Research on Support vector machine prediction on surrounding rock deformation based on fuzzy information granulation

On the basis of our realization to the world, granulation is one of the basis concepts. It refers to the whole divides into the parts. Information granulation has a key role in many methods and technical domains. The theory of fuzzy information granulation (TFIG) is elicited by man´s information granulation method and based this to infer. Support vector machine (SVM) is an effective method to predict the time series. This article studies the prediction of surrounding rock deformation by using SVM method based on TFIG. Combining with a project case, the algorithm is realized by Matlab program. The results show that this new method can well and truly predict the bound of the next five day´s displacement of surrounding rock.