基于神经网络的飞机供水系统故障预测研究

Fault Prediction for Aircraft Water Supply System Based on Neural Network

对飞机水/废水系统进行基于数据驱动的故障预测算法仿真研究,鉴于系统故障原因、故障模式的复杂性,先基于AMESim软件建立系统仿真模型,并采取故障注入的方式得到系统关键部位故障数据(以空气压缩机为例)。在以上故障仿真模型的基础上,进一步采用神经网络故障预测方法来拟合系统性能衰退曲线。仿真验证表明,上述方法可以有效预测系统关键部件的性能衰退趋势及产品剩余寿命,便于后期维护保养,以保证系统的可靠性和安全性。

Data-driven fault prediction algorithm simulation for aircraft water/wastewater system was studied in the paper.In view of the complexity of the fault cause and fault mode of the system,a simulation model of the system was built based on AMESim software,and the fault injection method was used to get the fault data of key parts of the system(air compressor as an example).Based on the above fault simulation model,a fault prediction method based on neural network was further used to fit the performance degradation curve of the system.The simulation results indicate the higher precision of performance decline trend and remaining life,which can facilitate later maintenance to ensure the reliability and security of the system.