基于P-SA-PSO算法的概率积分法参数反演

Parameter Inversion of Probability Integral Method Based on P-SA-PSO Algorithm

针对目前概率积分法参数反演存在的问题,为了更加准确稳定地解算出概率积分法参数,在经典粒子群算法的基础上,引入自适应惯性权重以及模拟退火算法中的Metropolis准则,提出了一种基于P-SA-PSO算法的概率积分模型参数反演方法。论文通过仿真实验,与遗传算法(GA)、经典粒子群算法(PSO)以及加入自适应惯性权重优化的粒子群算法(P-PSO)进行对比分析,结果表明:新算法反演各参数的精确性、稳定性和收敛速度优于其他3种算法。将新算法引入工程实例中,进一步体现了新算法在反演概率积分模型参数上的可行性。

Aiming at the problems existing in the parameter inversion of probability integral method,in order to solve the parameter of probability integral method more accurately and stably,a method of parameter inversion of probability integral model based on P-SA-PSO algorithm was proposed by introducing adaptive inertia weight and Metropolis criterion in simulated annealing algorithm on the basis of classical particle swarm optimization algorithm.The results show that the accuracy,stability and convergence speed of the new algorithm are better than those of the other three algorithms.The feasibility of the new algorithm in inverting the parameters of probability integral model is further demonstrated by introducing the new algorithm into an engineering example.