基于灰色关联分析与GA-BP神经网络的拉斗铲生产能力预测  

Production capacity prediction of dragline based on grey correlation analysis and GA-BP neural network

作　　者：赵红泽[1,2] 王宇新 李淋 郭帅 王金瑞[3] 任志辉 Zhao Hongze;Wang Yuxin;Li Lin;Guo Shuai;Wang Jinrui;Ren Zhihui(School of Energy and Mining Engineering,China University of Mining&Technology-Beijing,Beijing 100083,China;State Key Laboratory for Geomechanics and Deep Underground Engineering,Beijing 100083,China;Heidaigou Open-pit Coal Mine,Shenhua Group Zhungeer Energy Co.,Ltd.,Zhungeer Inner Mongolia 010300,China)

机构地区：[1]中国矿业大学(北京)能源与矿业学院,北京100083 [2]深部岩土力学与地下工程国家重点实验室,北京100083 [3]神华准格尔能源有限责任公司黑岱沟露天煤矿,内蒙古准格尔010300

出　　处：《矿业科学学报》2020年第1期58-66,共9页Journal of Mining Science and Technology

基　　金：国家重点研发计划(2018YFC0808301);国家十二五科技支撑计划(2015BAK38B01);深部岩土力学与地下工程国家重点实验室开放基金(SKLGDUEK1923)

摘　　要：为了解决拉斗铲生产能力的测量问题,本文提出一种基于灰色关联结合GA-BP神经网络的预测方法。对影响拉斗铲生产能力的12个因素进行灰色关联分析,选取实动时间、出动率、有效抛爆量和有效抛掷率4个灰色关联度大于0.7的影响因素作为输入变量,拉斗铲月生产能力作为输出变量,建立了GA-BP神经网络和BP神经网络预测模型。结果表明,GA-BP神经网络最大相对误差为8.786%,平均相对误差为3.385%,平均相对误差方差为0.0156,迭代次数为18次,各项性能均优于常规BP神经网络。GA-BP神经网络模型对拉斗铲生产能力预测的泛化性能更好,精度更高,为拉斗铲生产能力的预测提供了一种较为有效的方法。In order to solve the measurement problem of the production capacity of the dragline,a prediction method based on gray correlation combined with GA-BP neural network is proposed.The gray correlation analysis is carried out on the 12 influencing factors affecting the production capacity of the dragline,and the actual working hours,the out rate,the effective casting blast amount,and effective throw-out rate are selected.Four influencing factors greater than 0.7 were used as input variables and the monthly production capacity of the dragline was used as the output variable to establish GA-BP neural network and BP neural network prediction model.The results show that the maximum relative error of GA-BP neural network is 7.525%,the average relative error is 3.52%,the average error variance is 0.0156,and the number of iterations is 18 times.The performance is better than the conventional BP neural network.The GA-BP neural network model provides better and more accurate generalization performance for the production capacity of the dragline,and provides a more effective method for predicting the production capacity of the dragline.

关 键 词：灰色关联分析 BP神经网络 遗传算法 拉斗铲生产能力 预测 

分 类 号：TD[矿业工程]