Intelligent Geological Subsidence Prediction Based on PSO-RBF Neural Network

Land subsidence is a threat to the safe and stable operation of transmission lines, and it is very likely to lead to the movement and deformation of transmission line tower foundations. In this paper, a particle swarm optimization radial basis function neural network (PSO-RBF) method is used to predict the geological settlement. For RBF neural network hidden layer center, width varies according to different applications, PSO-RBF uses particle swarm algorithm to find the optimal parameters, optimize the RBF neural network mechanism and the global convergence of the algorithm, and then establish the PSO-RBF neural network prediction method for land subsidence. This paper verifies the applicability of pso-rbf neural network in surface subsidence and compares the results with the traditional BP neural network, RBF neural network, and PSO-BP neural network pairs for predicting land subsidence, and the results show that the prediction results perform better than the other three models in the application of the land subsidence prediction case.