摘要
针对矿山边坡预测模型精度低等问题,提出一种由主成分分析(PCA)、灰狼算法(GWO)和支持向量机(SVM)组合的混合模型(PCA-GWO-SVM)。首先,采用PCA对原始数据进行降维去噪;其次,通过GWO算法优化支持向量机参数;最后,通过SVM实现矿山边坡变形的预测。矿山边坡实例表明PCA-GWO-SVM模型具有更高的预测精度。
Aiming at the low precision of mine slope prediction model,a hybrid model(PCA-GWO-SVM)consisting of principal component analysis(PCA),grey wolf algorithm(GWO)and support vector machine(SVM)was proposed.Firstly,PCA was used to reduce the noise of the original data.Secondly,the support vector machine parameters were optimized by GWO algorithm.Finally,the deformation prediction of mine slope was realized by SVM.The mine slope example showed that the PCA-GWO-SVM model had higher prediction accuracy.
作者
解洪伟
朱东丽
Xie Hongwei;Zhu Dongli(Surveying and Mapping Institute of Guangdong Non-ferrous Metals Geological Bureau,Guangzhou 510080,China;Guangzhou Urban Planning Survey Design and Research Institute,Guangzhou 510060,China)
出处
《矿山测量》
2020年第1期63-66,共4页
Mine Surveying
关键词
支持向量机
矿山边坡
变形监测
主成分分析
灰狼算法
support vector machine
mine slope
deformation monitoring
principal component analysis
grey wolf algorithm