论文部分内容阅读
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。
In order to improve the accuracy and computational efficiency of the slope stability estimation method, HIOA-optimized MK-TCRVM (HIOA-MK-TCRVM) is established by combining HIOA with MK- TCRVM) algorithm, and use it to estimate the stability of rock slope and soil slope. At the same time, other slope stability estimation models based on single-kernel two-class correlation vector machine and support vector machine (SVM) are established and compared with the HIOA-MK-TCRVM algorithm for accuracy and sparseness. Finally, the effect of HIOA algorithm on optimizing parameters of MK-TCRVM algorithm is analyzed. The results show that the accuracy of HIOA-MK-TCRVM algorithm is 100%, and the accuracy of HIOA-MK-TCRVM algorithm is superior to other slope stability estimation models. The correlation vector of HIOA-MK-TCRVM algorithm Accounting for less than 25% of the training samples, the model sparsification is obvious. After the genetic operation is added to the HIOA algorithm, the evolution speed and the optimal solution are all improved.