基于SSA-BP的露天矿山边坡位移变形预测

Research on Slope Displacement Prediction of Open-Pit Mine Based on SSA-BP

针对传统的BP神经网络在预测露天矿山边坡位移变形时存在的局限性,构造了一种基于麻雀搜索算法(SSA)和BP神经网络相结合的边坡位移变形预测模型,先利用麻雀搜索算法对传统的BP神经网络进行权值与阈值的优化,再将麻雀搜索算法优化后的BP神经网络算法(SSA-BP)运用于露天矿山边坡位移的预测。为了验证算法的可行性,将SSA-BP预测模型与WOA-BP、BP以及Elman预测模型针对露天矿山边坡位移变形量的预测结果进行比较。实验结果表明:SSA-BP预测模型针对露天矿山边坡位移变形量的预测相比其他三种模型,其迭代速度快、寻优能力强;通过预测精度评价指标来看,SSA-BP算法的R2、RMSE、MAPE、MAE、MSE明显优于另外三组算法。为露天矿山边坡位移变形预测提供了一种新的思路和方法。

Aiming at the limitations of traditional BP neural network in predicting slope displacement and deformation in open-pit mines,a slope displacement and deformation prediction model based on the combination of sparrow search algorithm(SSA)and BP neural network was constructed.First,Sparrow search algorithm was used to optimize the weight and threshold of traditional BP neural network.Then the BP neural network algorithm(SSA-BP)optimized by Sparrow search algorithm is applied to the prediction of slope displacement in open-pit mine.To verify the feasibility of the algorithm,the SSA-BP prediction model was compared with the WOA-BP,BP and Elman prediction models for slope displacement and deformation of open-pit mine.The experimental results showed that compared with the other three models,the SSA-BP prediction model hadbetter iteration speed and optimization ability for slope displacement and deformation prediction.According to the prediction accuracy evaluation index,the R 2,RMSE,MAPE,MAE and MSE of SSA-BP algorithm wereobviously superior to the other three groups of algorithms.It provides a new idea and method for slope displacement and deformation prediction in open-pit mine.