A comparative research based on three different algorithms for fault diagnosis in gas turbine

In view of disadvantage of neural network in gas turbine fault diagnosis, other algorithms are used in this article to improve fault diagnosis in gas turbine. With the application of MATLAB programming, three different algorithms, which are Back Propagation (BP) neural network, Particle Swarm Optimization-Back Propagation (PSO-BP) neural network and Lenvenberg Marquardt-Back Propagation (LM-BP) neural network, are studied and used to compare the performance of gas turbine fault diagnosis. In the models of this paper, the results show that gas turbine fault diagnosis based on LM-BP neural network has the best diagnosis identification ratio and the fastest diagnosis identification rate.