A method, device and storage medium for locating faults in a power distribution network

Abstract

The present invention discloses a method, device, and storage medium for locating faults in a power distribution network containing a source, in the technical field of power system fault location. The method comprises: obtaining an evaluation function for the power distribution network containing a source under a fault state based on acquired fault current information of the power distribution network containing a source; utilizing the evaluation function as the fitness function of an improved particle swarm optimization algorithm, utilizing candidate solutions for the fault section of the power distribution network containing a source as particles, updating the particle speed and position through the improved particle swarm optimization algorithm, performing fault location on the power distribution network containing a source, and obtaining the fault section of the power distribution network containing a source; utilizing the improved particle swarm optimization algorithm to update the particle speed and position using adaptive inertia weights; and constraining the output of the improved particle swarm optimization algorithm using pre-set distribution network stable operation constraints. The present invention can quickly and accurately determine the fault section of the power distribution network containing a source, achieving high-precision fault location on the power distribution network containing a source.

Description

Technical Field

[0001] The present invention relates to a method, a device and a storage medium for locating a fault in a power distribution network containing a source, and belongs to the technical field of power system fault locating. Background Art

[0002] In recent years, with the rapid development of my country's national economy and the continuous improvement of people's living standards, the demand for electricity has gradually increased. As the hub between the transmission system and electricity users, the distribution system plays a key role in the distribution of electricity. In the context of carbon peak and carbon neutrality, in order to accelerate energy transition, the proportion of distributed generation (DG) in the distribution network has increased significantly. This has led to an increasingly complex network structure, increased network losses, and wasted resources.

[0003] Fault location in distribution networks is the initial step in grid fault self-healing technology. A fast and accurate location solution is a prerequisite for ensuring the safety and reliability of distribution networks. Fault location in distribution networks is divided into fault point location and fault section location. Fault point location primarily measures the distance between the monitoring point and the fault point to determine the fault's location. This method, also known as fault distance measurement, traditionally includes the traveling wave method, the impedance method, and the S-injection method. Fault section location, on the other hand, locates the fault section between two sectionalizers. This method significantly aids in fault isolation after a fault occurs.

[0004] Currently, distribution network fault location algorithms are mainly divided into two categories. One is a matrix algorithm based on the intelligent feeder terminal unit (FTU) of the distribution network, also known as the direct method, which has problems such as low fault tolerance; the other is artificial intelligence algorithms, such as particle swarm optimization, genetic algorithm, harmony search algorithm, immune algorithm, simulated annealing algorithm, etc. As more and more DGs are connected to the distribution network, the distribution system has changed from a unidirectional flow to a multidirectional flow, and the network structure has become increasingly complex. Due to certain limitations of the direct method, it is no longer applicable to actual engineering practices. Therefore, artificial intelligence algorithms are gradually being used in distribution network fault location. However, some existing distribution network fault location methods based on artificial intelligence algorithms have problems such as high algorithm complexity, slow calculation speed, and insufficient fault location accuracy. Therefore, how to provide a distribution network fault location method with higher accuracy and faster calculation speed is a problem that needs to be solved by those skilled in the art. Summary of the Invention

[0005] In order to solve the problems existing in the prior art, the present invention proposes a method, device and storage medium for locating faults in a source-containing distribution network, which uses an improved particle swarm optimization algorithm to locate faults in the source-containing distribution network and obtain accurate and reliable distribution network fault sections.

[0006] In order to solve the above technical problems, the present invention adopts the following technical means:

[0007] In a first aspect, the present invention proposes a method for locating a fault in a power distribution network including a source, comprising the following steps:

[0008] According to the acquired fault current information of the source distribution network, an evaluation function of the source distribution network under the fault state is obtained;

[0009] The evaluation function is used as the fitness function of the improved particle swarm optimization algorithm, and the candidate solutions of the fault section of the source distribution network are used as particles. The speed and position of the particles are updated by the improved particle swarm optimization algorithm to locate the fault of the source distribution network and obtain the fault section of the source distribution network;

[0010] The improved particle swarm optimization algorithm uses adaptive inertia weights to update the speed and position of particles;

[0011] The output result of the improved particle swarm optimization algorithm is constrained by pre-set distribution network stable operation constraint conditions.

