6 October 2021 in IEEE Transactions on Power Electronics
Lithium-ion batteries are the main energy source of devices, and the estimation of their state-of-health (SOH) has become a hot point in prognostics and health management. However, many existing methods assume that training and testing data follow the same distribution. The model based on dataset under one working condition may be ineffective for the dataset under another working condition due to the distribution discrepancy. Thus, this article proposes a novel battery health prognostic model based on transfer learning. First, a novel transfer learning-based prognostic model, called deep domain adversarial network, is developed for SOH estimation of Lithium-ion batteries. Second, an unsupervised feature alignment metric is proposed, where maximum mean discrepancy and correlation alignment are considered simultaneously. Moreover, a generative adversarial learning is developed to guide the feature generator to provide the domain-invariant features. Finally, a novel feature generator, called dense bidirectional gated recurrent unit, is proposed to extract discriminate features from sensor signals. The effectiveness of DDAN for SOH estimation is verified on a battery dataset. The experimental results indicate that DDAN can effectively predict SOH of Lithium-ion battery, and significantly improve the performance of feature learning and knowledge transferring.
2 June 2021 in IEEE Transactions on Transportation Electrification
Battery health monitoring is crucial to reduce unexpected maintenance and realize maximum productivity. However, batteries generally work under dynamic and complex working conditions (i.e., different domains) in industrial applications. The distribution difference exists not only between source and target domain but also between multisource domains, it is a challenging problem to monitor battery health deterioration under multiple working conditions. This study proposes a novel transfer learning-based method, multisource domain adaption network (MSDAN), to address the problem of battery health degradation monitoring under different working conditions. First, a battery health degradation monitoring method is proposed based on multisource transfer learning. The knowledge learned from multiple operational conditions is transferred to the target condition. Second, multisource domain adaption is proposed to address the problem of distribution discrepancy of different domains, where unsupervised feature alignment metric and maximum mean discrepancy (MMD) are utilized to evaluate the difference between target domain and each source domain. Multisource adversarial learning is adopted to guide feature generator to provide domain-invariant features of different domains. Third, a feature generator is proposed to extract common and domain-specific features of different domains. The experimental results illustrate that MSDAN has a better performance of battery health prognostics than other single-source domain adaption methods.
29 April 2022 in Applied Soft Computing
13 July 2021 in IEEE/ASME Transactions on Mechatronics
Vibration signals are widely utilized for machinery fault diagnosis. However, the fault-related components (i.e., impulse) in vibration signals are often buried by strong background noises due to complex working conditions. Thus, it is challenging to directly extract discriminate features from vibration signals. In this article, a novel deep neural network (DNN), multiscale weighted morphological network (MWMNet), is proposed to extract impulses from vibration signals. First, a novel morphological layer is smoothly embedded in DNN as a signal processing layer to extract impulses and filter out the noise. Second, multiple branches with different structure element (SE) scales are employed to respectively extract impulses. An adaptive weighted fusion is utilized to enhance the scales that provide strong impulsive components. Third, a morphological operator, called average-hat transform, is adopted in MWMNet, and both positive and negative impulses can be extracted from vibration signals. The effectiveness of MWMNet is validated by the experiments of gearbox fault diagnosis and bearing fault diagnosis. The experimental results show that MWMNet can learn the fault-related features and filter out noise from vibration signals well. The comparison results illustrate that MWMNet has a better fault diagnosis performance than those DNNs, e.g., one-dimensional convolutional neural network, densely connected convolutional network, and residual network.
27 July 2022 in IEEE/ASME Transactions on Mechatronics
Machinery fault diagnosis is crucial to ensure production safety. Deep learning has been widely investigated in intelligent fault diagnosis. However, the mechanical equipment generally runs in a complex and polytropic condition. It is challenging to perform machinery fault diagnosis under polytopic conditions. In this article, a novel multisource domain adaptation method is proposed for machinery fault diagnosis under different working conditions. First, a deep negative correlation multisource domains adaptation network (DNC-MDAN) is proposed, where the information from multisource domains is transferred into target domain based on the multisource feature alignment and adversarial learning. Second, in order to make full of the multisource domains information, an ensemble classifier corresponding to multisource domains is developed, where a DNC learning is performed in the ensemble classifier. Third, an end-to-end feature generator, discrete cosine convolutional block, is proposed in DNC-MDAN, where discrete cosine transform is embedded in the network smoothly for noise reduction. In order to validate the effectiveness of DNC-MDAN for machinery fault diagnosis, three cases are considered in this article. In comparison with other transfer learning methods, DNC-MDAN has a better fault diagnosis performance.
7 July 2022 in IEEE Systems Journal
Deep neural networks have been widely applied in machinery health prognostics due to their powerful feature learning capacity. However, in many existing remaining useful life (RUL) prediction methods, the data distribution shift between multiple working conditions is ignored. In fact, there are inevitably certain differences between different working conditions. How to transfer knowledge to different but related domains for machine RUL prediction is not considered well. Thus, this article proposes a novel domain adaptation method named selective adversarial adaptation network (SAAN) for machine RUL prediction. A selective feature interaction is utilized in SAAN to implement knowledge selection, and the selective weights can be updated according to different data distribution. A new feature generator, residual convolutional long short-term memory is utilized for sequential information extraction from time-series input. Finally, two experiment on engine and bearing are utilized to validate the performance of SAAN for RUL prediction. The comparison results demonstrate that SAAN can implement knowledge transferring for machine RUL prediction.