基于HIOA-MK-TCRVM算法的边坡稳定性估计模型

发布时间：2018-04-20 02:07

本文选题：多核二分类相关向量机(MK-TCRVM) + 边坡稳定性　；参考：《中国安全科学学报》2017年01期

【摘要】：为提高边坡稳定性估计方法的精度及计算效率,将混合智能优化算法(HIOA)与多核二分类相关向量机(MK-TCRVM)算法相结合,建立HIOA优化的MK-TCRVM(HIOA-MK-TCRVM)算法,并用其估计岩质边坡及土质边坡稳定性。同时,基于单核二分类相关向量机、支持向量机(SVM)等算法建立其他的边坡稳定性估计模型,并与HIOA-MK-TCRVM算法进行精度与稀疏性对比分析。最后,分析HIOA算法优化MK-TCRVM算法参数的效果。结果表明,HIOA-MK-TCRVM算法对训练集与测试集边坡稳定性估计的准确率均达到100%,其精度优于其他边坡稳定性估计模型;HIOA-MK-TCRVM算法的相关向量数占训练样本数的25%以内,模型稀疏化效果明显;向HIOA算法中加入遗传操作后,其进化速度及最优解均得到较好的改善。
[Abstract]:In order to improve the accuracy and computational efficiency of the slope stability estimation method, a HIOA optimized MK-TCRVM-based MK-TCRVM-based MK-TCRVM-based algorithm is established by combining the hybrid intelligent optimization algorithm (HIOAA) with the multi-kernel two-classification correlation vector machine (MK-TCRVM-based). The stability of rock and soil slopes is estimated by using the algorithm. At the same time, other slope stability estimation models are established based on single-core two-classification correlation vector machine, support vector machine and SVM, and the accuracy and sparsity are compared with HIOA-MK-TCRVM algorithm. Finally, the effect of HIOA algorithm to optimize the parameters of MK-TCRVM algorithm is analyzed. The results show that the accuracy of HIOA-MK-TCRVM algorithm for slope stability estimation of training set and test set is 100, and the accuracy of HIOA-MK-TCRVM algorithm is better than that of other slope stability estimation models. The correlation vector number of HIOA-MK-TCRVM algorithm is less than 25% of the training samples, and the sparse effect of the model is obvious. After adding genetic operation to HIOA algorithm, the evolutionary speed and optimal solution are improved.
【作者单位】：东华理工大学测绘工程学院;武汉大学测绘学院;
【基金】：国家自然科学基金资助(41374007) 江西省自然科学基金资助(20151BAB213031) 江西省教育厅科学技术研究项目(GJJ150592)
【分类号】：TU43;TP18

【相似文献】