Simulation Study Of Short-circuit Fault Location In Ship’s Active Distribution Network Based On Optimization Algorithms

In today’s world environment of energy saving and emission reduction,the shipping sector in various countries is constantly expanding the application of green energy on ships.Distributed generations are constantly being researched,developed and exploited for their advantages of green,sustainable,low cost and convenient energy storage.A ring-radial network structure with distributed generations consisting of main distribution boards and power stations is called a ship’s active distribution network.Aiming at the problems of slow convergence,long localization time,low localization accuracy and poor population quality of existing artificial intelligence optimization algorithms in solving the fault localization problem of ship’s active power distribution network,this thesis proposes various fault localization methods for ship’s active power distribution network based on improved new intelligent optimization algorithms.Firstly,the research background,research significance and current status of the topic are described,and the impact of distributed generations connected to the ship’s distribution network on traditional fault location methods is introduced.According to the bi-directional current characteristics of the short-circuit current when a fault occurs in the ship’s active distribution network,the coding method and switching function of the distribution system are defined differently from the traditional methods,and the evaluation function of the analytic model with fault-tolerant performance is proposed as an iterative basis for improving the optimization algorithm.Secondly,based on the update mechanism and theory of various new intelligent optimization algorithms,the coding method,main parameters and population update process are designed and improved,and the performance of the improved algorithms is tested to verify their effectiveness and practicality.(1)Based on the mechanism of quantum genetic algorithm: introducing quantum computing theory into genetic algorithm,adopting double-chain quantum bit encoding for the algorithm population;dynamically adjusting the rotation angle size of the quantum rotation gate according to the difference between the global optimal individual fitness value and the current individual fitness value,and using quantum non-gates for chromosome variation operation,which accelerates the convergence speed and enhances the global search ability of the improved algorithm;(2)Based on the mechanism of multiverse optimization algorithm: discrete encoding of the universe population;secondly,incorporating the adaptive elite strategy into the multiverse population iteration of the improved algorithm;designing the update mechanism of nonlinearly varying wormhole existence probability(WEP)and travel distance rate(TDR)to improve the ability of the algorithm to search for the optimal universe in the early stage and adjust the accuracy of the optimal detection distance in the later stage;finally,enhancing the local search ability of the improved algorithm by adaptive mutation operations to enhance the local search capability of the improved algorithm;(3)Based on the mechanism of slime bacteria optimization algorithm: according to the discrete domain characteristics of fault location problem,binary encoding of slime bacteria population,chaotic mapping of initial population using Chebyshev theory,and improving the behavior and morphological change process during slime bacteria searching for food,incorporating bit operation update mechanism to retain valuable information,improving population quality,and enhancing the ability of the algorithm to jump out of the local optimal solution and global search.Finally,through the simulation experiments of locating single point,combination and information distortion fault cases in ship’s active distribution network by three improved algorithms proposed based on the mechanism of intelligent optimization algorithm,combined with traditional algorithms,the algorithm performance is evaluated based on the algorithm convergence,locating time,locating accuracy and population quality situation as the evaluation criteria to verify the significant performance advantages of the improved intelligent optimization algorithm in the fault location problem.