基于MMAS-BP的煤与瓦斯突出强度预测  

Prediction of Coal and Gas Outburst Intensity Based on MMAS-BP

作　　者：程加堂[1] 艾莉[1] 徐绍坤[1] 

机构地区：[1]红河学院工学院

出　　处：《中国安全科学学报》2011年第9期77-81,共5页China Safety Science Journal

基　　金：红河学院科研项目(10XJY117)

摘　　要：为提高煤与瓦斯突出强度的预测精度及预测速度,用最大最小蚂蚁系统和BP神经网络相结合的方法进行预测模型设计。根据煤与瓦斯突出强度及其主要影响因素之间的关系数据,建立其神经网络的预测模型。以网络的权值和阈值为自变量,网络误差为目标函数,通过蚁群算法的迭代运算,搜索出误差的全局最小值,以实现BP神经网络的初始权值、阈值优化,并用优化后的网络进行瓦斯突出强度的预测。实例结果表明,MMAS-BP算法的预测值均方差为0.089,约为BP神经网络的0.1倍,且输出稳定性好,适用于煤与瓦斯突出强度的预测。To improve the precision and efficiency of coal and gas outburst intensity,a neural network model combined with max-min ant system is presented.According to the data of outburst intensity and its main impact factors,the neural network prediction model is established.Taking BP network weights and threshold values as decision variables,and prediction error as objective function,the global minimum prediction error could be gotten through multiple generation computation of ant colony so the initial values of BP network can be optimized,and then the prediction model can be accomplished via the optimized neural network.An application example shows that the method avoids the randomness of selecting initial values,and that the mean square error gotten by MMAS-BP algorithm is 0.089 with a good output stability,about 0.1 times the size of the BP neural network,and the outputs stability.The model is suitable for the prediction of coal and gas outburst intensity.

关 键 词：最大-最小蚂蚁系统(MMAS) 神经网络 煤与瓦斯突出强度 优化 预测 

分 类 号：X936[环境科学与工程—安全科学]