基于PSOBP-AdaBoost模型的瓦斯涌出量分源预测研究  

Research on prediction of gas emission quantity with sub sources basing on PSOBP-Ada Boost

作　　者：温廷新[1,2] 孙雪[1] 孔祥博[2] 田洪斌[2] 

机构地区：[1]辽宁工程技术大学系统工程研究所,辽宁葫芦岛125105 [2]辽宁工程技术大学工商管理学院,辽宁葫芦岛125105

出　　处：《中国安全科学学报》2016年第5期94-98,共5页China Safety Science Journal

基　　金：国家自然科学基金资助(713711091);辽宁省社会科学基金资助(L14BTJ004)

摘　　要：为准确预测瓦斯涌出量,选取某煤矿的开采煤层、临近煤层、采空区3个瓦斯涌出源作为实例研究,将BP神经网络、粒子群算法(PSO)、Ada Boost迭代提升算法和瓦斯涌出分源预测法相结合,建立基于PSOBP-Ada Boost算法的瓦斯涌出量分源预测模型,并将其与BP神经网络算法进行比较分析。结果表明,PSOBP-Ada Boost算法预测的3个瓦斯涌出源平均相对误差分别为3.24%,2.11%,3.21%;BP神经网络的平均相对误差分别为6.73%,3.19%,4.27%,基于PSOBP-Ada Boost模型的预测精度明显优于BP神经网络模型。In order to predict the quantity of gas emission from different sources in a mine,the coal seam being mined,the adjacent coal seam and the gob of a certain coal mine in China were taken as the gas emission sources,and a PSOBP-Ada Boost based model was built for predicting the quantity. Effectiveness of the model was verified with the data acquired in the mine.It was found that accuracy of prediction of the model is better than that of the BP neural network model.

关 键 词：瓦斯涌出量 分源预测 BP神经网络 粒子群算法(PSO) ADA Boost迭代算法 误差 

分 类 号：X936[环境科学与工程—安全科学]