A power distribution network fault location method based on improved ant colony algorithm

Abstract

The application relates to a power distribution network fault positioning method based on an improved ant colony algorithm, which comprises the following steps: a feeder automatic unit acquires current information, and a main station analyzes switch overcurrent information; all switches are regionally divided; a switch function and an evaluation function in the algorithm are constructed; according to the fault information after the regional division, the improved ant colony algorithm is adopted to perform fault positioning, and a positioning result is output. The application has the advantages that: the cross variation operation is introduced in the power distribution network fault positioning for the first time, the limitation that the traditional ant colony algorithm is prone to falling into local optimization in fault positioning is overcome, the characteristics of the genetic algorithm parallel operation are adopted, the limitation that the traditional ant colony algorithm converges slowly in the initial stage in fault positioning is overcome, the speed and accuracy of fault positioning are improved, and the response time is shortened.

Description

Technical Field

[0001] This invention relates to the field of distribution network automation, specifically to a method for fault location in distribution networks based on an improved ant colony algorithm. Background Technology

[0002] Users have increasingly higher requirements for power distribution networks in terms of power supply safety and quality. The safety, reliability, flexibility, and economy of distribution networks directly impact the normal production and daily life of enterprises. Once a distribution network fails, it can cause serious social losses and greatly disrupt people's lives. Therefore, research on high-level automatic fault location in distribution networks has significant social and economic benefits.

[0003] With the development of distribution network automation in my country, the amount of fault transmission information is increasing: electrical quantity information, switch information, protection information, etc. How to effectively utilize this information to achieve rapid and efficient fault location, and maximize the potential of the distribution network, is the trend in distribution network automation development. Existing distribution network fault location methods all have their own advantages and disadvantages, requiring a site-specific approach and appropriate measures based on local conditions.

[0004] Ant colony optimization (ACO) for fault location primarily transforms the fault location problem into a Traveling Salesman Problem (TSP). It utilizes ACO for local and global optimization to determine the faulty section. While it offers good positive feedback and fault tolerance, it suffers from challenges in constructing the evaluation function, an excessive number of demand species, complex internal parameter settings, and a large number of iterations. It is suitable for distribution networks with simple structures and automatic feeder units. Summary of the Invention

[0005] To address the above issues and overcome the shortcomings of traditional ant colony algorithms in fault location, this solution provides a fault location method for distribution networks based on an improved ant colony algorithm. This solution aims to solve the problems of long search times and low efficiency in existing technologies for fault location in distribution networks, and provides an improved ant colony algorithm to overcome the limitations of traditional ant colony algorithms in solving distribution network fault location problems.

[0006] The technical solution adopted in this invention is: a distribution network fault location method based on an improved ant colony algorithm, comprising the following steps:

[0007] Step 1: The automatic feeder unit acquires and reports current information, and the main station summarizes and analyzes the switch overcurrent information;

[0008] Step 2: Divide all switches into zones, input the basic information of the distribution network, such as ring network, branch marker switches, tie switches, etc., and retain the areas that need to be located for fault according to the distribution network type;

[0009] Step 3: Construct the switching function and evaluation function in the algorithm;

[0010] Step 4: Use a genetic algorithm to obtain a better solution and form an initial pheromone matrix accordingly;

[0011] Step 5: Based on the fault information after region division and the initial pheromone, the improved ant colony algorithm is used to locate the fault and output the location result.

[0012] Furthermore, the method of obtaining faulty lines using the improved ant colony algorithm is as follows: First, generate m ants that pass through the paths selected by each switch and record them to form an iterative population. Randomly select individuals for crossover and mutation operations. Then, these m ants simultaneously use the constructed evaluation function to calculate the evaluation value of the switch information after the region is divided. Use the roulette wheel method to select individuals and record individuals with low evaluation values. If the termination condition is not met, continue iterating for the next generation of the population until the termination condition is met. The solution with the smallest evaluation value is the optimal solution, where the part with a value of 1 is the faulty line.

[0013] Furthermore, the initialization variables required by the algorithm include population size, maximum number of generations, mutation probability, pheromone evaporation coefficient, number of elite ants, and maximum number of iterations.

[0014] Furthermore, the termination condition is that the number of iterations reaches the set maximum number of iterations, wherein the maximum number of iterations and the population size are determined according to the complexity of the distribution network, and the population size is approximately 1.5 times the number of nodes in the distribution network.

