基于CPSO算法的岩石蠕变模型非定常参数反演分析

针对粒子群优化(PSO)算法具有全局寻优能力强、无梯度信息、收敛速度快、算法简单但易陷入局部最优解且初始化解的质量不高的特点,利用混沌的遍历性,把混沌机制和粒子群优化算法结合起来,对粒子群优化算法进行了改进,提出了混沌粒子群优化算法,并利用混沌粒子群优化(CPSO)算法对岩石蠕变本构模型的非定常参数进行了反演分析,算例结果表明,采用该混沌粒子群优化算法反演非定常参数是可行的.

Back analysis of non-stationary parameters of rock creep model based on chaos particle swarm optimization algorithm

The particle swarm optimization(PSO) algorithm has high capabability in global optimization,a high rate of convergence and other advantages,including its lack of gradient information and simple solution process.But it is prone to provide a local optimum solution and bad initialization solution.Because of its ergodicity,the chaos mechanism was combined with the PSO algorithm,providing an improved PSO algorithm,the chaos particle swarm optimization(CPSO) algorithm.A case study shows that the CPSO algorithm is practical for the back analysis of non-stationary parameters of a rock creep model.