金属矿山岩层移动角选取的BP神经网络模型及工程应用

BP NEURAL NETWORK MODEL TO PREDICT MOTION ANGLE OF METAL MINE AND ITS APPLICATION IN ENGINEERING

研究岩层移动与沉陷最重要的角值参数是移动角,正确确定岩层移动角与移动范围对于矿山工程布置具有非常重要的意义。本文在分析国内外大量金属矿山崩落法与充填法开采岩层移动研究成果的基础上,采用改进的BP神经网络算法建立了金属矿山岩层移动角选取的模型,并对模型的预测精度进行了验证。结果表明,所选取的各因素之间与岩层的移动之间存在着较强的非线性规律,实际输出与期望输出基本吻合。将所建立的网络模型应用于狮子山铜矿地下开采岩层移动角的预测,取得了较好的效果。

The most important parameter to study the strata movement and subsidence is moving angle of strata.Correct determination of the angle of strata movement has the extremely vital significance for mining engineering arrangement.Based on a large number of research results of moving angle of strata in domestic and foreign caving and filling mine,the model for calculating the displacement angles of strata is established by applying the improved BP neural network model,and the precision of the model is further validated.The results show that there is a strong nonlinear charateristeics between the select factors and strata movement angle,the anticipant and the actual output results are very similar.The model is applied in predicting moving angle of strata of Shizishan copper mine,and achieved considerable effects.