[0012] In combination with the first aspect, further, a method for obtaining an evaluation function of a source-containing distribution network under a fault state is as follows:

[0013] According to the positions of the sectionalizer switches in the source-containing distribution network, the source-containing distribution network is divided into sections to obtain a plurality of sections corresponding to the sectionalizer switches one by one;

[0014] encoding the segment switches and segments;

[0015] According to the code and the fault current information of the source distribution network, an actual state value and an expected state value of each section switch in the source distribution network under a fault state are obtained;

[0016] The evaluation function of the source distribution network under fault state is obtained according to the actual state value and expected state value of each section switch.

[0017] In combination with the first aspect, further, the expression of the evaluation function is as follows:

[0018]

[0019] Among them, F(S B ) is the evaluation function of the distribution network with source, S B is the state value of the distribution network with source, I j is the actual state value of the jth section switch in the source distribution network, is the expected state value of the jth section switch in the source distribution network, ω is the evaluation weight, S B (j) is the state value of the j-th section feeder in the source distribution network, S B(j) = 1 means the jth section feeder is faulty, S B (j) = 0 means that the jth section feeder is normal, j = 1, 2, ..., N, where N is the total number of section switches in the source distribution network.

[0020] Combined with the first aspect, further, according to the fault current information of the source distribution network, the actual state value of each section switch is obtained, I j The expression is:

[0021]

[0022] In combination with the first aspect, the improved particle swarm optimization algorithm is further used to update the speed and position of particles to locate the fault of the source distribution network, and the fault section of the source distribution network is obtained, including:

[0023] Initialize network parameters and particle swarm parameters, wherein the particle swarm parameters include the position and velocity of particles;

[0024] Calculating the fitness value of each particle according to the evaluation function and the position of the particle;

[0025] According to the fitness values ​​of all particles, the individual optimal position of each particle and the optimal position of the population of all particles are obtained;

[0026] Update the adaptive inertia weight of each particle according to the fitness value of all particles;

[0027] updating the velocity and position of each particle according to the adaptive weight;

[0028] According to the updated particle position, the new fitness value of each particle is calculated using the evaluation function;

[0029] According to the new fitness values ​​of all particles, update the individual optimal position of each particle and the optimal position of the population of all particles;

[0030] When the iterative convergence conditions are met, the optimal position of the iteratively converged population is output and the fault section of the distribution network containing the source is obtained.

[0031] In combination with the first aspect, further, updating the adaptive inertia weight of each particle according to the fitness value of all particles includes:

[0032] According to the fitness values ​​of all particles, the average fitness value of the population at the current moment is obtained

[0033] According to the fitness value f of the optimal position of the population at the current moment max and the average fitness value Calculate the weight selection ratio of each particle. The expression is as follows:

[0034]

[0035] Among them, k i represents the weight selection ratio of particle i, i = 1, 2, ..., M, M is the total number of particles in the particle swarm;

[0036] According to the weight selection ratio and the fitness value of each particle, the adaptive inertia weight of each particle is updated. The expression is as follows:

[0037]

[0038] Among them, P i represents the adaptive inertia weight of particle i, f i Represents the fitness value of particle i.

[0039] In combination with the first aspect, further, in the improved particle swarm optimization algorithm, when the distance between the position of particle i and the optimal position of the swarm at the current moment is less than a preset threshold, chaotic search is used to update the position of particle i, i = 1, 2, ..., M, where M is the total number of particles in the particle swarm.

[0040] In combination with the first aspect, further, the stable operation condition of the distribution network includes node voltage constraint conditions and branch current constraint conditions.

[0041] When the improved particle swarm optimization algorithm meets the iterative convergence conditions, the power flow calculation of the source distribution network is performed according to the iteratively converged optimal position of the population, and the voltage value and current value corresponding to the iteratively converged optimal position of the population are obtained; it is judged whether the voltage value and current value meet the stable operation conditions of the distribution network. If so, the fault section of the source distribution network is output according to the iteratively converged optimal position of the population. If not, the improved particle swarm optimization algorithm is reused for particle swarm optimization.

[0042] In a second aspect, the present invention provides a fault location device for a power distribution network including a source, comprising:

[0043] Data acquisition module, used to obtain fault current information of the source distribution network;

[0044] A model building module is used to obtain an evaluation function of the source-containing distribution network under a fault state based on the acquired fault current information of the source-containing distribution network;

[0045] a fault location module, configured to utilize the evaluation function as a fitness function of an improved particle swarm optimization algorithm, utilize candidate solutions of the fault section of the source distribution network as particles, update the velocity and position of the particles through the improved particle swarm optimization algorithm, perform fault location on the source distribution network, and obtain the fault section of the source distribution network;

[0046] In the fault location module, the improved particle swarm optimization algorithm uses adaptive inertia weights to update the speed and position of particles;

[0047] The fault location module is further configured to constrain the output result of the improved particle swarm optimization algorithm through preset distribution network stable operation constraint conditions.

