Forecasting Model of Blasting Vibration Acceleration Peak-value Based on SVM

Dong Li · Tance yu kongzhi xuebao · 2010

Aiming at the problem of low generalization capability in forecasting the parameters of blasting vibration,a new forecasting model based on combination of grey method and SVM was proposed.The model calculated and confirmed the main factors affecting the blasting vibration acceleration peak-value by grey method.By replacing the ERM rule with SRM rule,it solved the practical problems such as small sample,non-linear and partial infinitesimal,and improved the forecasting precision.The calculation instances showed that the average absolute relative error between the forecasting values(by SVM model) and real values was less than 5%,which proved that the SVM based forecasting model was feasible and effective.

Read the paper · More papers on PaperTik