基于MABC-SVR的边坡可靠性分析

(1.长安大学 地质工程与测绘学院,陕西 西安 710054; 2.西安市热力集团有限责任公司,陕西 西安 710016)

边坡可靠性; 改进的人工蜂群算法; 支持向量机回归; 蒙特卡罗法

Slope reliability analysis based on MABC-SVR
HAO Jianbin1,WEI Xingmei1, WANG Fen1,2

(1.School of Geology Engineering and Geomatics, Chang'an University, Xi'an 710054, China; 2.Xi'an District Heating Group, Xi'an 710016, China)

slope reliability; modified artificial bee colony algorithm(MABC algorithm); Support Vector Machine Regression(SVR); Monte Carlo method

DOI: 10.15986/j.1006-7930.2020.02.002

备注

针对原始人工蜂群算法收敛较慢、易陷入局部最优的缺点,提出一种新的改进人工蜂群算法,并利用支持向量机建立了边坡安全系数预测模型(MABC-SVR).在此基础上,结合蒙特卡罗法对典型边坡算例(ACADS)进行边坡可靠性分析,计算该边坡的可靠性指标与失效概率,计算结果与已有结果基本接近,且整个建模和模拟运行过程用时不到26 s,与单纯使用蒙特卡罗法相比,用时大大缩短,可见该可靠性分析方法是科学可行的.

Aiming at the disadvantage that the original artificial bee colony algorithm converges slowly and easily falls into local optimum, a new modified artificial bee colony algorithm(MABC)is proposed, and a slope safety factor prediction model(MABC-SVR)was established by combining the support vector machine.On this basis, a typical slope example(ACADS)is analyzed by Monte Carlo method to calculate the reliability index and failure probability of the slope.The calculation results are basically close to the existing results, and the whole modeling and simulation operation process is less than 26 s.Compared to just using the Monte Carlo method, the time is greatly shortened. It can be seen that the analytical method is scientific and feasible.