[0048] In a third aspect, the present invention proposes a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for locating a fault in a source-containing distribution network described in the first aspect is implemented.

[0049] The following advantages can be obtained by adopting the above technical means:

[0050] The present invention proposes a method, device, and storage medium for locating faults in a power distribution network containing a source. Through an improved particle swarm optimization algorithm, the fault section of the power distribution network containing a source can be quickly and accurately obtained, achieving high-precision fault location in the power distribution network containing a source, providing favorable support for fault isolation and power restoration in the power distribution network containing a source, and improving the safety of the power distribution network containing a source. The present invention uses an adaptive inertia weight optimization algorithm to prevent the particle swarm optimization algorithm from falling into a local optimum, thereby improving the convergence speed and convergence performance of the particle swarm optimization algorithm. The present invention also constrains the output results of the particle swarm optimization algorithm through pre-set distribution network stable operation constraints, ensuring that the fault section ultimately output meets the basic operation requirements of the distribution network, further improving the accuracy of fault location. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 This is a flow chart of the steps of the method for locating a fault in a power distribution network containing a source according to the present invention;

[0052] Figure 2 Schematic diagram of the optimization process of the improved particle swarm optimization algorithm in an embodiment of the present invention;

[0053] Figure 3 is a simplified schematic diagram of a traditional power distribution network in an embodiment of the present invention;

[0054] Figure 4 This is a simplified schematic diagram of the distribution network after DG is connected to the grid in an embodiment of the present invention;

[0055] Figure 5 Schematic diagram of the IEEE 33-node power distribution system in an embodiment of the present invention. DETAILED DESCRIPTION

[0056] The technical solution of the present invention will be further described below with reference to the accompanying drawings:

[0057] It is necessary to explain that: the particle swarm optimization algorithm (PSO) originated from the study of the predation behavior of bird flocks. Its core idea is to use the cooperation and information sharing between each individual in the population to make the entire population move from disordered to orderly evolution in the solution space of the optimization target, and finally obtain the global optimal value. The algorithm has a simple mathematical principle, is easy to program, and has a fast convergence speed, but it is easy to fall into local optimality and show "premature" phenomenon. There are many improved studies on particle swarm optimization algorithms, such as binary particle swarm optimization algorithm (BPSO), but the improved algorithm still has defects. The present invention makes further improvements on the basis of BPSO, introduces adaptive proportional selection strategy and chaotic search into the algorithm, enhances the convergence performance of the algorithm, and at the same time, constrains the output results of the algorithm according to the characteristics of the source distribution network itself to obtain accurate and reliable fault location results.

[0058] Example 1:

[0059] This embodiment introduces a method for locating faults in a power distribution network containing a source. Figure 1 As shown, the specific steps include:

[0060] Step A: According to the acquired fault current information of the source-containing distribution network, an evaluation function of the source-containing distribution network in a fault state is obtained.

[0061] Using section switches as nodes, the traditional distribution network can be simplified to Figure 3 , where S1, S2, S3, S4, and S5 are section switches, and L1, L2, L3, L4, and L5 switches correspond to sections. When DG is connected to the grid, the distribution network can be simplified to Figure 4 , where DG1 and DG2 represent distributed power sources connected to the distribution network, and K1 and K2 represent switches that control the switching of distributed power sources.

[0062] In an embodiment of the present invention, based on the normal operating conditions of the distribution network, that is, the network is radial and there are no isolated islands or ring networks, the source-containing distribution network is segmented according to the positions of the sectionalizers in the source-containing distribution network, and multiple segments corresponding one-to-one to the sectionalizers are obtained. The sectionalizers and segments are then coded. Considering the impact of DG grid connection, a source-containing distribution network model is established according to the logical relationship between the sectionalizers and the faulty segments in the distribution line. The source-containing distribution network model includes a switching function and an evaluation function.

[0063] The present invention uses fault current information collected by the intelligent feeder terminal unit (FTU) of the distribution network to obtain the actual state value and expected state value of each section switch in the source-containing distribution network under the fault state. By bringing the actual state value and expected state value of each section switch into the source-containing distribution network model, an evaluation function of the source-containing distribution network under the fault state can be obtained.

