Comparisons of BP-ANN Models of Mining Subsidence Prediction

摘要：

Mining subsidence prediction is a key step of some work, such as mining under buildings, rivers and railways, environment control of mining area and so on. The main factors influencing subsidence were comprehensively analyzed. Then the models to calculate surface subsidence were established by different methods of back propagation artificial neural network (BP-ANN). A large amount of data of observation stations of a certain mining field was used as learning and training samples to train and test the BP-ANN models. The results and calculating speed were compared and analyzed, it is proved that using BP-ANN mo del to predict the mining subsidence is much more reasonable. The BP-ANN method reduces artificial factors during the prediction, simplifies the complex problems and makes the results more reasonable and creditable.

关键词： mining subsidence prediction BP-ANN comparison

作者: FAN Hongdong DENG Kazhong WEI Hao

作者单位: College of Environment and Spatial informatics,China University of Mining and Technology (Xuzhou),Xuzhou 221008,Jiangsu,China

会议类型: 国际会议

会议名称: The 2008 International Symposium on Safety Science and Technology(2008年安全科学技术国际会议)

会议地点: 北京

会议语种:英文

页码: 2428-2431

在线出版日期: 2008-09-24（万方平台首次上网日期，不代表论文的发表时间）