引入平均最好位置的量子粒子群算法及在航空发动机非线性模型求解中的应用

QPSO Imported Average Best Position and Application on Solving Nonlinear Model of Aeroengine

为改善粒子群算法的全局搜索能力和计算精度,在标准粒子群算法和量子理论的基础上,将平均最好位置引入粒子的状态更新过程,提出了1种基于量子行为改进的粒子群算法及不同的收缩-扩张因子取值策略。采用该量子粒子群算法对建立的典型双轴涡扇发动机部件级非线性模型进行了求解,结果表明:分段线性递减收缩-扩张因子适用于复杂的航空发动机隐式模型,其收敛效率和精度都较高,具有一定的工程应用价值。

To improve global searching ability and calculation accuracy of Particle Swarm Optimization （PSO） algorithm, the average best place into particles status updated on the standard PSO algorithm and quantum theory. An improved QPSO and corresponding different contraction-expansion factor deciding strategies were proposed, The algorithm was used to solve nonlinear mathematical model of typical two spool turbofan engine. The results show that piecewise linear diminishing eontraction--expausion factor is more suitable for complicated implicit model of aemengines, its convergence efficiency and precision are above normal. The proposed QPSO is of engineering application value.