Analyzing Surrounding Rock Failure Risk of Undersea Tunnel Based on GA-ANN
The undersea tunnel calculation has characters of uncertainty and randomicity. The numerical simulation expends much time and can not express the explicit relation between parameters and responding value, which brings along difficulties in uncertain calculation. The paper proposed a surrounding rock failure probability calculation method based on genetic algorithm(GA) and artificial neural network (ANN). According to random distribution of each parameter, adopts orthogonal and uniformity design principles and produces data samples by numerical tests for training ANN model, the GA is used to optimize the ANN structure. Then, the GA-ANN mapping model combining Monte-Carlo sampling method can be used to calculate rock failure probability which can directly take usage of current large scale numerical simulation software. Calculation sample study shows the method is feasible.
Undersea tunnel fluid-solid coupling failure risk genetic algorithm artificial neural network
AN-NAN JIANG CHUN-YAN BAO CHUN-AN TANG YONGBIN ZHANG
Institute of Highway and Bridge,Dalian Maritime University,Dalian,116026,P.R.China School of Civil and Hydraulic Engineering,Dalian University of Technology Dalian,116024,P.R.China
国际会议
大连
英文
369-373
2009-10-20（万方平台首次上网日期，不代表论文的发表时间）