基于退火单亲算法的压气机叶片排序

A opmizing sequencing of compressor vanes based on simulated annealing partheno genetic algorithm

压气机叶片排序问题属于完全NP难题,本文针对遗传算法在解决排列组合问题中的早熟收敛,提出了一种新的算法——退火单亲算法。并用退火单亲算法对压气机排序问题进行建模,在无需人工微调的情况下得到了比其他随机优化算法精度更高的叶片排序结果。实验结果表明退火单亲算法与遗传算法相比,具有寻优效果好、耗时少的优点,同时减少了航空发动机安全事故的发生概率。因此,将退火单亲算法运用在压气机叶片排序问题的求解上具有良好的有效性和可行性。

Optimizing sequencing of compressor vanes is related to the NP - completed problem. The precocious and convergence is inevitably emerging when Genetic algorithm is used in the combinatorial optimization. In this paper, a modified SA- PGA is suggested to build up a optimizing sequencing of compressor vanes, and a optimization result is found. The optimization result is superior to others without artilicial working. And,the results show that the improved Genetic algorithm （SA - PGA）is superior to traditional Genetic algorithm in both the result and the time. At the same time, it can reduce the probability of accident happened in engine rotor. So ,it is proved that the application of SA - PGA in optimizing sequencing of compressor vanes has a good validity and feasibility.