The Power System Fault Diagnosis Based On DTW And Modified Artificial Fish Swarm Algorithm

With the continuous improvement of the power system interconnection,if the fault elements in the power grid fail to be removed in time,a large-scale blackout will happen.It is very necessary to identify fault components timely and effectively and remove fault components.The existing fault diagnosis method based on switch information has high diagnostic accuracy when the information of the protection or circuit breaker is complete,and the diagnosis results will be affected when the information of the protection and circuit breaker is incomplete or when the fault is incorrect.The electrical information has natural advantages compared with the switch information,and the information contained in it is more abundant.The research of power grid fault diagnosis based on electrical information has attracted the attention of the industry.In this paper,a fault diagnosis method based on DTW(Dynamic Time Warping)algorithm and artificial fish clustering algorithm is proposed,and the fault line is diagnosed by using this method.By using the switch information,the analytic model of power system fault diagnosis is set up,and the improved artificial fish swarm algorithm is used to find the optimal solution of the analytical model,and it is of great theoretical significance to identify the fault element.A power grid fault diagnosis method based on DTW and artificial fish swarm clustering algorithm is proposed.The current sampling value of the suspected line in the power outage area is first obtained from recorded data.According to the suspected lines,each suspected line in the power outage area is compared with the comprehensive current of a reference line outside the outage area,and the DTW algorithm is used to obtain the transverse difference values of the suspected fault lines.For each suspected line,the longitudinal difference degree of the line is obtained by using the DTW algorithm to calculate the longitudinal difference degree of the three weeks before and after the fault.The differential value of the horizontal and longitudinal difference of each line is sent into the artificial fish clustering algorithm,which is divided into fault class and normal class,and the line in the fault class is diagnosed as fault line.Simulation experiments on IEEE39 bus system show that the algorithm can accurately identify fault components and is not affected by fault location,transition resistance and the fault type.In this paper,an improved artificial fish swarm algorithm is applied to analyze the power grid fault diagnosis model.Compared with the basic artificial fish swarm algorithm,the improved artificial fish swarm algorithm is added to the mutation factor,which reduces the probability of the algorithm to stay in the local optimal position,has a stronger ability to jump out of the local extremum,overcomes the problem of multiple solutions when the fault element is more.The actual action vector of the protection and circuit breaker is determined through the collection of the action information of the protection and circuit breaker after the fault occurred.The desired action state vectors are given by the suspected elements in the power outage area according to the action rule,and the actual action direction and the desired action state vector are brought into the analytical model of the fault diagnosis of the power grid.The improved artificial fish swarm algorithm is used to optimize the fault components.The typical fault instances of test system are simulated to verify the effectiveness of the method.