论文部分内容阅读
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。
The traditional methods of solving nonlinear equations and co-working equations in aeroengine component models do not always converge due to their dependence on initial values. First of all, based on the characteristics of the nonlinear equations in the model, a particle swarm optimization algorithm with HPSO1 is proposed by introducing the convergence factor, passive aggregation pressure factor and adaptive inertia weight into the algorithm. The solution to the common equations in the model is to introduce the convergence factor and passive aggregation pressure factor into the basic particle swarm optimization algorithm based on the basic particle swarm optimization algorithm and put forward another hybrid particle swarm optimization algorithm HPSO2. Experimental results show that both HPSO1 and HPSO2 overcome the dependence on the initial value, and thus it is effective to solve the engine component model.