基于粒子群算法的发动机部件模型求解

Application of Particle Swarm Optimization in Obtaining Solution of Aeroengine Component-Level Model

关键词:  粒子群算法  收敛因子  被动聚集压力因子  自适应惯性权重

Key words:  Particle swarm optimization  Constriction factor  Passive congregation factor  Adaptive inertia weight

摘要:

传统求解航空发动机部件模型中的非线性方程和共同工作方程组的方法因对初值的依赖使得模型并不总能收敛。首先针对模型中的非线性方程求解的特点,在带邻域的粒子群算法的基础上,将收敛因子、被动聚集压力因子和自适应惯性权重引入算法,提出一种混合粒子群算法HPSO1。而对模型中的共同工作方程组的求解,则是在基本粒子群算法的基础上,将收敛因子和被动聚集压力因子引入基本粒子群算法,提出另一种混合粒子群算法HPSO2。实验仿真表明HPSO1和HPSO2均克服了对初值的依赖性,因而对发动机部件模型求解是有效的。

Abstract:

The convergence of conventional solutions to nonlinear equation and component matching equations of aeroengine component model depends on the initial values. Based on the peculiarity of nonlinear equation of the model, a hybrid Particle Swarm Optimization(PSO) algorithm HPSO1 is proposed, which integrates constriction factor, passive congregation factor and adaptive inertia weight on the basis of neighbor PSO. As for component matching equations,another hybrid Particle Swarm Optimization(PSO) algorithm HPSO2 is proposed, which integrates constriction factor, passive congregation factor on the basis of basic PSO.The simulation results demonstrate that both HPSO1 and HPSO2 overcome the dependence on initial values and they are effective in obtaining solution of aeroengine component-level model.