基于GA-BP神经网络的露天矿边坡变形预测分析

矿区边坡在各种因素的影响下，将会发生变形，但变形超过一定限度时，会对矿区产生很大的危害，开展边坡变形预测分析，能在一定程度上预防灾害的发生。文中在充分考虑BP神经网络初始权值和阈值难以确定，造成模型系统进入局部最小化，导致预测精度不高等问题的基础上，提出GA-BP神经网络预测模型，解决了普通网络模型在权值和阈值上的不足，并以越堡露天矿边坡变形监测点JC31、JC33、JC36为研究对象，分别采用灰色理论模型、BP神经网络模型以及GA-BP模型进行预测，研究结果表明:GA-BP网络模型较灰色模型和BP模型的预测值与实际值更吻合，预测精度更高，其平均相对误差最小，较其他两种方法预测精度提高了10倍以上，表明该方法具有一定的可靠性和可行性。 

Prediction and analysis of open pit slope deformation based on a GA-BP neural network

Due to various factors, mine slopes will be deformed. When the deformation exceeds a certain limit, it will cause great harm to the mining area. It can prevent disasters to a certain extent by carrying out slope deformation prediction and analysis. On the basis of fully considering that it is difficult to determine the initial weight and threshold of the BP neural network, which causes the model system to enter the local minimization and leads to the problem of low prediction accuracy, a GA-BP neural network prediction model that solves the deficiency of the ordinary network model in weight and threshold is proposed. Taking the slope deformation monitoring points JC31, JC33 and JC36 of the Yuebao open-pit mine as the research object, the gray theory model, BP neural network model and GA-BP model were used to predict the slope deformation. The results show that compared with the gray model and BP model, the predicted value of the GA-BP network model is more consistent with the actual value, the prediction accuracy is higher, and its average relative error is the smallest, which is more than 10 times higher than the other two methods, indicating that this method has certain reliability and feasibility.