摘要
针对目前边坡安全系数估算模型存在的问题,提出了一种基于改进的灰狼优化算法对支持向量机模型参数进行寻优,建立IGWO-SVM模型,对边坡安全系数进行估算。结果表明,在82个样本实例中,前71个为训练样本,后11个为预测样本,预测结果与改进的BP、GP、GWO-SVM预测模型进行对比。经对比,IGWO-SVM模型的平均绝对误差(Mean Absolute Error,MAE)、平均相对误差(Mean Absolute Percentage Error,MAPE)、均方根误差(Root Mean Square Error,RMSE)均最小,预测精度和预测效率均较高,该模型能有效地对边坡稳定性状态进行预测。IGWO-SVM模型估算精度更高,具有一定的应用价值。
Aimed at the problems existing in the current slope safety factor estimation model,this paper proposes an improved gray wolf optimization algorithm to optimize the parameters of the support vector machine model,and finally establishes the IGWO-SVM model,which is applied to slope safety factor estimation.The engineering examples show that the first 71 samples are training samples and the last 11 samples are prediction samples.The prediction results are compared with those of the improved prediction models of BP,GP and GWO-SVM.The MAE,MAPE and RMSE of the IGWO-SVM model are the smallest among all the models,which have higher prediction accuracy and prediction efficiency.This model can effectively predict the stability state of a slope.IGWO-SVM has higher estimation accuracy and enjoys certain application value.
作者
欧阳东生
OUYANG Dongsheng(Surveying and Mapping Institute of Guangdong Non-ferrous Metals Geological Bureau,Guangzhou 510080,China)
出处
《江西冶金》
2022年第6期108-114,共7页
Jiangxi Metallurgy
关键词
边坡
安全系数
支持向量机
灰狼优化算法
slope
slope safety factor
support vector machine
gray wolf optimization algorithm
