2(n),...,h j (n) is the Halton sequence value corresponding to each dimension;

[0157] Mapping to the search space:

[0158] x i,j =lb j +h j (n)·(ub j -lb j )

[0159] Where: ub j is the upper bound of the j-th dimension of the search space; lb j is the lower bound of the j-th dimension of the search space; x i,j is the actual position of the i-th mayfly individual (including both females and males) in the j-th dimension

[0160] Introduce a variable step size strategy, specifically:

[0161] Use a larger step size in the early stages of the search (high exploratory) and gradually reduce the step size in the later stages of the search (fine search). Dynamic adjustment of the step size helps to strike a balance between global and local search.

[0162] At this time, the speed calculation formula of the male mayfly in step S32 of the mayfly algorithm is changed to:

[0163]

[0164] The speed calculation formula of the female mayfly in step S33 is changed to:

[0165]

[0166] Where: χ(t) is a dynamic step size factor, which can be adaptive during the iteration process, so that the algorithm can develop some specific areas in the later stage of iteration, and adapt itself through the following formula.

[0167]

[0168] Where: g max , g minis the upper and lower limits of the step factor; it is the current number of iterations; it max is the maximum number of iterations.

[0169] After the mayflies mate in step S35, a normally distributed random disturbance is added to the selected offspring. The specific formula is:

[0170]

[0171] Where: β is the random disturbance intensity; ρ(-1.1) is the standard normal distribution with mean 1 and variance -1.

[0172] Step S4 uses the improved mayfly algorithm to solve the mathematical model of protection optimization setting. The specific solution process includes:

[0173] There is no difference in the optimization setting model of inverse time overcurrent protection between distribution networks without DG and distribution networks with single or multiple DG. The difference is that for distribution networks with DG, when calculating the fault current I flowing through the protection ij When the contribution of DG to the short-circuit fault current is considered, the optimized setting value of the protection parameter obtained by the present invention is applicable to the distribution network without DG, with a single type of DG or with multiple types of DG.

[0174] S41, initialize the optimization problem, the parameters include the size of the mayfly population, the variable dimension, the maximum number of iterations it _max wait;

[0175] S42, initializing the position of the mayfly population, and bringing the position of each mayfly in the mayfly population into the objective function, calculating the fitness value of each mayfly in the initial population, and determining the current optimal mayfly position and its fitness value according to the calculation result;

[0176] S43, updating the position and speed of the mayfly population, the male mayflies with the best fitness continue the wedding dance, and the female mayflies that are not attracted by the male mayflies fly randomly;

[0177] S44. Sort by fitness value, and mate with male and female mayflies to produce the same number of offspring.

[0178] S45, after each iteration, check whether the maximum number of iterations has been reached. If the maximum number of iterations has not been reached, return to step S43 for the next iteration; otherwise, go to step S46.

[0179] S46, output the optimal mayfly position, obtain the optimal solution, and the program ends. Example

[0180] The present invention simulates and analyzes the three-phase short circuit fault at the midpoint of the line. The IEEE 39-node system is as follows: Figure 3 The simulation results of the method proposed by the present invention and the traditional method are shown in Table 1.

[0181] Table 1 Main / backup protection action time and protection coordination in IEEE 39 bus system

[0182]

[0183] As shown in Table 1, the method of the present invention is better than the traditional method as a whole, and performs better in the action time of the main protection and backup protection, and the degree of protection coordination. This shows that the optimization strategy adopted by the present invention can more efficiently deal with the three-phase short-circuit fault at the midpoint of the line, and improve the reliability and fault handling capability of the system.

[0184] The degree of iterative convergence of different algorithms is as follows: Figure 4 As shown, according to Figure 4 It is shown that the present invention performs faster in terms of global convergence speed, and its convergence value is lower than that of the traditional algorithm, indicating that the present invention is more efficient in the process of minimizing the objective function. In summary, the algorithm is superior to other comparative algorithms in terms of convergence speed and global search performance. The present invention has the ability to adjust the protection setting in real time, and can adjust the parameters of the protection system in real time according to the operating status of the distribution network and the changes in the fault type, thereby improving its sensitivity. In the case that there is a solution to the optimization problem of the inverse time overcurrent protection setting, by optimizing and setting the inverse time overcurrent protection parameters, it can not only ensure the selectivity of the protection, but also effectively improve the speed of the protection. This helps to quickly restore power supply, urgently handle fault areas, and better meet actual needs. Through the application of the present invention, the distribution network can more reliably identify and protect faults, thereby improving the reliability and safety of the power supply system.