[0064] The evaluation function represents the difference between the fault current information and the expected value of the switch. The essence of the distribution network fault location method of the present invention is to use an intelligent algorithm to optimize the minimum value of the evaluation function. Due to the network complexity brought about by DG grid connection, the traditional evaluation method may suffer from information distortion and information loss during the optimization process. Therefore, the present invention constructs a new evaluation function. The expression of the evaluation function is as follows:

[0065]

[0066] Among them, F(S B ) is the evaluation function of the distribution network with source, S B is the state value of the distribution network with source, I j is the actual state value of the jth section switch in the source distribution network, is the expected state value of the jth section switch in the source distribution network, ω is the evaluation weight, ω takes values ​​within (0,1), usually 0.5, SB(j) is the state value of the j-th section feeder in the source distribution network, S B (j) = 1 means the jth section feeder is faulty, S B (j) = 0 means that the jth section feeder is normal, j = 1, 2, ..., N, where N is the total number of section switches in the source distribution network.

[0067] In a traditional single-source distribution network, when a fault occurs at a certain point, the fault current is encoded in only two states: whether the fault current flows or not. That is, the section switch has only two states: 1 and 0. Since the connection of DG to the grid affects the power flow direction of the distribution network after the fault, the traditional fault current encoding is no longer applicable. This invention improves the traditional model. The improved encoding rules based on the fault current direction are as follows:

[0068] When the fault current direction at the jth section switch is positive, I j Take "1"; when the fault current direction at the jth section switch is negative, I j Take "-1"; when no fault current is detected at the jth section switch, I j Taking "0", it can be expressed as:

[0069]

[0070] Attach Figure 4 For example, if a fault occurs in section L3, the fault current of the section switch S3 flows from the system power supply G to the load, which is the positive direction and is "1"; the fault current of the section switch S4 flows from the distributed power supply DG1 to the load, which is the reverse direction and is "-1".

[0071] The expected state of the switch changes according to the actual situation of the distribution network. The present invention can obtain the expected state of each switch according to the switches and sections after segment coding. Figure 4 For example, The expression is as follows:

[0072]

[0073] Traditional switching functions are often used to locate faults in single power supply networks. However, with the current large-scale access of distributed generators (DGs) to distribution networks, traditional functions are no longer applicable. Therefore, this paper constructs a new switching function that takes DG switching into account. Compared with traditional functions, this function can dynamically change according to the access conditions of different distributed power sources, thereby accurately reflecting the relationship between the fault current information of each switch and the feeder interval.

[0074] The switch function of the present invention is as follows:

[0075]

[0076] Among them, I j (s) is a new switch function; ∏ is a logical OR operation; I ju (s), I jd (s) is the upper and lower switching function value of the jth segment switch; M1 and N2 represent the number of power sources and feeder intervals at the upper end, respectively; N1 and M2 represent the number of power sources and feeder intervals at the lower end, respectively; S j,Gu , K u They represent the state from the upper power supply to the node and the access coefficient of the power supply in this interval. If the power supply is detected, K u It is recorded as "1", otherwise it is recorded as "0".

[0077] With the continuous development of DG, a large number of DGs are connected to the power grid, which will to a certain extent increase the randomness of the power grid load forecast and have a certain impact on the normal operation of the distribution network. If DGs cannot be reasonably incorporated, the stability of the distribution system will be reduced. Figure 5 It is a standard IEEE 33-node distribution system, in which multiple DGs are added. The grid-connected parameters of DGs are shown in Table 1.

[0078] Table 1

[0079]

[0080] In the fault section location of the distribution network containing DG, DG is generally treated as a PQ node. The randomness of its output is not considered. Instead, the capacity and output of DG are given. The grid connection of DG has a supporting effect on the network, which can reduce network losses and optimize the operation of the entire distribution system.

[0081] Step B: Using an improved particle swarm optimization algorithm to locate the fault in the distribution network containing the source, the fault section of the distribution network containing the source is obtained. Specifically, an evaluation function is used as the fitness function of the improved particle swarm optimization algorithm, and candidate solutions of the fault section of the distribution network containing the source are used as particles. The improved particle swarm optimization algorithm updates the particle velocity and position, and the particle position that minimizes the fitness index is found, thereby obtaining the fault section of the distribution network containing the source.

