摘要
依据某露天矿2006～2007年汽车运输能力统计数据,运用BP神经网络的非线性逼近能力、高度的鲁棒性和容错性及自学习和数据挖掘等特性建立了汽车运输能力时间序列动态预测模型。经过样本数据的训练和检验,表明该模型拟合精度较好,预测结果与实际值吻合度较高,可用于汽车运输能力的预测,为合理有效编制矿山采掘计划、制定经济合理的生产计划及矿山生产过程中采用计算机模拟对汽车运输能力的随机计算提供理论依据,从而提高矿山的经济效益。
Based on the statistical data reflecting the truck＇s transportation capacity from 2006 to 2007 in a certain opencut mine, the paper established a time series model for dynamically forecasting the truck＇s transportation capacity with BP neural network that has the characteristics of the nonlinear approximation ability, high degree of robustness, fault tolerance, self＋learning and data mining. It shows that the model has fine fitting accuracy and the prediction result is close to the actual value through the training and checking of sample data, so the model can be used to forecast the truck＇s transportation capacity. The studies can provide theoretical basis for the constitution of reasonable and effective mining plans, economical production plans as well as random calculations for the truck＇s transportation capacity in the course of mining production with the aid of computer simulation, finally economic benefits of the mine is to be enhanced.
出处
《矿山机械》
北大核心
2010年第15期50-52,87,共4页
Mining & Processing Equipment
