The Stability Classification System For The Surrounding Rocks In Mining Roadway Based On Ann Method

There are many factors that affect the stability of the wall rock of mining roadway, and it exhibits a nonlinear relationship between these factors and the classification of the analysis about the wall rock of mining roadway.So expert develop many methods on the stability of the wall rock such as fuzzy accumulate method, grey system theoretical method.The artificial neural net provide us new theoretical method on stability of wall rock. It is a new information process system based on the preliminary knowledge on the organizing structure and mechanism about human,s brain, and it has powerful recognition and classifying ability to the external input samples which can perfectly solve the problem on apporching to nonlinear curve.We investigate the data of the wall rock of mining roadway in YanZhou mining area and analysis all factors which affect the wall rock. Based on analyzing the major factors. We establish the BP neural net model which to make classifying recognition on stability of wall rock of mining roadway. Then we improve the training speed and generalization ability of this model because the traditional BP neural net is very slow and easy to be trained excessively. So, by this new model, we implement the scientifically classification on stability of the wall rock of mining roadway in YanZhou mining area and provide the theoretical reference which to design the anchor shank sporting scientifically, economical, and effectively. At last we design a set of software used in neural net training, classifiying of mining roadway, designing of anchor shank supporting and data base management by the programming method combining structure with object—oriented .Thus the application of research achievement is solved effectively.