Data-driven bearing-fault diagnosis methods have become a research hotspot recently. These methods have to meet two premises: (1) the distributions of the data to be tested and the training data are the same; (2) there are a large number of high-quality labeled data. However, machines usually work under different working conditions in practice, which challenges these prerequisites due to the fact that the data distributions under different working conditions are different. In this paper, the one-dimensional Multi-Scale Domain Adaptive Network (1D-MSDAN) is proposed to address this issue. The 1D-MSDAN is a kind of deep transfer model, which uses both feature adaptation and classifier adaptation to guide the multi-scale convolutional neural network to perform bearing-fault diagnosis under varying working conditions. Feature adaptation is performed by both multi-scale feature adaptation and multi-level feature adaptation, which helps in finding domain-invariant features by minimizing the distribution discrepancy between different working conditions by using the Multi-kernel Maximum Mean Discrepancy (MK-MMD). Furthermore, classifier adaptation is performed by entropy minimization in the target domain to bridge the source classifier and target classifier to further eliminate domain discrepancy. The Case Western Reserve University (CWRU) bearing database is used to validate the proposed 1D-MSDAN. The experimental results show that the diagnostic accuracy for the 12 transfer tasks performed by 1D-MSDAN was superior to that of the mainstream transfer learning models for bearing-fault diagnosis under variable working conditions. In addition, the transfer learning performance of 1D-MSDAN for multi-target domain adaptation and real industrial scenarios was also verified.

Addressing the problem of online fault detection of a temperature sensor, a fault detection algorithm based on quadratic wavelet transform is proposed in this paper. First, the discrete wavelet transform is used to extract noise from the process signal; since the process noise signal is related to the internal structure and status of the sensor, a Butterworth low‐pass filter is used to filter out the high‐frequency electrical interference signal to obtain the process noise signal. Second, the continuous wavelet transform is used to detect abrupt fault from the process noise signal. Experimental results show that the noise‐analysis‐based fault detection algorithm (quadratic wavelet transform) is better than the process‐signal‐analysis‐based approach and that the quadratic wavelet transform is feasible for thermowell drop fault detection of the temperature sensor. © 2018 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.

Electromagnetic coils are a key component in a variety of systems and are widely used in many industries. Because their insulation usually fails suddenly and can have catastrophic effects, degradation monitoring of coil insulation systems plays a vital role in avoiding unexpected machine shutdown. The existing insulation degradation monitoring methods are based on assessing the change of coil high-frequency electrical parameter response, whereas the effects of the insulation failure mechanisms are not considered, which leads to inconsistency between experimental results. Therefore, this paper investigates degradation monitoring of coil insulation systems under thermal loading conditions from a creep point of view. Inter-turn insulation creep deformation is identified as a quantitative index to manifest insulation degradation changes at the micro level. A method is developed to map coil high-frequency electrical monitoring parameters to inter-turn insulation creep deformation in order to bridge the gap between the micro-level and macro-level changes during the incipient insulation degradation process. Thermally accelerated tests are performed to validate the developed method. The mapping method helps to determine the physical meaning of coil electrical monitoring parameters and presents opportunities for predictive maintenance of machines that incorporate electromagnetic coils.

One of the biggest challenges for the fault diagnosis research of industrial robots is that the normal data is far more than the fault data; that is, the data is imbalanced. The traditional diagnosis approaches of industrial robots are more biased toward the majority categories, which makes the diagnosis accuracy of the minority categories decrease. To solve the imbalanced problem, the traditional algorithm is improved by using cost-sensitive learning, single-class learning and other approaches. However, these algorithms also have a series of problems. For instance, it is difficult to estimate the true misclassification cost, overfitting, and long computation time. Therefore, a fault diagnosis approach for industrial robots, based on the Multiclass Mahalanobis-Taguchi system (MMTS), is proposed in this article. It can be classified the categories by measuring the deviation degree from the sample to the reference space, which is more suitable for classifying imbalanced data. The accuracy, G-mean and F-measure are used to verify the effectiveness of the proposed approach on an industrial robot platform. The experimental results show that the proposed approach’s accuracy, F-measure and G-mean improves by an average of 20.74%, 12.85% and 21.68%, compared with the other five traditional approaches when the imbalance ratio is 9. With the increase in the imbalance ratio, the proposed approach has better stability than the traditional algorithms.

Electromagnetic coils are one of the key components of many systems. Their insulation failure can have severe effects on the systems in which coils are used. This paper focuses on insulation degradation monitoring and remaining useful life (RUL) prediction of electromagnetic coils. First, insulation degradation characteristics are extracted from coil high-frequency electrical parameters. Second, health indicator is defined based on insulation degradation characteristics to indicate the health degree of coil insulation. Finally, an insulation degradation model is constructed, and coil insulation RUL prediction is performed by particle filtering. Thermal accelerated degradation experiments are performed to validate the RUL prediction performance. The proposed method presents opportunities for predictive maintenance of systems that incorporate coils.