基于蚁群优化神经网络的巷道围岩稳定性预测  

Roadway's Surrounding Rocks Stability Predicting Based on Neural Nnetwork Using Ant Optimization Algorithm

作　　者：孙洁[1] 刘晓悦[1] 

机构地区：[1]河北理工大学计算机与控制学院,河北唐山063009

出　　处：《微电子学与计算机》2008年第5期117-119,共3页Microelectronics & Computer

基　　金：国家自然科学基金项目(50674093)

摘　　要：巷道围岩的稳定性是保证煤矿安全开采的重要因素.在对影响巷道稳定性因素综合分析基础之上,针对神经网络处理非线性对象的优点,对样本数据进行神经元网络模型训练,并利用蚁群算法对神经网络进行快速优化,避免网络陷入局部极小值.仿真结果表明网络训练误差及收敛速度达到对顶底板和两帮的移近率的合理预测,进而可以确定巷道的稳定性状态。以移近率的大小和巷道稳定性为依据,选择合理的支护方式,对煤矿安全生产提供科学指导.The stability of the roadway＇s surrounding rocks stability is an important factor to guarantee coal mining safety. In this paper, on the basis of comprehensive analysis of the effects factors of roadway＇ s stability, considering of the advantages of dealing with non-linear object of neural network, the neural network model is trained by the sample data, and the using of ant algorithm to optimize neural network rapiddly to avoid Local minimum. The simulation results show that net- work training error and convergence speed reach a reasonable forecast for the roof and floor and the two sides moved closer to the rate, which determine the stability of the state roadway. Based on Closer to the size and rate and stability roadway, the reasonable supporting way is choosed to provide scientific guidance for production safety in coal mines.

关 键 词：蚁群算法 神经元网络 巷道稳定性 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]