基于bi-LWCA-ENN煤与瓦斯突出危险性预测  

Prediction of Coal and Gas Outburst Risk Based on bi-LWCA-ENN

作　　者：付华[1] 司南楠 鲁俊杰[1] 王雨虹[1] 徐耀松[1] 

机构地区：[1]辽宁工程技术大学电气工程与控制工程学院,辽宁葫芦岛125105

出　　处：《传感技术学报》2016年第8期1222-1228,共7页Chinese Journal of Sensors and Actuators

基　　金：国家自然科学基金项目(51274118);辽宁省教育厅基金项目(L2012119);辽宁省科技攻关项目(2011229011)

摘　　要：为了提高煤与瓦斯突出的预测精度,以实现准确、可靠的瓦斯突出危险性预测,提出一种双层狼群算法(LWCA)优化Elman神经网络模型进行模式分类与预测,建立煤与瓦斯突出的双层LWCA-ENN预测模型。分析煤与瓦斯突出机理和影响因素,提取相关数据样本,筛选稳定的特征子集作为特征向量训练模型,算法通过对Elman神经网络的权值、阈值寻优,建立了基于bi-LWCA-ENN算法的预测模型并结合矿井监测数据进行实例分析。试验结果表明:煤与瓦斯突出的bi-LWCA-ENN模型稳定性好,收敛速度快,有效地实现了瓦斯突出危险性预测。In order to improve the prediction accuracy of coal and gas outburst, realize the accurate and reliable gas outburst danger prediction, this paper put forward a method that use bi-Leader Wolves Colony Algorithm （LWCA）to optimize Elman neural network for pattern classification and prediction, and model of coal and gas outburst was es- tablished by bi-LWCA-ENN algorithm. On the analysis of the mechanism and influencing factors of coal and gas outburst based on the data samples that extracted by application of feature selection as the characteristic vector, a bi-Leader Wolves Colony Algorithm （LWCA）was merged with Elman neural network to optimize weight and thresh- old, with the data of mine actual monitoring to experiment and analysis. The results show that the bi-LWCA-ENN model of coal and gas has good stability and fast convergence rate, realized the gas outburst prediction effectively.

关 键 词：煤与瓦斯突出危险性 双层狼群算法 ELMAN神经网络 特征选择 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] TP212[自动化与计算机技术—计算机科学与技术]