| 基于GS-PSO-SVM的边坡稳定性预测模型 |
| |
| 引用本文： | 黄俊,刘小生.基于GS-PSO-SVM的边坡稳定性预测模型[J].中国矿业,2020,29(6). |
| |
| 作者姓名： | 黄俊 刘小生 |
| |
| 作者单位： | 江西理工大学建筑与测绘工程学院,江西 赣州341000,江西理工大学建筑与测绘工程学院,江西 赣州341000 |
| |
| 基金项目： | 国家自然科学基金项目（面上项目，重点项目，重大项目） |
| |
| 摘 要： | 针对传统边坡稳定性预测模型的不足，提出一种基于网格搜索和粒子群优化的支持向量机模型（GS-PSO-SVM）。为了解决支持向量机参数选取问题，先利用网格搜索法粗略寻优确定参数范围，然后利用粒子群二次寻优。利用该模型对边坡实例预测，39个实例样本中，30个为训练样本，剩下9个作为预测样本，以岩石重度、黏聚力、内摩擦角、边坡角、边坡高度、孔隙水压力6个边坡稳定性影响因素作为输入，边坡稳定性状态作为输出，预测结果与单独的网格搜索法、粒子群算法和遗传算法优化的支持向量机模型对比。结果表明，GS-PSO-SVM模型分类准确率100%，有更好地预测精度和更高的预测效率，该模型能有效的对边坡稳定性状态预测。
|
| 关 键 词： | 支持向量机 网格搜索法 粒子群算法 边坡稳定 |
| 收稿时间： | 2019/4/13 0:00:00 |
| 修稿时间： | 2020/6/16 0:00:00 |
Slope stability prediction model based on GS-PSO-SVM |
| |
| HUANG Jun and LIU Xiao-sheng.Slope stability prediction model based on GS-PSO-SVM[J].China Mining Magazine,2020,29(6). |
| |
| Authors: | HUANG Jun and LIU Xiao-sheng |
| |
| Affiliation: | School of Architectural and Surveying Mapping Engineering,Jiangxi University of Science and Technology,School of Architectural and Surveying Mapping Engineering,Jiangxi University of Science and Technology |
| |
| Abstract: | Aiming at the shortcomings of traditional slope stability prediction model, a support vector machine model (GS-PSO-SVM) based on grid search and particle swarm optimization is proposed. In order to solve the problem of parameter selection of support vector machine, the grid search method is used to roughly optimize the parameter range, and then the particle swarm optimization is used. Using this model to predict the slope example, 30 of the 39 sample samples are training samples, and the remaining 9 are used as prediction samples, with rock gravity, cohesion, internal friction angle, slope angle, slope height, and porosity. The influence factors of the six slope stability of water pressure are taken as input, the slope stability state is taken as the output, and the prediction result is compared with the separate grid search method, particle swarm optimization algorithm and genetic algorithm optimization support vector machine model. The results show that the classification accuracy of GS-PSO-SVM model is 100%, and it has better prediction accuracy and higher prediction efficiency. The model can effectively predict the slope stability state. |
| |
| Keywords: | support vector machine grid search method particle swarm optimization slope stability |
|
| 点击此处可从《中国矿业》浏览原始摘要信息 |
|
点击此处可从《中国矿业》下载全文 |
|
