Enhanced Quantum Neural Network Integrated in Health State Estimation for Aircraft Fuel System

As the core part of the airplane, aircraft fuel system has attracted much attention due to ever increasing demand on the enhancement of safety and reliability. In order to guarantee the flight safety and achieve condition based maintenance, the health states of aircraft fuel system should be estimated accurately. This paper illustrates the working principle and the core components of aircraft fuel system. A performance degradation model of aircraft fuel system is built based on the erosion of the boost pump. In this paper, an enhanced quantum neural network (EQNN)is developed to estimate the health state. The basic quantum neural network (QNN)has an inherently excellent capability for classification problem with overlapping boundary. In order to improve the performance, the loss function of the QNN is redesigned. Finally, EQNN is applied to estimate the health state of the aircraft fuel system. For validating of the performance of the proposed method, the proposed method is compared with other two classic neural networks.