基于神经网络的采空塌陷预测

Prediction of mining collapse based on neural network

摘要: 依据某煤炭开采区的勘察资料, 综合考虑影响采空塌陷的主要因素, 建立了预测采空塌陷的 BP神经网络模型。该模型结构为 7-10-2型。优化学习参数后, 用该模型对采空区塌陷进行了预测分析, 结果与实际情况完全吻合, 表明 BP神经网络模型应用于采空塌陷预测领域是行之有效的。

Abstract: It is introduced that the BP neural network model of prediction of mining collapse is established based on the survey′s data and main factors at a mine field where the collapse had happen.The model structure is 7-10-2.After several learning parameters are optimized,the prediction of mining collapse at the field is made with this model.The results show that BP neural network model is feasible and effective in prediction of mining collapse.