Study On Power Distribution System Fault Location Based On Intelligent Algorithm

With the continuous upgrading of smart grid,China puts forward higher requirements for the quality and reliability of power supply,therefore,timely and accurate fault location and solution are necessary aiming at reducing economic loss when the grid distribution network fails.From this perspective of view,efficient and reliable fault localization technique plays an extremely important role in the safe operation of smart grid distribution network.At present,the fault location techniques on FTU based smart grid distribution network are mainly categorized into direct approaches and indirect approaches,and the indirect methods incorporating metaheuristics of computational intelligence become hotspots of fault location research in smart grid due to its general-purpose characteristic,quick-solution within tolerance and good performance with robustness.However,the intelligent algorithms in these indirect approaches to fault location of smart grid have their inborn defects,which inevitably leads to low convergence speed and bad accuracy.Aiming at tackling these problems,the thesis proposed three intelligent algorithm based fault location methods including a novel Particle Swarm Optimization(PSO)based fault location algorithm,a novel Differential Evolution based fault location algorithm and a novel QUasi-Affine TRansformation Evolution(QUATRE)based fault location algorithm in smart grid.These three methods were verified under both the universal test suites for meta-heuristic algorithm validation and the fault location mode in smart grid.The main highlights of my work are listed as follows:(1)Firstly,the fault location models of radial distribution network and distribution network with distributed generation were constructed.The thesis analyzed the mechanism of overcurrent when phase to phase fault occurred in distribution network,and considered the situation that a large number of distributed generators were incorporated into the distribution network,then,the fault location models of radial distribution network and distribution network with distributed generators were constructed respectively.By adjusting the coding style of the solutions and their corresponding objectives,the fault location problem of the distribution network in smart grid can be transformed into an optimization problem.After the modeling,the thesis also conducted a deep investigation of the feasibility of the solutions.(2)Secondly,three intelligent algorithm based fault localization techniques were proposed.In order to tackle the low convergence and bad accuracy of the existing PSO variants,a novel adaptive strategy of parameters and the fully-informed search scheme were advanced in the novel PSO variant,namely,PSO-sono algorithm.In order to tackle the poor population diversity and slow convergence speed in existing DE variants,a cooperative mutation strategy and adaptive parameter control were proposed in the novel DE variant,namely CS-DE algorithm.In order to enhance the exploration capacity and optimization accuracy,an adaptive strategy of evolution matrix and the corresponding parameter control were introduced in the novel QUATRE algorithm,namely,E-QUATRE.(3)Finally,validations of the three algorithms were conducted not only under the universal test suites but also under the fault location model of distribution network in smart grid on Matlab platform.The experiment results showed that the proposed algorithms were competitive with their state-of-the-art variants under the universal testbeds.Meanwhile,excellent performance can be secured under the fault location model as well,the PSO-sono based technique ranked first,the E-QUATRE based technique ranked second and CS-DE based technique ranked third.In summary,the three intelligent algorithms proposed in the thesis can improve the efficiency and accuracy of the existing intelligent algorithm based fault location approaches in smart grid,therefore,the quality and reliability of power supply in our model can be secured by employing these techniques.