Empirical mode reconstruction: Preserving intrinsic components in data augmentation for intelligent fault diagnosis of civil aviation hydraulic pumps

A problem in data-driven fault diagnosis of civil aviation hydraulic pumps is that the faulty samples are much fewer than the healthy samples. To solve this problem, this paper develops a data augmentation method, namely empirical mode reconstruction (EMR), to augment faulty samples which preserve the intrinsic components in the original real samples. A significant property of the developed EMR is that the augmented samples are different but share very similar characteristics and category with the corresponding real samples, to properly guide the training of deep learning models, with the ultimate goal of yielding high diagnostic accuracies. First, the faulty training samples are converted to a series of intrinsic mode functions using empirical mode decomposition. Second, an intrinsic mode function is randomly selected and rescaled with a weight randomly selected from a properly predefined range. Third, these intrinsic mode functions are used to reconstruct the 1-dimensional samples, which serve as the augmented samples. Besides, the mean values and standard deviations of the augmented samples are kept the same with the corresponding original sample. Finally, the efficacy of the developed EMR in imbalanced fault diagnosis of civil aviation hydraulic pumps is validated through a group of experimental comparisons. (C) 2021 Elsevier B.V. All rights reserved.