Trajectory Similarity Matching and Remaining Useful Life Prediction Based on Dynamic Time Warping

Remaining useful life prediction based on trajectory similarity is a typical example of instance-based learning. Hence, Euclidean distance has the problems matching and low accuracy. Therefore, an engine remaining (RUL) method dynamic time warping (DTW) proposed. First, aiming at problem structure complexity multiple monitoring parameters, principal component analysis used to reduce dimension multisensor signals. Then, system performance degradation extracted kernel regression. After obtaining database, carried out DTW. finding best curve, RUL can be predicted. Finally, proposed verified by public aeroengine simulation dataset NASA, compared with several representatives high-precision literature methods same dataset, which verifies effectiveness method.