三峡大学学报（自然科学版）2011，Vol.33Issue(6)：41-45,56,6.
粒子群优化-BP神经网络对岩爆的预测
Prediction of Rock Bursts Based on Particle Swarm Optimization-BP Neural Network
摘要
Abstract
Rock burst, as a basic kind of geological disaster, usually occurs in high geostress zones. There is no perfect prediction theory or occurrence mechanism so far. In this study, by choosing the main influence factors of rock bursts, BP neural network is used to train and predict the samples of rock bursts. Because initial weight values and threshold of the neural network have great effects on efficiency of learning and prediction of results, the prediction of testing samples using BP neural network are not satisfactory. The particle swarm optimization(PSO) algorithm is utilized to optimize the initial weight values and threshold of BP neural network. The improved BP neural network has a good prediction result for rock bursts. The results indicate that it is feasible to predict rock bursts based on PSO-BP neural network in practical engineering.关键词
岩爆预测/粒子群优化(PSO)/BP神经网络Key words
rock bursts prediction/particle swarm optimization(PSO) /BP neural network分类
建筑与水利引用本文复制引用
张强,王伟,刘桃根..粒子群优化-BP神经网络对岩爆的预测[J].三峡大学学报（自然科学版）,2011,33(6):41-45,56,6.基金项目
国家自然科学基金项目(51109069) （51109069）
中央高校基本科研业务费专项资金资助项目(2009B14014) （2009B14014）
