Slope stability analysis based on quantum-behaved particle swarm optimization and least squares support vector machine

Given the complexity and uncertainty of the influencing factors of slope stability, its accurate evaluation is difficult to accomplish using conventional approaches. This paper presents the use of a least square support vector machine (LSSVM) algorithm based on quantum-behaved particle swarm optimization (QPSO) to establish the nonlinear relationship of slope stability. In the proposed QPSO-LSSVM algorithm, QPSO is employed to optimize the important parameters of LSSVM. To identify the local and global optimum, three popular benchmark functions are utilized to test the abilities of the proposed QPSO, the nonlinearly decreasing weight PSO, and the linearly decreasing weight PSO algorithms. The proposed QPSO exhibited superior performance over the other aforementioned algorithms. Simulation results obtained from PSO-LSSVM, QPSO-LSSVM, and LSSVM algorithms are compared in a case. Case analysis shows that QPSO-LSSVM has the quickest search velocity and the best convergence performance among the three algorithms, and is therefore considered most suitable for slope stability analysis.