Aircraft Complex System Diagnosis Based on Design Knowledge and Real-Time Monitoring Information

摘要

In current aircraft maintenance, diagnostic and troubleshooting procedures may sometimes consume great time and energy due to large uncertainty especially for highly integrated systems. This study is intended to reduce the diagnostic uncertainty by considering both design knowledge and operational monitoring information. An aircraft brake system is selected as the typical electromechanical system with frequent fault occurrence. A hierarchical Bayesian network is constructed based on fault mode and effect analysis and system composition. This Bayesian network modeling enables a combination of fault mode and effect analysis on safety analysis as prior knowledge and real-time monitoring events as observation information. A detailed example on posterior update is illustrated followed by sensitivity analyses in parameter setting. This intelligent fault diagnostic approach has the potential to improve the efficiency and accuracy of aircraft system diagnosis.