Application and comparison of prediction models of support vector machines and back-propagation artificial neural network for debris flow average velocity
YU Guo-qiang, Maosheng Zhang, Genlong Wang, Liang Pei · Institute of Geographic Sciences and Natural Resources Research, CAS Institutional Repository · 2012
To investigate the average velocity of viscous debris flow and coupling relationship of influence factors, different methods for researching debris flow are assessed. The SVM and BPANN models are proposed for predicting average velocity of viscous debris flow and building a predictive model. The two corresponding computer programs are compiled by the MATLAB program. Based on real time monitoring data of debris flow in the Jiangjia gully, relative advantages and disadvantages of the two models for predicting the average velocity are compared. The results show that both SVM and BPANN have sufficiently high accuracy in reproducing (fitting) the average velocity of viscous debris flow. However, in the validation phase, comparison of predictive accuracies of the SVM and BPANN models indicates that the former is superior to the latter in forecasting the average velocity. The extrapolating ability and predicting capability have been validated by using the support vector machine prediction model. The SVM model expresses well the complicated coupling relationship of debris flow velocity, and is more suitable for SVM prediction. Therefore, application of this method to such prediction is feasible and practical. It is also complementary and ideal for traditional research methods of debris flow, and is accurate scientific basis for prevention of debris flow.