改进的粒子群算法在可靠性优化设计中的应用

针对粒子群优化算法在处理约束问题时产生的不可行解,引用基于多级罚函数的约束处理方法。为了改进罚函数粒子群算法易早熟、后期收敛慢、易陷入局部最优解的缺点,提出了动态改变惩罚系数的改进粒子群算法。应用于几个经典的测试函数,都在较少的迭代次数内得到了高精度的优化解,验证了算法的有效性。以某一机械零部件的可靠性优化为例,建立了基于改进粒子群算法的可靠性优化设计模型。结果表明:该方法能快速有效地解决可靠性优化设计问题,计算结果明显优于常规的多级罚函数法。

Application of improved particle swarm optimization algorithm in design of reliability optimization

Considering particle swarm optimization(PSO) algorithm in dealing with infeasible solution arising from the constrained problems,this paper quotes from the constraint processing method based on multistage punishment function.In order to improve the problems of prematurity,slow convergence and falling into local,this paper presents an improved PSO that changes punish coefficient dynamically.Applied to some of the classic test function,all gets the optimal solution of high precision in less iteration times,and verifies the effectiveness of the algorithm.Take optimization design of reliability of mechanical parts and assemblies as an example,and establishment of its mathematical model based on improved particle swarm optimization algorithm.The results show that this method can quickly and effectively solve the problem of reliability design,the result is better than the conventional multistage punishment function.