[0185] It should be noted that the above content only illustrates the technical idea of ​​the present invention and cannot be used to limit the protection scope of the present invention. For ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications all fall within the protection scope of the claims of the present invention.

Claims

1. A method for setting the inverse time overcurrent setting value of a distribution network based on an improved mayfly algorithm, characterized in that: The steps include: S1. Taking the action time and protection coordination of the main protection and backup protection of the distribution network as the objective function, combined with the constraints of protection selectivity and sensitivity requirements, a mathematical model for protection optimization setting is established; S2. Based on the mayfly algorithm, the Halton sequence is used to generate an initial mayfly population with super uniform distribution, and a variable step size strategy is introduced to rationalize the search mechanism and add random disturbances to obtain an improved mayfly algorithm. S3. The improved mayfly algorithm is used to solve the mathematical model of protection optimization setting to obtain the coordination strategy between the main protection and the backup protection.

2. According to claim 1, a method for setting the inverse time overcurrent setting value of a distribution network based on an improved mayfly algorithm, characterized in that: In step S1, the objective function is: In the formula: A is the number of main protection devices; C is the number of backup protection devices; L is the total number of fault lines; When a short circuit fault occurs on the current line j, protection i acts as the main protection action time; is the action time of protection k as backup protection when short circuit occurs in the current line j; R CO For the purpose of protection coordination, if the main protection device fails to operate in time, the backup protection device should be able to intervene quickly in a short time to avoid false operation and refusal to operate; λ1, λ2, and λ3 are weight coefficient values.

3. The method for setting the inverse time overcurrent setting value of a distribution network based on the improved mayfly algorithm according to claim 1, characterized in that: In step S1, the constraints of protection selectivity and sensitivity are considered, specifically: In order to ensure the selectivity requirements between the main protection and the backup protection, the time difference between the main and backup protection meets the following requirements: Where: ΔT is the minimum value of the time difference between the main protection and the backup protection. In order to ensure the reliability of the protection, ΔT cannot be too small; In order to ensure that the relay can operate without error at the maximum starting load current and can operate accurately at the minimum fault current, the relay should have a certain sensitivity. Its sensitivity S j The specific constraints are: Where: I op is the starting current of the relay; I l·min is the minimum starting load current of the line; I l·max is the maximum starting load current; The relay action time limit is: T i,min ≤T≤T i,max Where: T i,min is the lower limit of the relay action time; T i,max is the upper limit of the relay action time; T is the relay action time value; The range of relay time setting coefficient is: TDS i,min ≤TDS≤TDS i,max Where: TDS is the relay time setting coefficient; TDS i,min It is the lower limit of the relay time setting coefficient; TDS i,max It is the upper limit value of the relay time setting coefficient; The protection device should meet the selectivity requirement, that is, when a fault occurs, it can accurately select the fault section and avoid false operation and refusal to operate. The selectivity can be expressed by constraining the false operation rate and refusal to operate rate: P fault (I)≥η1 P reject (I)≤η2 Where: P fault (I) is the probability of malfunction of the protection device; P reject (I) is the probability of the protective device refusing to operate; η1 and η2 are the upper limits of the allowable false operation rate and rejection rate.

4. The method for setting the inverse time overcurrent setting value of a distribution network based on an improved mayfly algorithm according to claim 1, characterized in that: In step S3, the mayfly algorithm is specifically as follows: S31 Initialization Initialize the positions and velocities of male and female mayfly populations, Randomly generate n m Male mayflies f Position x of female mayflies m and x f ; The search space of mayfly i is expressed as: i ~U(lb,ub) Where: lb, ub are the lower and upper limits of the search space respectively; Initialize the speed v of male and female mayflies m and v f ; S32. Activities of male mayflies Assumptions is the position space of male mayfly m at the tth iteration, by moving to the current position Add Speed Arriving at the new position, that is: Where: z1 and z2 are the weight coefficients of individual learning and social learning respectively; pbest mj is the optimal position of m mayflies in dimension j; gbest j is the optimal position of the entire mayfly population in dimension j; c1 and c2 are random numbers in the interval [0,1]; ω1 is the inertia weight of male mayflies, In order to enhance the population diversity of the mayfly algorithm, an additional local search is performed on the male mayfly with the best fitness through a weighted neighborhood exploration strategy. The formula is as follows: Where: The current position of the best male mayfly; δ search radius, usually a small positive number, dynamic adjustment can enhance the search ability; is a uniformly distributed random number in the interval [-1,1]; is the position of a randomly selected individual in the neighborhood; η is the neighborhood learning factor, which controls the influence weight of the optimal individual and the random neighborhood individual; S33. Activities of female mayflies Female mayflies are ranked according to their fitness function. The best females will be attracted to the best males, and the second-best females will be attracted to the second-best males. Assume is the position space of the female mayfly f at the tth iteration, by transforming the velocity Add to the current position to change the position, that is: Where: ω2 is the inertia weight of the female mayfly; ∈ is the random flight coefficient; r is a random value in the range of [-1,1]; is a standard normal distribution random variable; a1 is the female learning factor; c3 is a random number in the interval [0,1]; is the speed of female mayfly f at the tth iteration; is the position space of male mayfly m at the tth iteration; is the position space of the female mayfly f at the tth iteration; is the speed of the female mayfly f after the t+1th iteration; is the objective function value of the strategy corresponding to the position of male mayfly m after t iterations; is the objective function value of the strategy corresponding to the position of the female mayfly f after t iterations; S34, Mayfly speed limit Where: V max v is the maximum speed limited by the mayfly; i represents the set of speeds of male and female mayflies; S35, Mayfly mating Female mayflies are ranked according to their fitness values. The best female mates with the best male, and the second-ranked female mates with the second-ranked male. After mating, two offspring are produced. The calculation formula is: Where: x offspring1 、x offspring2 is the specific position of the two offspring; x male is the position corresponding to the male mayfly; x female is the position of the corresponding female mayfly; α is the mating weight coefficient, Where: f male and f female is the fitness of male and female mayflies.

