Study on performance measurement and fault diagnosis of engine in real-time based on neural network

In accordance with the historical data and the real-time data collected from the different sensors, a system of performance inspecting and fault diagnosis for engines in real-time was built up based on the theory of neural network. The features of this system are: (1) it combines the real-time data collecting with ANN to diagnose the fault. (2) It can diagnose faults from different aspects and different layers by several different ways. (3) The function of self-learning is realized by ANN module. Using this system, users can analyze and diagnose the faults quickly and easily. The hardware structure, the software function and the principles of working and practicability of this system were introduced. The functions, methods and actualized principles of the system were expounded emphatically. It was proved by the experiments that the system had merits both high accuracy and economical practicability.