Parameter inversion of probability integral method based on robust PSO algorithm

In order to solve the problem that the predicted parameters of Particle Swarm Optimization (PSO) algorithm are not accurate due to the interference of gross error data when retrieving the parameters of probability integral method, this paper proposes to invert the parameters of probability integral method by combining the robust estimation and PSO algorithm. In order to reduce the influence of gross error data on parameter inversion results, a weighted fitness function model was constructed by means of the fitting error between the predicted subsidence value and the measured value, and the weight of each observation point was determined by IGGⅢ scheme. The measured data and simulation experiment data are used to verify the anti-gross error ability of the method and the accuracy of the inversion results. The experimental results show that the influence of the gross error data on the inversion results is effectively eliminated when the robust PSO algorithm is used for parameter inversion, and the mining subsidence prediction results are more robust and accurate.