基于神经网络模型的采场巷道收敛预测研究

Prediction of Convergence of Mining Roadways Based on Neural Network Model

根据余化寺矿的生产实际情况,分析影响巷道围岩收敛的主要因素,利用神经网络的非线性、学习和记忆等功能,建立了针对采场巷道收敛的神经网络预测模型。通过对采场巷道围岩收敛的现场监测,采用训练样本训练网络模型,并用检验样本对模型进行检验,预测模型性能好,预测精度高。

Based on the production practice at Yuhuasi Mine,the major factors influencing the convergence of surrounding rocks in roadways are analyzed.Making use of the non-linearity, self-study and memory of artificial neural network,a BP neural network model is founded for predicting the convergences of roadways. By surveying the convergence of surrounding rocks in roadways,the proposed model is trained by learnt samples and tested by check samples.It is found that the prediction model is of high quality and high precision.