基于BP神经网络的矿车运行时矿井摩擦阻力的预测

Prediction of mine friction resistance during tramcar running based on BP neural network

为了快速有效地确定矿车等运输设备在巷道内运行时矿井摩擦阻力的变化情况,克服模拟软件计算量和现场实测工作量大的问题,以巷道风流速度、矿车运行速度、阻塞比、矿车长度4个矿车运行时巷道摩擦阻力的影响因素作为切入点,采用动网格技术模拟得到矿车在巷道内运行时有关矿井摩擦阻力的数据,以此为样本构建基于BP神经网络的矿井摩擦阻力预测模型,运用MATLAB软件进行网络训练,并将BP神经网络预测值与FLUENT模拟值进行对比。研究结果表明:BP神经网络结构比较简单,能以较快速度收敛,预测值与模拟值最大误差在7%以内,该神经网络模型用于求解矿车等运输设备在巷道内运行时摩擦阻力的变化情况是可行的。

In order to quickly and effectively determine the change of mine friction resistance when the tramcar and other transportation equipment are running in the mine roadway,and overcome the problems of large amount of calculation in simulation software and large amount of field measurement workload,four influencing factors of roadway friction resistance including the roadway wind speed,tramcar running speed,blocking ratio and tramcar length were taken as the starting point,and the data about the mine friction resistance when the tramcar ran in the roadway were obtained through the dynamic grid technique simulation.Taking this as a sample,a prediction model of mine friction resistance based on BP neural network was constructed,then the network was trained with MATLAB software,the predicted value of the BP neural network is compared with the FLUENT simulation value.The results showed that the BP neural network had simple structure and fast convergence speed,and the error between the predicted values and the simulated values was within 7%.The neural network model is feasible to solve the change of friction resistance of tramcar and other transportation equipment when they are running in the roadway.