基于BP神经网络的工作面周期来压预测

Forecast of Periodic Weighting in Working Face Based on Back-propagation Neural Network

针对现有工作面周期来压预测方法在算法结构复杂、计算量较大等问题,探索一种高效、准确、易于使用的工作面周期来压预测方法。以淮南潘集矿区20个已采工作面作为工程研究背景,采用BP神经网络预测原理,在分析工作面周期来压的影响因素基础之上,设计出一种基于BP神经网络的工作面周期来压预测方法,通过MATLAB编程实验表明BP神经网络在工作面周期来压的预测中具有较高的精度。

In order to solve the problems of complex algorithm structure and heavyworkload of calculation in periodic weighting forecasting,a kind of high efficient,accurate and easy to use method of periodic weighting forecasting was developed.20 working faces in Huainan Panji mining area were taken as study objects,adopting BP neural network principles,on the basis of analysis of influencing factors on periodic weighting,a kind of periodic weightingforecasting method based on BP neural network was developed.Experiments based on MATLAB programming showed that by BP neural network periodic weighting prediction is precisious.