Forecasting Slope Displacement of the Agricultural Mountainous Area Based on the ACO-SVM Model

Abstract

Due to the combined influence of complex engineering geological conditions and environmental factors from agricultural mountainous areas, the evolution of slope deformation is complicated and nonlinear. Support vector machine (SVM) technology could effectively solve the technical problems of small sample, high dimension, and nonlinear, so it is applied to data mining of the measured slope displacement and the prediction and analysis of the slope deformation trend. In order to avoid blindness of human choice of SVM parameters and to improve the prediction accuracy and generalization ability of the model, an ACO-SVM model is built by adopting an improved ant colony algorithm (ACO) to optimize parameters in association with the rolling forecasting method of displacement time series. The model was applied to two engineering examples. The research results showed that the ACO-SVM model was correct with high accuracy. The ACO-SVM model had higher accuracy of prediction and stronger generalization ability than optimizing SVM based on the genetic algorithm or particle swarm optimization. The forecasting results were more reasonable. It has certain engineering application values for slope deformation prediction.