A hierarchical structure built on physical and data-based information for intelligent aero-engine gas path diagnostics

Future trends in engine health management (EHM) systems are information fusion, advanced analytical methods, and the concept of Intelligent Engines. Machine Learning (ML)-based aero-engine gas path diagnostic methods promising under motivation these trends. However, previous ML-based structures rarely applied actual engineering practice because they purely mathematical lack physical insight or limited by error accumulation problem. Developing an accurate, flexible interpretable intelligent method has always posed a challenge, especially when knowledge is also available for more information. Instead modifying applying existing ML classification regression, this study proposes novel hierarchical to get into systems, build hierarchies automatically, recommend structures. The proposed evaluated against NASA model high-bypass two-spool turbofan engine. NASA's blind test case results show that Kappa Coefficient 0.693 at least 0.008 higher than other open literature. It been proved can quantify dependence relationships between fault classes enhanced information, best structure reduced complexity, solve problem improved accuracy. could support condition monitoring effectively exploiting data-based interpretability, flexibility, visibility, accuracy, reliability.