The invention discloses a civil
airplane APU performance assessment and fault pre-warning method, and relates to the technical field of
equipment state monitoring. According to the method, based on QAR data of an
airplane, a health
state model of an important parameter in an APU is dug out by a
machine learning technology; then the predicted value of the health
state model is subtracted from the actual measured value of the APU
state parameter, and thus the deviation value of the
state parameter is acquired; then the deviation values of the several key parameters are fused into a health indexby using a regression model, and the
health index is used for representing the degradation degree of APU performance, and fault pre-warning is timely emitted according to a set warning value. The method provided by the invention is applied to the civil
airplane APU, particularly the APU that is faulted due to performance degradation, the problem of being lack of airplane APU performance assessmentmeans is solved, the performance state of the APU can be accurately assessed and the APU fault can be found at the early stage and the pre-warning can be timely emitted, and unplanned issuing eventscan be reduced; meanwhile, performance
assessment data is from the existing civil airplane QAR data, and the method is relatively strong in practicability and
operability.