| In order to ensure the smooth operation of the distribution network system,we need to quickly determine the fault range of the distribution network and accurately locate it.The development of Wireless Sensor Network(WSN)provides a new direction for fault location of distribution network.Because it has the advantages of automatic networking,low energy consumption and high reliability,it adapts to the complex network architecture of the distribution network.Therefore,it can accurately locate the fault area and improve the efficiency of distribution network location fault detection.Aiming at the problem of distribution network fault location,this essay designs and develops a distribution network fault location simulation system based on Distance Vector-Hop(DV-Hop)algorithm,which integrates a variety of efficient and high-precision algorithms to make the distribution network fault location detection more rapid and visual.In addition,the system can effectively reduce the human and material costs caused by actual hardware testing.The main work of this essay will be described in detail below.First,to solve the problem of low accuracy of sensor node fault location in WSN,we propose a heuristic optimization algorithm DECHDV-Hop algorithm combining elite selection and adaptive strategy.The algorithm uses different weighting factors to modify the average jump distance of anchor nodes,and uses the elite selection operator to enhance the optimization ability of the algorithm for the optimal individuals,and then designs a crossover mutation strategy to increase the diversity of population individuals,to prevent falling into the local optimum in the later iterative optimization.Finally,the Simulated Annealing Algorithm(SA)is used to find the optimal solution and obtain the estimated coordinates.Through simulation and analysis,it is found that the localization error of unknown nodes of our proposed algorithm is effectively reduced compared with DV-Hop and CPDV-Hop localization algorithms,and in the generation optimization simulation,DECHDV-Hop algorithm converges faster and has stable optimization ability.Secondly,in view of the more complex terrain environment in the actual application scenario,we propose a DV-Hop localization algorithm based on the chaotic adaptive Fruit Fly Algorithm(FOA)based on the Anisotropic Network(AN).The algorithm transforms the calculation of node localization into a mathematical optimization problem.First,the candidate solution is changed to increase the scope of individual optimization of fruit fly melanogaster to avoid falling into local optimization.Then the search step of FOA is improved to enhance the ability of individual local search.After completing an iterative optimization,the chaotic strategy is used to update the individual position and increase the population diversity.After simulation,compared with other comparison algorithms,our proposed algorithm constructs a diverse elite pool and effectively improves the localization accuracy in AN.When the anchor ratio is 20%and the node communication radius is 200 meters,the average localization error of CAFOA-DV-Hop in the H-shaped topology is about 0.29 R.Similar results can be seen in the W-shaped network,with an average error of 0.27 R.The precision and efficiency of the O-shaped network are even better,and the average error is almost one third of DV-Hop(0.21R: 1.09R).Finally,based on two kinds of localization algorithms and models related to the distribution network,the distribution network fault location system is developed using MATLAB GUI.The system includes login interface,WSN-based localization operation module and AN-based localization module,and also carries out a comprehensive functional test of the system.The test results show that the designed functions of the distribution network fault location system have achieved the expected goals,and can achieve efficient and rapid distribution network fault location,and the interface is clear and easy to operate. |
PDF Full Text Request