[0082] like Figure 2 As shown, the specific operations of step B are as follows:

[0083] Step B01: Initialize network parameters and particle swarm parameters. Network parameters include the maximum number of iterations, learning factors C1, C2, and particle speed limit V. max The particle swarm parameters include the particle position, velocity, and particle swarm population size M.

[0084] In the present invention, 0 and 1 are used to represent the status word of each section in the distribution network, 0 represents no fault, 1 represents fault, and all sections are encoded to form N-dimensional particles. Figure 4 For example, if a fault occurs in section L2, the final fault section result obtained by the algorithm when the fault occurs is: [0 1 0 0 0 0 0].

[0085] Step B02: Get the S corresponding to the current particle according to the initialized particle position B The value of (j) is then substituted into formula (1) to calculate the initial fitness value of each particle through the evaluation function. According to the initial fitness values ​​of all particles, the initial individual optimal position of each particle and the initial population optimal position of all particles are obtained. Specifically, the fitness value of each particle is compared. The smaller the fitness value, the better the particle position corresponding to the fitness value.

[0086] Step B03: Update the adaptive inertia weight of each particle according to the fitness value of all particles.

[0087] In the present invention, the position of the particle is related to the speed of the particle. The position of each particle can only be 0 or 1. The greater the speed of the particle, the greater the probability that the particle position is 1. The speed of the particle is related to the inertia weight. The present invention further improves the traditional BPSO and proposes a new adaptive inertia weight P. The adaptive inertia weight P can dynamically change with the size of the particle fitness, which can avoid the algorithm from falling into local optimality and slow convergence.

[0088] The expression of adaptive inertia weight is as follows:

[0089]

[0090] Among them, Pi represents the adaptive inertia weight of particle i, f i represents the fitness value of particle i, k i Represents the weight selection ratio of particle i, i = 1, 2, ..., M, and M is the total number of particles in the particle swarm.

[0091] In the adaptive inertia weight, the smaller the weight selection ratio, the smaller the selection pressure, and particles with poor fitness have a chance to survive, so the search range will be expanded; on the contrary, the larger the weight selection ratio, the greater the selection pressure, and particles with poor fitness will be eliminated, so the search range will be reduced.

[0092] Weight selection ratio k i The expression is as follows:

[0093]

[0094] in, Indicates the average fitness value of the population at the current moment, f max Indicates the fitness value of the optimal position of the population at the current moment.

[0095] If the algorithm focuses too much on global search, it may cause the search range to be too large, reducing the search efficiency; if the algorithm focuses too much on local search, it may fall into the local optimal solution, resulting in the algorithm being unable to find a better solution. In order to better balance the relationship between global search and local search, the present invention uses P value (adaptive inertia weight) and k value (weight selection ratio) to improve the algorithm. At different stages of algorithm iteration, the k value can be dynamically adjusted, and then the P value can be adjusted, and finally the algorithm search range can be adjusted. In the first stage of iteration, since the population is randomly generated, the differences between individuals are large, and the gap between the average fitness of the population and the optimal fitness is large, so the k value calculated according to formula (6) is very small, which can expand the search range and avoid the premature phenomenon of the particle swarm; in the second stage of iteration, the k value will be adaptively adjusted to speed up the population to reach the global optimal speed; in the third stage of iteration, the average fitness value of the population is close to the optimal fitness, the k value is close to infinity, and the pressure of particle selection becomes infinite, thereby promoting the algorithm to converge to the global optimal solution. The adaptive inertia weight of the present invention can simultaneously improve the convergence speed and convergence accuracy of the algorithm, thereby improving the accuracy of fault location in the source distribution network.

[0096] Step B04: Update the speed and position of each particle according to the adaptive weight.

[0097] The update formula of particle velocity is as follows:

[0098]

[0099] in, represents the velocity of particle i in the dth dimension at the t+1th iteration, represents the velocity of particle i in the dth dimension at the tth iteration, r1 and r2 represent random numbers in [0,1]. represents the optimal position of particle i in the dth dimension at the tth iteration, represents the optimal position of the entire particle swarm in the dth dimension at the tth iteration, represents the position of particle i in the dth dimension at the tth iteration, d = 1, 2, ..., N.

[0100] The position of each particle Take 0 or 1, and update the particle position according to the updated particle velocity. The expression is as follows:

[0101]

[0102] in, represents the position of particle i in the dth dimension at the t+1th iteration, r represents a random number in [0,1], Sigmoid(x)=1 / 1+e -x .

