Interval-valued Blind Source Separation Applied to AI-based Prognostic Fault Detection of Aircraft Engines
Alvaro Martinez, Luciano Sanchez and Ines Couso
The design of user-friendly plots of Equipment Health Management (EHM) data for prognostic fault detection of aircraft engines is addressed. EHM plots link trend shift signatures, originated in cruise data of the engine being diagnosed, either with prototypes of specific known events or abnormal signatures derived from service data. Abnormalities are expressed as thresholds that must not be exceeded. EHM data, prototype and abnormality signatures are regarded as a mix of different sources and transformed with a new computational procedure that extends Blind Source Separation to interval-valued data.