Research On Fault Location Of Distribution Network Based On Improved Immune Network And Its Algorithm

According to the latest requirements of the "14th Five-Year Plan" and the 2035 Vision Outline,we should speed up the development of clean energy,persist in the development of distributed power,and steadily advance the construction of coastal nuclear power,so as to achieve the goal of green transformation of production and lifestyle.The security of the distribution network is an important factor in maintaining the reliability of the power system.In the case of the steady development of the power system and the continuous growth of its scale,it is particularly important to ensure the normal operation of the distribution network.The access of distributed power sources has brought tremendous changes to the distribution network architecture,also made traditional distribution network fault detection and processing methods no longer applicable.Power outages due to errors in the distribution network can let to a big effect on the production and life of the society.Therefore,it is meaningful for engineering to locate faults timely and precisely.The application result of diverse basic intelligent algorithms in simulation are compared and analyzed,and a new improved immune network model is proposed based on the immune algorithm.Firstly,a comparative experiment was carried out on various algorithms to analyze the characteristics and operating effects of various algorithms.Based on the phenomenon that the population tends to become singular from the iteration of the immune algorithm to the later stage,combined with the analysis of the failure location data in the experimental results,a memory cell information differentiation mechanism is proposed.This way can availably ensure the multiformity of the population as well as promote the local search capacity of the algorithm,and it can also lessen the likelihood of falling into the local optima in the iterative process.Secondly,based on the traits of immune algorithm and the specialty of particle swarm algorithm,a modified method is designed with the information differentiation mechanism of memory cells.This way uses the advantage of particle swarm algorithm that can quickly find the best point,so that let it to participate in updating the memory information of the immune algorithm,effectively avoiding the unstable result of a single particle swarm algorithm.The improved immune algorithm performs well in simulated fault experiments such as singular failure,multiple failures and information distortion,and can accurately locate the fault occurrence section even in extremely complex fault conditions.Due to the improved immune algorithm adopts two parallel iterative methods of algorithms,which leads to a longer running time of the algorithm,a multi-layer fault location model is designed on this basis.First,the distribution network model is reduced for the first time according to the access of distributed power sources,and the reduced-dimensional distribution network model is simplified again according to the principle of equivalent nodes.The improved immune algorithm is used to locate the fault of the simplified distribution network model,locate the area where the fault occurs,and then perform the second fault location of its internal nodes based on the information of the fault area.Experimental results show that adopting a hierarchical fault location model can significantly improve the iteration speed of the algorithm while ensuring the accuracy of fault location.