[0103] Step B05: In the later stage of the improved particle swarm optimization algorithm, the convergence speed and convergence accuracy often fail to meet the expected requirements. Since chaotic variables have initial value sensitivity and ergodicity, the present invention combines the Logistics mapping formula in chaos optimization with the improved particle swarm optimization algorithm to perform chaotic search on examples with poor fitness in the later stage of iteration, thereby increasing the search accuracy of particles and accelerating the convergence speed.

[0104] When the distance d between the position of particle i and the optimal position of the population at the current moment i If the distance d between the position of particle i and the optimal position of the population at the current moment is less than the preset threshold, it means that particle i is poor, and chaotic search is used to update the position of particle i. i If the value is not less than the preset threshold, the position of particle i is kept unchanged. In the embodiment of the present invention, the threshold is 10 -2 .

[0105] d i =(X i -X gbest ) 2 (9)

[0106] Among them, X i represents the position of particle i, X gbest represents the optimal position of the population.

[0107] The present invention maps the position of the particle into the chaotic space, updates the position of the particle through the Logistics formula, and then reversely maps it to obtain the particle position of the chaotic search.

[0108] In the prior art, there are applications of using chaos search to initialize particle swarms. However, the chaos search algorithm requires adjusting multiple parameters, such as chaotic system parameters, search range, etc. Using chaos search to initialize the population can easily increase the complexity of the algorithm and the difficulty of debugging. Once the chaos search parameters are adjusted incorrectly, it will affect the global search, resulting in low search accuracy. Moreover, using the random value of the chaos search as the initial value, the population takes a long time to reach a stable state, affecting the convergence speed of the algorithm. In the late iteration, the present invention performs chaos search on a small number of particles with poor fitness. When the distance between the current particle and the optimal solution is less than a fixed value, it means that the probability that the particle can obtain the optimal solution is relatively high. In this case, if the speed and position of the particle are updated using formulas (7) and (8), it may fall into the local optimum and deviate from the optimal solution. The present invention uses chaos search to expand the search accuracy of the particle. On the one hand, the amount of chaos search is not large, which can make the algorithm converge faster. On the other hand, even if the parameters of the chaos search are wrong, it will not affect the global search and can still achieve high search accuracy.

[0109] Step B06: Calculate the new fitness value of each particle using the evaluation function based on the particle positions updated in steps B04 and B05.

[0110] Step B07: Update the individual optimal position of each particle and the optimal position of the population of all particles according to the new fitness values ​​of all particles.

[0111] Step B08: Determine whether the iterative convergence condition is met. If so, output the optimal position of the population after iterative convergence. If not, return to step B03 and continue updating particles until the iterative convergence condition is met. In this embodiment of the present invention, the iterative convergence condition is reaching the maximum number of iterations.

[0112] Step B09: Before outputting the final fault section of the source-containing distribution network, constrain the optimal position of the population iteratively converged in step B08 by using pre-set distribution network stable operation constraints.

[0113] When the improved particle swarm optimization algorithm meets the iterative convergence conditions, the power flow calculation of the source distribution network is performed according to the iteratively converged optimal position of the population, and the voltage value and current value corresponding to the iteratively converged optimal position of the population are obtained; it is judged whether the voltage value and current value meet the stable operation conditions of the distribution network. If so, the fault section of the source distribution network is output according to the iteratively converged optimal position of the population. If not, the improved particle swarm optimization algorithm is reused for particle swarm optimization.

[0114] In the present invention, the stable operation conditions of the distribution network include node voltage constraint conditions and branch current constraint conditions.

[0115] (1) The node voltage constraint is:

[0116] U jmin ≤U j ≤U jmax (10)

[0117] Among them, U j Indicates the voltage value of node j in power flow calculation, U jmin and U jmax are the minimum and maximum values ​​of the voltage at node j, where node j represents the jth segment switch node in the source distribution network.

[0118] If the voltage at a certain node cannot meet the upper and lower voltage limits, the voltage needs to be corrected until it meets the voltage constraint conditions. Common methods include: adjusting the active and reactive power of the node, increasing or decreasing the output power of the generator, adjusting the transformer ratio, etc. The adjustment method can be adjusted according to the specific situation of each location.

[0119] U j '=ω1P j +ω2Q j +ω3P G +ω4k T (11)

[0120] Among them, U j ' represents the voltage at node j after correction, P j , Q j represent the active and reactive power at node j respectively; P G is the output active power of the generator; k T is the transformer ratio, and the weights ω1, ω2, ω3, and ω4 can be adjusted according to actual requirements.

