基于信息融合遗传算法的航空发动机气路故障诊断

针对航空发动机由传统定期维修向视情维修转变的发展趋势,研究了遗传算法在气路性能参数估计中的应用,提出了改进的基于信息融合的遗传算法,将多源信息融合到遗传算法中,用以减小遗传算法的搜索范围.仿真结果表明:通过融合多源信息,有效克服遗传算法在寻优过程中陷入局部最优的情况,从而提高了气路性能参数的估计精度,能将估计误差控制在5%以内.

Aero-engine gas path fault diagnosis based on genetic algorithm of information fusion

For the development trend of aero-engines from traditional regular maintenance to condition-based maintenance, the genetic algorithm was studied in the gas path performance parameters estimation, and the improved genetic algorithm based on information fusion was proposed. The multi-source information was fused with genetic algorithm to reduce search range of genetic algorithm. The simulation results prove that the situation that genetic algorithm might fall into local optimum in optimization process can be effectively overcome by multi-source information fusion. The estimation accuracy of gas path performance parameters can be improved evidently, and the estimation error can be controlled within 5% or less.