Stability Classification of Surrounding Rock Based on Support Vector Machine Classification Theory

Abstract

The classification of rock surrounding has an important significance for guiding design and construction of underground engineering. Past classification systems have not successfully been used because of complex classification index and much unascertained information.In order to better define stability classification of surrounding rock, support vector machine classification(SVM) theory is used. SVM is a good diagnosis method for small sample stability classification of surrounding rock system, which overcomes some deficiencies of traditional methods. Surrounding rock classification data collected are defined as training and forecasting samples, which can classify stability type surrounding rock. Therefore, the results show that the SVM model can be used to predict the classification of underground engineering surrounding rock, and the prediction accuracy is reliable and feasible.. It is a new way to be used for predicting the classification of surrounding rock of underground engineering.