Prediction of Activity Intensity of Ground Fissures Based on BP Neural Network——A Case Study of Xi'an Ground Fissures
I Jianjun · Journal of Catastrophology · 2010
Ground fissure is one of the typical disasters. It can do great harm to urban construction, including surface and underground constructions. According to the monitoring data of Xi'an ground fissures and research results in recent years,features and main influence factors of activity of Xi'an ground fissures and prediction of activity intensity of ground fissures are analyzed by combining BP neural network with Geographic Information Systems. Cluster analysis is introduced to the BP neural network learning process,and at the same time,inertia factor is used to speed up the convergence rate of learning. The resulting can be shown in geographic information system in a visualized manner. After the actual data validation,it is proved to be an effective method.