A Study On The Predicting Model Of Gray Neutral Network And Support Vector Machine

Prediction is necessary in modern times for us to master the changes in future, which is based on the study of research data and statistic data to discover the regulation of things' development and changes.Combining several predicting methods as a whole would let the model uses the advantages and avoids the disadvantages of each method,this will increasing the model's predicting accuracy comparing to a single predicting method.This paper introduced the theory and modeling process of gray neutral network and support vector machine(SVM).Through the comparison and analysis on gray system vs.neutral network and SVM vs.neutral network,we confirmed the possibility of integrating the above three predicting model as a whole.Two new predicting models were proposed in this paper.The first one is a gray compensating neutral network based on structural risk(GRBFNN).The core concept of it was to use the structural risk minimization principle into the modeling process of rbf neutral network so that the rbf function center can be derived by calculate the support vectors of the model.And then used the rbf neutral network based on structural risk to compensating GM(1,1) model.The second one is a gray support vector machine based on gray relational grades(GSVM).The core concept of it was to use the gray relational grades(GRA) to extract the main factors of model.And then took these main factors as the input factors to establish a GSVM model.Applying the two models into a wind path response system and a ultimate bearing capacity of bridge pile predicting system respectively.The experiment results indicated the increasing of predicting accuracy and generalization ability of the two new models.