基于混沌优化神经网络的冲击地压预测模型  

Combined optimization model of rock-burst prediction based on chaos optimization and BP neural networks

作　　者：尹光志[1] 谭钦文[1] 魏作安[1] 

机构地区：[1]重庆大学资源及环境科学学院

出　　处：《煤炭学报》2008年第8期871-875,共5页Journal of China Coal Society

基　　金：国家重点基础研究发展计划(973)基金资助项目(2005CB221502);国家自然科学基金重点资助项目(50534080);国家自然科学基金资助项目(50374084)

摘　　要：为了改进BP神经网络用于冲击地压预测的精度和泛化能力,利用BP算法和混沌优化算法优缺点的互补性,构建了一种组合式优化预测模型(COBP).将该模型应用于重庆砚石台煤矿冲击地压的预测,结果显示,该模型既利用混沌优化帮助BP算法克服了易陷入局部极值的缺点,又利用BP算法克服了基本混沌优化局部搜索能力有限和有时不能搜索到全局最优的缺陷.Based on the complementarities of BP neural network and chaos optimization algorithm, a new hybrid optimization model was presented. This model integrates chaos optimization algorithm with BP algorithm, which not only has a BP algorithm＇ s quick local search capability, but also can converge strongly to the global optimal result by use the chaos optimization＇ s global search character. Firstly, the model can effectively overcome the BP algorithm＇ s shortages such as local optimization, slowly convergence and instability etc. Secondly, it also can get rid of the shortage of the basic chaos optimization which sometimes couldn＇ t find the global optimization result. Finally, this new model is applied to predict rock burst in Yanshitai Coal Mine. The results show that it is an effective and feasible method to predict rock burst.

关 键 词：混沌优化 BP神经网络 冲击地压预测 优化模型 

分 类 号：TD324[矿业工程—矿井建设]