Prediction analysis of destroyed coalseam floor depth based on v-SVR algorithm

摘要

There are main six factors controlling failure zone floor depth including mining depth, coal seam inclination angle, mining thickness, workface inclined length, coal floor anti-destructive capacity, fault or fracture zone. Inorder to prediet failure depth of coal seam floor, based on the analysis of the factors influencing the failure depth of coal seam floor, a model to predict the failure depth is established by applying the theory of v-SVR algorithm. A large amount of on-site observed data was used as learning and training samples. Optimized the parameters of the model by grid-search method and tested the model performance. The model considered comprehensive affected factors and considered the nonlinear relationship between factors and target value.The results show that v-SVR algorithm model predictive value is more closer to measured datas compared to empirical formula calculated value.