基于GD-GEP算法的航空发动机振动监控

Vibration Monitoring of Aircraft Engine Based on GD-GEP

摘要:

针对传统基因表达式编程算法(GEP)在进行函数挖掘时易陷入局部最优以及收敛精度不高等问题,提出了一种基于基因多样性的GEP算法,该算法采用基于基因多样性的初始种群生成策略,引入了种群更新策略,使得种群基因多样性增加;仿真试验表明,该算法具有更好的全局搜索能力和更高的解精度;基于某型飞机飞行数据记录器记录的飞行数据,利用GD-GEP算法进行航空发动机模型辨识,将辨识得到的模型用于发动机振动值监控,真实的振动故障数据验证了该方法能够成功监测到振动故障。

Abstract:

In order to escape from local optimum and get a higher accuracy, GEP based on gene diversity is proposed, which using initial population producing strategy based on gene diversity, introducing the population update strategy, improved genes diversity of population. The simulation results show that GD_GEP has a better global search ability and higher accuracy. Using the flight data recorded by FDR of a certain type of aircraft, the model of aircraft engine based on GD-GEP is set up and used to monitor the vibration of engine, real malfunction data proved the method to be effective.