Prediction of Mining-induced Seismic Events Using an Evolutionary-neural Computing Approach
This paper presents a hybrid evolutionary neural learning approach for analysis of mining-induced seismic system. In this method, a back-propagation neural network is used to learn the mapping relationship between system input and output from monitored data; Together with the local gradient descent algorithm, genetic algorithms is employed and a nested search procedure using different coding schemes is proposed to search the global optimal network topology and its connecting weights; In addition, an over fitting minimizing procedure was added for improving the prediction accuracy on unseen data. Application to the Laohutai mine is carried out and the results show the great performance of the proposed hybrid approach, both in learning and generalization.
artificial intelligence evolutionary computation back-propagation neural networks genetic algorithms nonlinear behaviour mining seismicity
Chengxiang Yang Xiating Feng
the School of Resources & Civil Engineering, Northeastern University, Shenyang 110004 P.R.China
国际会议
武汉
英文
2007-09-21（万方平台首次上网日期，不代表论文的发表时间）