[0015] In summary, compared with the prior art, the above-described technical solutions conceived by this invention can achieve the following beneficial effects:

[0016] The improved ant colony algorithm draws on the parallel computing features of the genetic algorithm, simultaneously calculating the evaluation function and updating the pheromone for m ants, thus shortening the running time, improving the speed of fault location, and reducing the response time. The improved ant colony algorithm introduces the crossover and mutation operation of the genetic algorithm, improving the randomness of the ant colony search. Therefore, it can solve the problem that the traditional ant colony algorithm may get stuck in local optima in fault location, thus improving the accuracy of fault location. Attached Figure Description

[0017] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0018] Figure 1 This is a flowchart of the improved ant colony algorithm.

[0019] Figure 2 This is a schematic diagram of the system architecture corresponding to this solution;

[0020] Figure 3This diagram illustrates a typical application example of a medium- or low-voltage single-source power distribution network in this solution.

[0021] Figure 4 This diagram illustrates a typical application example of a medium- and low-voltage power distribution network including distributed power sources, as presented in this solution.

[0022] Figure 5 This is a location result diagram for a typical medium- and low-voltage single-source power distribution network application of this scheme;

[0023] Figure 6 This is a location result diagram of a typical medium- and low-voltage power distribution network application including distributed power sources in this scheme. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0025] like Figure 1 As shown, this embodiment provides a distribution network fault location method based on an improved ant colony algorithm. Using current over-limit signals as a basis, the improved ant colony algorithm is used for distribution network fault location, including the following steps:

[0026] Step 1: The automatic feeder unit acquires and reports current information, and the main station summarizes and analyzes the switch overcurrent information;

[0027] Step 2: Divide all switches into zones, input the basic information of the distribution network, such as ring network, branch marker switches, tie switches, etc., and retain the areas that need to be located for fault according to the distribution network type;

[0028] Step 3: Construct the switching function and evaluation function in the algorithm;

[0029] Step 4: Use a genetic algorithm to obtain a better solution and form an initial pheromone matrix accordingly;

[0030] Step 5: Based on the fault information after region division and the initial pheromone, the improved ant colony algorithm is used to locate the fault and output the location result.

[0031] This embodiment uses an improved ant colony algorithm for fault location. However, the improved ant colony algorithm cannot directly manipulate parameters, so the power grid fault location problem must be encoded with 0-1, where overcurrent or line fault is represented as 1, and no overcurrent or normal line is represented as 0.

[0032] An improved ant colony algorithm is used for fault location. The switching function reflects the desired state of the switch. Its role is to transform the line state into whether the switch has fault current passing through it. Then, the fault current data of the switch is analyzed by the evaluation function to find the actual faulty line.

[0033] In step 1, for a single-source distribution network, the power source is used as a reference power source; for a multi-source distribution network, it is necessary to determine the upstream and downstream power sources of the switch, and use the upstream power source as the reference power source. The output power direction of the reference power source is set to the positive direction, and a directional power element is used to detect the current direction. The overcurrent information matrix is ​​then calculated according to the following formula.

[0034]

[0035] In the formula, I j I represents the current value flowing through switch j. set This indicates the preset current value for switch j.

[0036] In step 2, when dealing with a distribution network with many branches, a regional division approach can be adopted to reduce the computational dimensions.

[0037] In the single-source radial network, the branch with the most nodes in the simplified model is considered the main region, and the remaining lines are sub-regions. The switch closest to the main branch in each sub-region is used as the branch detection switch. It is marked as 1 when there is overcurrent information in the branch, and marked as 0 when there is no overcurrent information.

[0038] For a distribution network operating in open-loop mode with multiple power sources, the network is divided into several independent zones using tie switches as boundaries. Each zone is a complete single-power-source radial network, and each zone is identified by the switch closest to the power source. Similarly, a zone is marked as 1 when overcurrent information exists and as 0 when no overcurrent information exists.

[0039] For distribution networks containing distributed power sources, all branches containing power sources are treated as main regions in the simplified model, and the remaining lines are sub-regions. A branch is marked as 1 when it has overcurrent information and as 0 when it does not.

[0040] If the branch detection switch is marked as 0, the area is shielded and not included in the positioning calculation.

[0041] In step 3, the mathematical expression for the switching function is common for both single-source and multi-source applications. The switching function is defined as follows:

[0042]

[0043] In the formula, K represents the calculated result of the switching function for switch j, indicating the desired state of the switch; DGiThis indicates the situation where distributed generation is connected to the power grid. If DGi is connected to the power grid, it is 1; otherwise, it is 0. ju S is the union of fault states of the downstream lines of the switch; jd It is the union of the fault states of the upstream line of the switch.

