Fault detection and diagnosis play a crucial role in ensuring the reliability and safety of modern industrial systems. For safety and cost considerations, critical equipment and systems in industrial operations are typically not allowed to operate in severe fault states. Moreover, obtaining labeled samples for fault diagnosis often requires significant human effort. This results in limited labeled data for many application scenarios. Thus, the focus of attention has shifted towards learning from a small amount of data. Few-shot learning has emerged as a solution to this challenge, aiming to develop models that can effectively solve problems with only a few samples. This approach has gained significant traction in various fields, such as computer vision, natural language processing, audio and speech, reinforcement learning, robotics, and data analysis. Surprisingly, despite its wide applicability, there have been limited investigations or reviews on applying few-shot learning to the field of mechanical fault diagnosis. In this paper, we provide a comprehensive review of the relevant work on few-shot learning in mechanical fault diagnosis from 2018 to September 2023. By examining the existing research, we aimed to shed light on the potential of few-shot learning in this domain and offer valuable insights for future research directions.

Pumps are one of the most critical machines in the petrochemical process. Condition monitoring of such parts and detecting faults at an early stage are crucial for reducing downtime in the production line and improving plant safety, efficiency and reliability. This paper develops a fault detection and isolation scheme based on an unsupervised machine learning method, sparse autoencoder (SAE), and evaluates the model on industrial multivariate data. The Mahalanobis distance (MD) is employed to calculate the statistical difference of the residual outputs between monitoring and normal states and is used as a system-wide health indicator. Furthermore, fault isolation is achieved by a reconstruction-based two-dimensional contribution map, in which the variables with larger contributions are responsible for the detected fault. To demonstrate the effectiveness of the proposed scheme, two case studies are carried out based on a multivariate data set from a pump system in an oil and petrochemical factory. The classical principal component analysis (PCA) method is compared with the proposed method and results show that SAE performs better in terms of fault detection than PCA, and can effectively isolate the abnormal variables, which can hence help effectively trace the root cause of the detected fault.

The marine engine is a complex-structured multidisciplinary system that operates in a harsh environment involving high temperatures and pressures and gas/fluid/solid interactions. Many malfunctions and faults can occur to the marine engine and efficient condition monitoring is critical to ensure the expected performance. In this paper, a marine engine test rig is established and its process data are recorded, including various temperatures and pressures. Two data-driven models, i.e., principal component analysis and the sparse autoencoder, and a physics-based model are applied to the marine engine for two classic faults, i.e., lubrication oil filter blocking and cylinder leakage. Comparative studies and discussions are conducted regarding their performance in terms of anomaly detection and fault isolation. The data points collected for the filter blocking fault are generally two times higher than the fault thresholds set by the data-driven models. In the physics-based model, it is observed that the lubrication oil pressure falls from the predicted 3.2–3.8 bar to around 2.3 bar. For the cylinder leakage fault, the fault test data are nearly four times higher than the thresholds in the data-driven models. The exhaust gas temperature of the leaked cylinder falls from an estimated 150–200 °C to about 100 °C. The transferability and interpretability of these models are finally discussed. The findings of the present study offer insights into the two types of models and can provide guidance for the effective condition monitoring of marine engines.

Substation noise is a crucial factor that influences residents’ quality of life, especially in the densely residential areas. Despite small- and medium-sized transformer facilities having relatively low noise levels, due to their proximity to residential areas, they generate considerable annoyance, rendering them a focal point among environmental noise complaints. The predominant noise emitted by these facilities falls within the medium- and low-frequency spectrum range, and the conventional passive noise reduction techniques exhibit limited efficacy in attenuating such low-frequency noise. This study develops a real-time active noise control (ANC) system based on a digital signal processor, TMS320F28335, and various ANC methods, including Filtered-X Least Mean Squares (FxLMS), Normalized Filter-X Least Mean Squares (FxNLMS), and variable step-size FxLMS (VS-FxLMS), are evaluated for the low-frequency noise reduction. In addition, the substation noises at a residential community are measured, analyzed, and used as noise source together with a series of sinusoidal waves for evaluation of the ANC algorithms. Results show the ANC system are effective in attenuating most low-frequency noises (within 600 Hz) and the average noise reduction for the substation noises has achieved by more than 12 dB.

Intelligent fault diagnosis methods based on deep supervised learning require massive labeled data, which contradicts the typical engineering scenarios. Industrial datasets usually have low label rates, because labeling large amounts of data requires a lot of labor costs and expert knowledge. To address the challenge in fault diagnosis under extremely low label rate, this article proposes a novel semi-supervised learning framework based on multiscale and multilevel contrast(MMCL), where representation learning on massive unlabeled data is integrated with joint time-frequency (JTF) domain mechanism on few labeled data to provide targeted assistance for fault diagnosis. Specifically, we utilize MMCL method to pretrain the encoder, which combines the correlation between signals and time dependency within signals to enable robust and intrinsic representation for each timestamp of signals. Furthermore, based on the pretrained encoder, we design JTF domain mechanism to enhance the generalization of representation from both time domain and frequency domain, which further enhance the overall diagnostic performance. The performance of the proposed method was verified using the Case Western Reserve University (CWRU) open dataset, PHM dataset and a self-built experimental dataset. Extensive experimental results on these two datasets demonstrated that the proposed method outperformed existing semi-supervised diagnosis methods under extremely low label rate.