5. A method for setting the inverse time overcurrent setting value of a distribution network based on an improved mayfly algorithm according to claim 1 or 4, characterized in that: In step S3, the Halton sequence is used to initialize the male and female mayfly populations, specifically: For the jth dimension, use a prime number p d Generate the nth point in the Halton sequence as a cardinality: Where: a i For p j- The i-th digit in the base expansion; Initialize the mayfly population: x n =[h1(n),h2(n),…,h j (n)] h1(n),h2(n),...,h j (n) is the Halton sequence value corresponding to each dimension; Mapping to the search space: x i,j =lb j +h j (n)·(ub j -lb j ) Where: ub j is the upper bound of the j-th dimension of the search space; lb j is the lower bound of the j-th dimension of the search space; x i,j is the actual position of the i-th mayfly individual (including females and males) in the j-th dimension.

6. A method for setting the inverse time overcurrent setting value of a distribution network based on an improved mayfly algorithm according to claim 1 or 4, characterized in that: In step S3, a variable step size strategy is introduced to rationalize the search mechanism, specifically: In the early stages of the search, a larger step size is used, and in the later stages of the search, the step size is gradually reduced. At this time, the speed calculation formula of the male mayfly of the mayfly algorithm is changed to: The speed calculation formula of the female mayfly in the mayfly algorithm is changed to: Where: χ(t) is the dynamic step size factor.

7. The method for setting the inverse time overcurrent setting value of a distribution network based on the improved mayfly algorithm according to claim 6, characterized in that: The dynamic step size factor χ(t) is adapted by the following formula: Where: g max , g min is the upper and lower limits of the step factor; it is the current number of iterations; it max is the maximum number of iterations.

8. The method for setting the inverse time overcurrent setting value of a distribution network based on the improved mayfly algorithm according to claim 4, characterized in that: In step S3, random disturbance is added, namely: After the mayflies mate in step S35, a normally distributed random disturbance is added to the selected offspring. The specific formula is: Where: β is the random disturbance intensity; ρ(-1.1) is the standard normal distribution with mean 1 and variance -1.

9. The method for setting the inverse time overcurrent setting value of a distribution network based on the improved mayfly algorithm according to claim 1, characterized in that: In step S4, the improved mayfly algorithm is used to solve the mathematical model of protection optimization setting to obtain the coordination strategy between the main protection and the backup protection, which is specifically: S41, initialize the mayfly population size, variable dimension, and maximum number of iterations it _max ; S42, initializing the position of the mayfly population, and bringing the position of each mayfly in the mayfly population into the objective function, calculating the fitness value of each mayfly in the initial population, and determining the current optimal mayfly position and its fitness value according to the calculation result; S43, updating the position and speed of the mayfly population, the male mayflies with the best fitness continue the wedding dance, and the female mayflies that are not attracted by the male mayflies fly randomly; S44, sorting by fitness value, mating between male and female mayflies to produce the same number of offspring; S45, after each iteration, check whether the maximum number of iterations has been reached, if not, return to step S43 to perform the next iteration; otherwise, go to step S46; S46, output the optimal mayfly position, obtain the optimal solution, and the program ends.

Images