Slope Stability Prediction Method Based on Improved pso-rbf Algorithm

Abstract

Slope Stability Prediction Method Based on Improved PSO‑RBF Algorithm. The invention discloses a slope stability prediction model based on an improved PSO-RBF algorithm, and belongs to the technical field of slope stability prediction. The present invention initializes the radial basis neural network, adopts the particle swarm optimization algorithm based on the normal decay inertia weight factor to optimize the parameters of the radial basis neural network, and constructs a new model based on the optimal parameters of the radial basis neural network calculated according to the optimization algorithm. The prediction model of radial basis neural network and the prediction model of radial basis neural network are used to predict the slope stability. On the basis of the Gaussian function used in the traditional hidden layer, the present invention adds a radial basis function expansion speed control factor, which can adjust the variation trend of the parameters of the neural network in the iterative process and avoid sudden changes in the iterative process. Make the prediction accuracy of the trained model higher.

Description

technical field [0001] The invention belongs to the technical field of slope stability prediction, in particular to a slope stability prediction method based on an improved PSO-RBF algorithm. Background technique [0002] In recent years, with the continuous emergence of extreme weather events around the world, natural disasters have occurred frequently, among which landslides, mudslides and other disasters caused by slope instability have caused the loss of lives and property of many people. Therefore, it is very important to be able to effectively predict the stability of slopes. [0003] The prediction of slope stability is mainly to determine the combination of factors favorable to landslide action through the analysis of landslide conditions, and predict the possibility of landslides in the future on the region or in a certain section of slope according to the combination of these favorable factors. There are four types of slope stability evaluation methods commonly us...

AI Technical Summary

Problems solved by technology

[0004] Although compared with the original theory, there are still some problems in the PSO algorithm, such as easy to fall into local optimum, slow convergence speed in the later period, and low accuracy of the optimization results of the algorithm; the RBF neural network also has improper parameter settings that will lead to problems in the training model. problems such as underfitting or overfitting

At the same time, because there are many factors affecting the slope stability, if the relevant parameters and quantities cannot be reasonably selected, it is impossible to accurately predict

Images