GA-BP和PSO-BP预测模型在九龙矿煤层底板突水预测中的应用研究  

Study on Application of GA-BP and PSO-BP Prediction Models in Prediction of Water Inrush from Coal Seam Floor in Jiulong Mine

作　　者：刘滢 卢兰萍[1] 王铁记 靳子栋 张会松 卫皓皓 LIU Ying;LU Lanping;WANG Tieji;JIN Zidong;ZHANG Huisong;WEI Haohao(School of Civil Engineering,Hebei University of Engineering,Handan 056038,China;Fengfeng Group,Jizhong Energy Group,Handan 056038,China;Jiulong Mine,Handan Baofeng Co.,Ltd.,Fengfeng 056200,China)

机构地区：[1]河北工程大学土木工程学院,河北邯郸056038 [2]冀中能源峰峰集团,河北邯郸056038 [3]邯郸市宝峰有限公司九龙矿,河北峰峰056200

出　　处：《煤炭技术》2024年第6期169-173,共5页Coal Technology

基　　金：国家自然科学基金资助项目(41902254);河北省自然科学基金生态智慧矿山联合基金资助项目(D2020402013)。

摘　　要：目前,煤层开采环境复杂,随着开采深度、开采强度的增加,面临多变的突水因素和复杂的突水机理,且各因素间相互联系的不确定性,使底板突水预测的难度不断增加。对GA-BP与PSO-BP两种组合优化方法进行描述、对比。两种组合优化方法克服了神经网络容易收敛到局部最小值,以及收敛速度慢的缺点,对煤层底板突水都能实现较高精度,具有强大的泛化能力。通过对两种组合优化方法的预测模型做对比,发现GA-BP模型更优于PSO-BP模型,证明GA-BP组合优化方法更适合对底板突水危险性进行预测。At present,the environment of coal seam mining is complex.With the increase of mining depth and mining intensity,it is faced with changeable water inrush factors and complex water inrush mechanism,and the uncertainty of mutual connection between various factors makes the difficulty of floor water inrush prediction increasing.Describes and compares the two combination optimization methods of GA-BP and PSO-BP.The two combined optimization methods overcome the shortcomings that the neural network is easy to converge to the local minimum and the convergence speed is slow.They can achieve high accuracy and strong generalization ability for water inrush from coal seam floor.By comparing the prediction models of the two combined optimization methods,it is found that the GA-BP model is better than the PSO-BP model,which proves that the GA-BP combined optimization method is more suitable for predicting the risk of floor water inrush.

关 键 词：GA-BP PSO-BP BP神经网络 组合优化方法 底板突水 

分 类 号：TD742[矿业工程—矿井通风与安全] TD745.21