[0121] (2) Branch current constraints:

[0122] I jmin ≤I j ≤I jmax (12)

[0123] Among them, I j Indicates the current of branch j in the power flow calculation. Branch j represents the branch in the jth section of the source distribution network. I jmin and I jmax is the minimum and maximum current limit allowed to flow through branch j.

[0124] If the current in a branch exceeds the limit, it needs to be corrected until it meets the current constraint conditions. Common methods include: adjusting active and reactive power, adjusting transformer ratio, adjusting conductor cross-sectional area, adjusting line length, etc.; the specific method can be adjusted according to actual conditions.

[0125] I j '=ω1P j +ω2Q j +ω3L S +ω4k T +ω5L j (13)

[0126] Among them, I j ' represents the corrected current, P j , Q j They represent the active and reactive power at branch j, L S is the cross-sectional area of ​​the conductor, k T is the transformer ratio, L j is the line length weight of branch j, ω1, ω2, ω3, ω4, and ω5 can be adjusted according to actual requirements.

[0127] The above two correction methods need to comprehensively consider the actual requirements of different levels of power grids, comply with various standards for stable operation of the power grid, and ensure the safe and stable operation of the distribution network.

[0128] The method of the present invention can quickly and accurately locate the fault section in the source distribution network based on the fault current information of the source distribution network. Compared with some existing algorithms, the method of the present invention has a fast convergence speed and good convergence performance, and achieves better fault location effect.

[0129] Example 2:

[0130] Based on the source distribution network fault location method introduced in Example 1, this embodiment introduces a source distribution network fault location device for implementing this method, which mainly includes a data acquisition module, a model building module and a fault location module.

[0131] The data acquisition module obtains the fault current information of the source distribution network through the intelligent feeder terminal equipment of the distribution network. The fault current information includes the current information flowing through each node (section switch) in the distribution network.

[0132] The model building module is mainly used to obtain the evaluation function of the source-containing distribution network under the fault state based on the acquired fault current information of the source-containing distribution network. The specific operation of the model building module is consistent with step A in Example 1.

[0133] The fault location module is primarily used to update the velocity and position of particles using an improved particle swarm optimization algorithm to locate faults in the source-containing distribution network and determine the faulted section of the source-containing distribution network. In the improved particle swarm optimization algorithm, the evaluation function is used as the improved particle swarm optimization algorithm's fitness function, and candidate solutions for the faulted section of the source-containing distribution network are used as particles. Adaptive inertia weights are used to update the particle velocity and position. The fault location module is also used to constrain the output of the improved particle swarm optimization algorithm using pre-set distribution network stability constraints.

[0134] The specific operation of the fault location module is consistent with step B in Example 1.

[0135] Example 3:

[0136] This embodiment introduces a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for locating a fault in a power distribution network with a source introduced in Embodiment 1 is implemented.

[0137] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0138] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0139] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0140] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0141] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which are all protected by the present invention.

Claims

1. A method for locating faults in a power distribution network, characterized in that: The steps include: According to the acquired fault current information of the source-containing distribution network, an evaluation function of the source-containing distribution network under the fault state is obtained; The evaluation function is used as the fitness function of the improved particle swarm optimization algorithm, and the candidate solutions of the fault section of the source distribution network are used as particles. The speed and position of the particles are updated by the improved particle swarm optimization algorithm to locate the fault of the source distribution network and obtain the fault section of the source distribution network; The improved particle swarm optimization algorithm uses adaptive inertia weights to update the speed and position of particles; The output result of the improved particle swarm optimization algorithm is constrained by pre-set distribution network stable operation constraint conditions; The adaptive inertia weight includes: According to the fitness values ​​of all particles, the average fitness value of the population at the current moment is obtained ; According to the fitness value of the optimal position of the population at the current moment and the average fitness value Calculate the weight selection ratio of each particle. The expression is as follows: ； in, represents the weight selection ratio of particle i, , M is the total number of particles in the particle swarm; According to the weight selection ratio and the fitness value of each particle, the adaptive inertia weight of each particle is updated. The expression is as follows: ； in, represents the adaptive inertia weight of particle i, Represents the fitness value of particle i.