[0044] In step 3, the evaluation function is defined as:

[0045]

[0046] Where N represents the number of segments; represents the expected value of the switch; w represents the weighting coefficient, usually taken as 0.5; S k This represents the fault information of the k-th switch of the input individual.

[0047] In step 5, the improved ant colony algorithm is used to obtain the faulty line as follows: First, m ants are generated and their paths selected by each switch are recorded to form an iterative population. Individuals are randomly selected for crossover and mutation. Then, these m ants simultaneously use the constructed evaluation function to calculate the evaluation value of the switch information after the region is divided. Individuals are selected using the roulette wheel method, and individuals with low evaluation values ​​are recorded. If the termination condition is not met, the next generation of the population continues to iterate until the termination condition is met. The solution with the smallest evaluation value is the optimal solution, and the part with a value of 1 is the faulty line.

[0048] like Figure 2 The diagram shows the system structure for fault location in this embodiment. The feeder automatic unit (FTU) collects current information and sends the collected information to the substation via the communication network. The substation then sends the information to the Supervisory Control and Data Center (SCADA), which is the master station. At the same time, the feeder terminal at the switch can also receive opening and closing commands from the SCADA. After the control center locates the faulty line, it issues opening and closing commands to each feeder terminal to isolate the faulty area or restore the non-faulty area.

[0049] like Figure 3 The diagram shows a simplified distribution network obtained after regional division when a short-circuit fault occurs on line L22 in a single-source distribution network. In the diagram, A represents the power source, and S... k For segmented switches, L i This is a simplified circuit design for a power distribution network. Based on actual current conditions, an improved ant colony algorithm was simulated using MATLAB, and the results are as follows. Figure 5 As shown, the fault area can be accurately located.

[0050] like Figure 4 The diagram shows a simplified distribution network obtained after regional division when a short-circuit fault occurs on line L3 in a distribution network containing distributed power sources. In the diagram, A represents the power source, and S... k For segmented switches, L iThis is a simplified circuit design for a power distribution network. Based on actual current conditions, an improved ant colony algorithm was simulated using MATLAB, and the results are as follows. Figure 6 As shown, the fault area can be accurately located.

[0051] In summary, this invention provides a distribution network fault location method based on an improved ant colony algorithm. When a distribution network fault occurs, the automatic feeder unit collects the fault current signal and uploads it to the master station. The master station then uses the improved ant colony algorithm described in this invention to quickly locate the fault. This invention improves the speed and accuracy of fault location, shortens response time, and reduces labor costs without requiring additional hardware investment.

[0052] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for fault location in a distribution network based on an improved ant colony algorithm, characterized in that, Includes the following steps: Step 1: The automatic feeder unit acquires and reports current information, and the main station summarizes and analyzes the switch overcurrent information; Step 2: Based on the distribution network topology, input the basic information of the distribution network, such as ring networks, branch marker switches, tie switches, etc., and dynamically divide the fault location area, retaining the area that needs to be located. Specifically, if it is a single-source distribution network, the branch with the most nodes is the main area, and the rest are sub-areas; if it is a multi-source ring network, the distribution network is divided into several independent areas with tie switches as the boundary; if it is a distribution network with distributed power sources, all branches with power sources are the main areas, and the rest are sub-areas. Step 3: Construct the switching function and evaluation function in the algorithm; Step 4: Use a genetic algorithm to obtain a better solution and form an initial pheromone matrix accordingly; Step 5: Based on the fault information and initial pheromone after region division, an improved ant colony algorithm is used for fault location. Specifically, this includes: generating m ants that traverse the paths selected by each switch and recording them to form an iterative population; randomly selecting individuals for crossover and mutation operations; using a pre-constructed evaluation function to calculate the evaluation value of the switch information after region division; using roulette wheel selection to select individuals and update pheromones; recording individuals with low evaluation values; if the termination condition is not met, continuing the iteration, and finally outputting the solution with the smallest evaluation value as the result of fault line location, where the part with a value of 1 is the fault line.

2. The method for fault location in a distribution network based on an improved ant colony algorithm according to claim 1, characterized in that, The algorithm requires initialization variables including population size, mutation probability, pheromone evaporation coefficient, number of elite ants, and maximum number of iterations.

3. The method for fault location in a distribution network based on an improved ant colony algorithm according to claim 1, characterized in that, The termination condition for the improved ant colony algorithm in step 5 is that the number of iterations reaches the preset maximum number of iterations, and the population size is 1.5 times the number of distribution network nodes.

Images