边坡失稳往往会造成巨大的经济损失和人员伤亡,构建科学有效的边坡稳定性评价模型对于边坡灾害防治具有重要意义。针对边坡评价涉及多重不确定性,探讨了基于正态云模型、果蝇优化算法与支持向量机耦合的边坡稳定性评价模型,即利用正态云模型描述果蝇个体飞行方向和飞行距离的随机性与模糊性,以改进果蝇优化算法,进而应用基于正态云模型的果蝇算法(CMFOA)求解支持向量机(SVM)分类模型的最优参数组合,并结合实例应用及与GA-SVM、PSO-SVM和网格法-SVM对比分析,验证模型的可靠性。实例应用及对比结果表明,CMFOA-SVM模型应用于边坡稳定性评价有效可行,且评价准确率高,同时CMFOA算法的参数寻优效率高,也为其他分类问题提供了新的参考。
Abstract
Since slope instability may cause huge economic losses and casualties, a rational and effective model for slope stability evaluation is of vital significance to disaster prevention. Considering multiple uncertainties of slope stability evaluation, we propose a slope stability evaluation model coupling fruit fly optimization algorithm (FOA), normal cloud model (CM), and support vector machine (SVM). The normal CM is used to optimize the FOA by describing the randomness and fuzziness in the foraging process of drosophila. The CM-based FOA, namely, CMFOA, is used to optimize the parameters of SVM classification model. The reliability of the proposed CMFOA-based SVM model is verified with a case study and comparisons with GA-SVM, PSO-SVM, and Grid Search-SVM. Results demonstrated that the proposed model is effective and feasible in slope stability evaluation. The accuracy and efficiency of parameter selection for classification using the CMFOA-SVM model are better than those of other models. It also overcomes the shortcoming of local optimal capability in traditional FOA and provides a new reference for other similar classification problems.
关键词
边坡稳定性 /
云模型 /
果蝇算法 /
支持向量机 /
评价指标
Key words
slope stability /
cloud model /
fruit fly optimization algorithm /
support vector machine /
evaluation indices
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 自然资源部地质灾害技术指导中心, 地调局环境监测院. 全国地质灾害通报[M]. 北京: 自然资源部, 2020.
[2] WANG Ming-wu, WANG Xiao, LIU Qiu-yan, et al. A Novel Multi-dimensional Cloud Model Coupled with Connection Numbers Theory for Evaluation of Slope Stability[J]. Applied Mathematical Modelling, 2020, 77: 426-438.
[3] 谢全敏, 夏元友. 岩体边坡治理决策的模糊层次分析方法研究[J]. 岩石力学与工程学报, 2003, 22(7): 1117-1120.
[4] 刘寒冰, 李国恒, 谭国金, 等. 基于时间序列的边坡位移实时预测方法[J]. 吉林大学学报(工学版), 2012, 42(增刊1): 193-197.
[5] 张正虎, 袁孟科, 邓建辉, 等. 基于改进灰色-时序分析时变模型的边坡位移预测[J]. 岩石力学与工程学报, 2014, 33(增刊2): 3791-3797.
[6] 夏元友, 李新平, 朱瑞赓. 基于人工神经网络的边坡稳定性工程地质评价方法[J]. 岩土力学, 1996, 17(3): 27-33.
[7] 刘沐宇. 冯夏庭. 基于神经网络范例推理的边坡稳定性评价方法[J]. 岩土力学, 2005, 26(2): 193-197.
[8] 兰海涛,李 谦,韩春雨. 基于广义回归神经网络的边坡稳定性评价[J]. 岩土力学,2009,30(11):3460-3463.
[9] FENG Xian-da, LI Shu-chen, YUAN Chao, et al. Prediction of Slope Stability Using Naive Bayes Classifier[J]. KSCE Journal of Civil Engineering, 2018, 22(3): 941-950.
[10] CRISTIANINI N. An Introduction to Support Vector Machines[M]. Cambridge: Cambridge University Press, 2000.
[11] CORTES C, VLADIMIR V. Support Vector Networks[J]. Machine Learning, 1995, 20(3): 273-297.
[12] PAN W T. A New Fruit Fly Optimization Algorithm: Taking the Financial Distress Model as An Example[J]. Knowledge-Based Systems, 2012, 26: 69-74.
[13] LI Chun-quan, XU Shao-ping, LI Wen, et al. A Novel Modified Fly Optimization Algorithm for Designing the Self-tuning Proportional Integral Derivative Controller[J]. Journal of Convergence Information Technology, 2012, 16: 69-77.
[14] PAN Quan-ke, SANG Hong-yan, DUAN Jun-hua, et al. An Improved Fruit Fly Optimization Algorithm for Continuous Function Optimization Problems[J]. Knowledge-Based Systems, 2014, 62: 69-83.
[15] 李德毅, 孟海军, 史雪梅. 隶属云和隶属云发生器[J]. 计算机研究与发展, 1995, 32(6): 15-20.
[16] 李德毅, 刘常昱. 论正态云模型的普适性[J]. 中国工程科学, 2004, 6(8): 28-34.
[17] WU Liang-hong, ZUO Ci-li, ZHANG Hong-qiang, et al. Bimodal Fruit Fly Optimization Algorithm Based on Cloud Model Learning[J]. Soft Computing A Fusion of Foundations, Methodologies & Applications, 2017, 21(7): 1877-1893.
[18] GEISSER S. The Predictive Sample Reuse Method with Applications[J]. Journal of the American Statistical Association, 1975, 70(350): 320-328.
[19] 李朋丽. 基于智能算法的陕西省公路边坡稳定性评价[D]. 西安: 长安大学, 2015.
[20] LIN Chih-jen, CHANG Chih-chung. LIBSVM: A Library for Support Vector Machines[J]. ACM Transactions on Intelligent Systems and Technology, 2011, 2(3): 1-39.
基金
国家重点研发计划项目(2017YFC1502405);国家自然科学基金项目(41172274)
PDF(1506 KB)