2. The method for locating faults in a power distribution network according to claim 1, wherein: The method to obtain the evaluation function of the source distribution network under fault state is: According to the positions of the sectionalizer switches in the source-containing distribution network, the source-containing distribution network is divided into sections to obtain a plurality of sections corresponding to the sectionalizer switches one by one; encoding the segment switches and segments; According to the code and the fault current information of the source distribution network, an actual state value and an expected state value of each section switch in the source distribution network under a fault state are obtained; The evaluation function of the source distribution network under fault state is obtained according to the actual state value and expected state value of each section switch.

3. The method for locating faults in a power distribution network according to claim 2, wherein: The expression of the evaluation function is as follows: ； in, is the evaluation function of the distribution network with source, is the state value of the distribution network with source, is the actual state value of the jth section switch in the source distribution network, is the expected state value of the jth section switch in the source distribution network, is the evaluation weight, is the state value of the j-th section feeder in the source distribution network, =1 indicates a feeder fault in the jth section, =0 means the feeder of section j is normal, , N is the total number of section switches in the source distribution network.

4. The method for locating faults in a power distribution network according to claim 2, wherein: According to the fault current information of the source distribution network, the actual state value of each section switch is obtained. The expression is: 。 5. The method for locating faults in a power distribution network according to claim 1, wherein: The improved particle swarm optimization algorithm is used to update the velocity and position of particles to locate the fault in the distribution network containing the source, and the fault section of the distribution network containing the source is obtained, including: Initializing network parameters and particle swarm parameters, wherein the particle swarm parameters include the position and velocity of particles; Calculating the fitness value of each particle according to the evaluation function and the position of the particle; According to the fitness values ​​of all particles, the individual optimal position of each particle and the optimal position of the population of all particles are obtained; Update the adaptive inertia weight of each particle according to the fitness value of all particles; updating the velocity and position of each particle according to the adaptive inertia weight; According to the updated particle position, the new fitness value of each particle is calculated using the evaluation function; According to the new fitness values ​​of all particles, update the individual optimal position of each particle and the optimal position of the population of all particles; When the iterative convergence conditions are met, the optimal position of the iteratively converged population is output and the fault section of the distribution network containing the source is obtained.

6. The method for locating faults in a power distribution network according to claim 5, wherein: In the improved particle swarm optimization algorithm, when the distance between the position of particle i and the optimal position of the population at the current moment is less than the preset threshold, the position of particle i is updated using chaotic search. , M is the total number of particles in the particle swarm.

7. The method for locating faults in a power distribution network according to claim 1, wherein: The stable operation conditions of the distribution network include node voltage constraints and branch current constraints; When the improved particle swarm optimization algorithm meets the iterative convergence conditions, the power flow calculation of the source distribution network is performed according to the iteratively converged optimal position of the population, and the voltage value and current value corresponding to the iteratively converged optimal position of the population are obtained; it is judged whether the voltage value and current value meet the stable operation conditions of the distribution network. If so, the fault section of the source distribution network is output according to the iteratively converged optimal position of the population. If not, the improved particle swarm optimization algorithm is reused for particle swarm optimization.

8. A fault location device for a power distribution network containing a source, characterized in that: include: Data acquisition module, used to obtain fault current information of the source distribution network; A model building module is used to obtain an evaluation function of the source-containing distribution network under a fault state based on the acquired fault current information of the source-containing distribution network; a fault location module, configured to utilize the evaluation function as a fitness function of an improved particle swarm optimization algorithm, utilize candidate solutions of the fault section of the source distribution network as particles, update the velocity and position of the particles through the improved particle swarm optimization algorithm, perform fault location on the source distribution network, and obtain the fault section of the source distribution network; In the fault location module, the improved particle swarm optimization algorithm uses adaptive inertia weights to update the speed and position of particles; The fault location module is further configured to constrain the output result of the improved particle swarm optimization algorithm by using preset distribution network stable operation constraint conditions; The adaptive inertia weight includes: According to the fitness values ​​of all particles, the average fitness value of the population at the current moment is obtained ; According to the fitness value of the optimal position of the population at the current moment and the average fitness value Calculate the weight selection ratio of each particle. The expression is as follows: ； in, represents the weight selection ratio of particle i, , M is the total number of particles in the particle swarm; According to the weight selection ratio and the fitness value of each particle, the adaptive inertia weight of each particle is updated. The expression is as follows: ； in, represents the adaptive inertia weight of particle i, Represents the fitness value of particle i.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for locating a fault in a power distribution network containing a source as claimed in any one of claims 1 to 7 is implemented.

Images