Bp Neural Network-based Aero-engine Failure Modes And Failure Impact Analysis

As one of the most active fields in Artificial Intelligence and very important methods in system reliability analyse, Nerve Network (NN) could be used in aero-engine reliability analyse, solve the problems of lacking data and experience accumulating in traditional Failure Mode, Effects, and Criticality Analysis (FMECA).This paper combines NN technology with traditional FMECA tightly, accomplishes tedious FMECA by NN simulation automatically. It discusses BPNN framework, and a method to manage reliability data for automatic BPNN simulation. Then, a program based on BPNN was developed to carry out FMECA for one turbine engine.The practice proved that this new method had merit of network application to get the shortage of traditional FMECA, and improved the reliability analyse efficiency and objectivity.