Effective fault diagnosis of rotating machinery plays a pretty important role in the enhanced reliability and improved safety of industrial informatics applications. Although traditional intelligent fault diagnosis techniques such as support vector machine, extreme learning machine and convolutional neural network might achieve satisfactory accuracy, a very high price is caused by marking all samples manually. In this paper, a novel fault diagnosis method of the rotating machinery is proposed by integrating semi-supervised generative adversarial nets (SSGANs) with wavelet transform (WT), namely WT-SSGANs. The proposed WT-SSGANs method involves two parts. In the first part, WT is adopted to transform 1-D raw vibration signals into 2-D time-frequency images. In the second part, the 2-D time-frequency images are inputted into the built SSGANs model to realize fault diagnosis with little labeled samples. The advantage of the built model is that the unlabeled samples might be made full use of through an adversarial learning mechanism. Finally, two case studies are implemented to verify the proposed method. The results indicate that it can achieve higher accuracy and use less labeled samples than other existing methods in literatures. In addition, its performance in stability is pretty good as well. Competitive and promising results are still achieved when working conditions are changed.

Accurate and timely bearing fault diagnosis is crucial to decrease the probability of unexpected failures of rotating machinery and improve the efficiency of its scheduled maintenance. Since convolutional neural networks (CNN) have poor feature extraction capability for sensor data with 1D format, CNN combined with signal processing algorithm is often adopted for fault diagnosis. This increases manual conversion work and expertise dependence while reducing the feasibility and robustness of the corresponding fault diagnosis method. In this paper, a novel sensor data-driven fault diagnosis method is proposed by fusing S-transform (ST) algorithm and CNN, namely ST-CNN. First of all, a ST layer is designed based on S-transform algorithm. In the ST layer, sensor data is automatically converted into 2D time-frequency matrix without manual conversion work. Then, a new ST-CNN model is constructed, and the time-frequency coefficient matrixes are inputted into the constructed ST-CNN model. After the training process of the ST-CNN model is completed, the classification layer such as softmax performs the fault diagnosis. Finally, the diagnosis performance of the proposed method is evaluated by using two public available datasets of bearings. The experimental results show that the proposed method performs the higher and more robust diagnosis performance than other existing methods.

Many existing fault diagnosis methods based on deep learning (DL) require numerous fault samples to train the diagnosis model. However, in industrial applications, rotating machines (RMs) operate in normal states for most of their service life with fault events being rare and thus failure samples are very limited. To solve the problem above, a novel wide residual relation network (WRRN) is proposed for intelligent fault diagnosis of the RMs. Specifically, the WRRN is trained by performing a series of learning tasks in RMs with sufficient samples to obtain knowledge about how to diagnose, and then it is directly transferred to realize fault task of the RM with small samples. In this method, a wide residual network-based feature extraction module is used to generate representative fault features from input samples, and a relation module is designed to calculate the relation score between the sample pairs so as to determine their categories. Extensive experiments are conducted on two RMs to validate the WRRN method. The results demonstrate that the WRRN can accurately identify the fault types of the RMs with only small samples or even one sample. The WRRN significantly outperforms the existing popular methods in diagnostic performance.

Aircraft engine failures or damages not only incur substantial financial losses but also present risks of injuries or even fatalities. Hence, it is of utmost importance to devise an effective method to predict potential failures in advance, thereby mitigating accidents and minimizing losses. This paper proposes a novel approach that combines a principal component analysis (PCA) with similarity methods to establish a degradation trajectory database and predict the remaining useful life (RUL) of new engines by identifying similar trajectories. Firstly, the data dimensionality is reduced using a PCA to create a health indicator. The validity of the reduced data is confirmed by calculating the Spearman correlation coefficient between the health indicator and the system RUL. During the similarity comparison process, the Manhattan distance is employed for the similarity calculation, and parameter optimization is performed on the length of selected time segments and the number of chosen similar trajectories to optimize the similarity of RUL prediction model, resulting in the optimal prediction results among all engine test sets. Notably, this paper introduces the feasibility of employing the Manhattan distance in similarity method-based prediction, which is superior to the commonly used Euclidean distance calculation method found in most literature. This finding offers innovative ideas and perspectives for advancing RUL prediction methodologies. By adopting the proposed approach, the occurrence of accidents and losses associated with aircraft engine failures can be substantially reduced, leading to enhanced safety and economic benefits.

With the fast growth of embedded and handheld intelligent devices, the demand for lightweight fault diagnosis (FD) models of wind turbine gearboxes (WTG) is increasing in actual industrial scenarios. The existing lightweight FD approaches suffer from poor diagnostic performance and low robustness in complex operating environments. In this paper, a FD approach is proposed based on a new lightweight FD network (LFDNet) for high diagnostic performance of WTG. In LFDNet, the stem module, stage module and downsample module are respectively redesigned to reduce the model complexity and maintain superior and robust diagnostic capabilities, which enable it to diagnose the faults of WTG precisely, even with noisy interference. In the validation experiments on a self-made dataset and an open-source dataset, the diagnostic accuracy achieves 99.41% and 99.69%, respectively, which is higher than multiple lightweight deep learning methods. In addition, the anti-noise experiment also shows that the proposed approach has good noise robustness.