Degradation signals that reflect a system's health state are important for diagnostics and health management of complex systems. However, degradation signals are often compounded and contaminated by measurement errors, making data analysis a difficult task. Motivated by the wear problem of magnetic heads used in hard disk drives (HDDs), this paper investigates Wiener processes with measurement errors. We explore the traditional Wiener process with positive drifts compounded with i.i.d. Gaussian noises, and improve its estimation efficiency compared with the existing inference procedure. Furthermore, to capture the possible heterogeneity in a population, we develop a mixed effects model with measurement errors. Statistical inferences of this model are discussed. The mixed effects model subsumes several existing Wiener processes as its limiting cases, and thus it is useful for suggesting an appropriate Wiener process model for a specific dataset. The developed methodologies are then applied to the wear problem of magnetic heads of HDDs, and a light intensity degradation problem of light-emitting diodes.

Prognostics of the remaining useful life (RUL) has emerged as a critical technique for ensuring the safety, availability, and efficiency of a complex system. To gain a better prognostic result, degradation information is quite useful because it can reflect the health status of a system. However, due to the lack of accurate information about the plants' degradation, the prognostic model is usually not well established. To solve this problem, this paper proposes a two-stage strategy that is in the context of data-driven modeling to predict the future health status of a bearing, where the degradation information was estimated by calculating the deviation of multiple statistics of vibration signals of a bearing from a known healthy state. Then, a prediction stage based on an enhanced Kalman filter and an expectation-maximization algorithm were used to estimate the RUL of the bearing adaptively. To verify the effectiveness of the proposed approach, a real-bearing degradation problem was implemented.

Recently, domain adaptation has received extensive attention for solving intelligent fault diagnosis problems. It aims to reduce the distribution discrepancy between the source domain and target domain through learning domain-invariant features. However, most existing domain adaptation methods mainly focus on global domain adaptation and overlook subdomain adaptation, which results in the loss of fine-grained information and discriminative features. To address this problem, in this article, a deep adversarial subdomain adaptation network is proposed. This network aligns the relevant distributions of subdomains by minimizing the local maximum mean discrepancy loss of the same categories in the source domain and target domain. Under the constraints of global domain adaptation and subdomain adaptation, the distribution discrepancy is reduced from the domain and category levels. Four transfer tasks under different machine rotating speeds and six transfer tasks on different but related machines were used to evaluate the effectiveness of the proposed method. The results demonstrated the robustness and superiority of the proposed method over five other methods.

Convolutional neural networks (CNNs) have demonstrated promising effectiveness in vibration-based fault diagnosis. However, the faulty characteristics are usually distributed on different scales and contaminated by noises from various sources. Therefore, it is still a challenging task for traditional CNNs to efficiently extract multiscale features and suppress unrelated noises in vibrational signals. In this article, a novel fault diagnosis framework called multiscale residual attention CNN (MRA-CNN) is proposed to learn discriminative multiscale features from vibrational signals and fully utilize the multiscale features to reduce noises. First, the raw vibrational signals are fed into the multiscale learning module (MLMod) to obtain multichannel feature maps generated with different kernel sizes. Second, the feature maps are passed through an efficient residual attention module (RAMod) to get the attention mask for weighing all locations of different channels of the multiscale feature maps to denoise. Third, to mitigate the information loss in attention, a new strategy called residual attention learning (RAL) is proposed to improve the feature extraction ability of RAMod, in which the learned attention mask itself is also regarded as feature maps by a shortcut connection. Experimental validation is conducted on two bearing datasets. The results show that the proposed method can learn more effective features from vibrational signals and deliver much higher accuracy than the seven state-of-the-art methods under highly noisy environments.

A hard disk drive (HDD) failure may cause serious data loss and catastrophic consequences. Online health monitoring provides information about the degradation trend of the HDD, and hence the early warning of failures, which gives us a chance to save the data. This paper developed an approach for HDD anomaly detection using Mahalanobis distance (MD). Critical parameters were selected using failure modes, mechanisms, and effects analysis (FMMEA), and the minimum redundancy maximum relevance (mRMR) method. A self-monitoring, analysis, and reporting technology (SMART) data set is used to evaluate the performance of the developed approach. The result shows that about 67% of the anomalies of failed drives can be detected with zero false alarm rate, and most of them can provide users with at least 20 hours during which to backup the data.

Predicting the impending failure of hard disk drives (HDDs) is crucial for preventing essential data from losing. In this paper, a two-step parametric method was developed to predict the impending failure of HDDs using the aggregate of statistical models. This method deals with the problem of failure prediction in two steps: anomaly detection and failure prediction. First, Mahalanobis distance was used for aggregating all the monitored variables into one index, which was then transformed into Gaussian variables by Box-Cox transformation. By defining an appropriate threshold, anomalies in HDDs were detected as a result. Second, a sliding-window-based generalized likelihood ratio test was proposed to track the anomaly progression in an HDD. When the occurrence of anomalies in a time interval is found to be statistically significant, indicating the HDD is approaching failure. In this work, we also derived a new cost function to adjust the prediction rate. This is important in a way to balance the failure detection rate and false alarm rate as well as to provide an advanced warning of HDD failures to the users, whereby the users can back up their data in time. Then the developed method was applied on a synthetic data set showing its effectiveness on predicting failures. To demonstrate the practical usefulness, this method was also applied on a real-life HDD data set. The result shows that our method could achieve 68% failure detection rate with 0% false alarm rate. This is much better than the results achieved by the state-of-the-art methods, such as support vector machine and hidden Markov models.