Abstract: In order to solve the problem of insufficient accuracy of transmission line grounding resistance value,an improved nonlinear echo state network based on particle swarm optimization is proposed to predict and analyze the grounding resistance of transmission line.Firstly,by introducing the readout layer,the standard echo state network is improved to obtain the nonlinear echo state network（NESN）. Secondly,the PSO algorithm is added to the dynamic inertia weight,dynamic learning factor and mutation disturbance to obtain the improved particle swarm optimization algorithm（IPSO）,so as to improve the population diversity and enhance the global search ability and local development ability of the population. Finally,a prediction model based on IPSO-NESN is established to optimize the parameters of NESN through the IPSO algorithm. The results show that IPSO-NESN has better prediction performance than ESN and PSO-NESN models,can be applied to predict the grounding resistance of transmission lines.