Accurate and robust health measurement for rolling bearings under variable working conditions has great significance to guarantee the safe and stable operation of rotating machinery. In this paper, a two-stage and working condition-robust health measurement method is proposed, systematically blending energy entropy theory, deep learning approach and transfer learning technology. In the first stage, a state boundary of energy entropy is systematically deduced based on the adaptive variational mode decomposition (VMD) – improved fruit fly optimization algorithm (IFOA) and statistical analysis principle to detect the abnormal states of bearings, where the IFOA is developed for searching the optimal parameters of VMD with high efficiency. In the second stage, if fault exists, a hybrid robust auto-encoder (HRAE) adopting the multi-layer and deep structure is constructed to strengthen the features extraction capacity and automatically capture valuable and robust fault features from original samples. Considering the insufficient labeled samples and significant data distribution discrepancy, a novel dynamic adversarial transfer network (DATN) is designed to extract the transferable and domain-invariant features between source and target datasets and achieve accurate fault identification. Specifically, a dynamic adversarial coefficient based on Wasserstein distance is provided in DATN to quantitatively evaluate the relative importance of marginal and conditional distributions. Extensive experiments on two rolling bearing datasets validate the more superior performance of the proposed method compared with other state-of-the-art identification models and transfer learning approaches.

No abstract