Deep learning-based anomaly-onset aware remaining useful life estimation of bearings

Systematic review of predictive maintenance and digital twin technologies challenges, opportunities, and best practices

Background

Maintaining machines effectively continues to be a challenge for industrial organisations, which frequently employ reactive or premeditated methods. Recent research has begun to shift its attention towards the application of Predictive Maintenance (PdM) and Digital Twins (DT) principles in order to improve maintenance processes. PdM technologies have the capacity to significantly improve profitability, safety, and sustainability in various industries. Significantly, precise equipment estimation, enabled by robust supervised learning techniques, is critical to the efficacy of PdM in conjunction with DT development. This study underscores the application of PdM and DT, exploring its transformative potential across domains demanding real-time monitoring. Specifically, it delves into emerging fields in healthcare, utilities (smart water management), and agriculture (smart farm), aligning with the latest research frontiers in these areas.

Methodology

Employing the Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) criteria, this study highlights diverse modeling techniques shaping asset lifetime evaluation within the PdM context from 34 scholarly articles.

Results

The study revealed four important findings: various PdM and DT modelling techniques, their diverse approaches, predictive outcomes, and implementation of maintenance management. These findings align with the ongoing exploration of emerging applications in healthcare, utilities (smart water management), and agriculture (smart farm). In addition, it sheds light on the critical functions of PdM and DT, emphasising their extraordinary ability to drive revolutionary change in dynamic industrial challenges. The results highlight these methodologies' flexibility and application across many industries, providing vital insights into their potential to revolutionise asset management and maintenance practice for real-time monitoring.

Conclusions

Therefore, this systematic review provides a current and essential resource for academics, practitioners, and policymakers to refine PdM strategies and expand the applicability of DT in diverse industrial sectors.

The combination model of CNN and GCN for machine fault diagnosis

Learning powerful discriminative features is the key for machine fault diagnosis. Most existing methods based on convolutional neural network (CNN) have achieved promising results. However, they primarily focus on global features derived from sample signals and fail to explicitly mine relationships between signals. In contrast, graph convolutional network (GCN) is able to efficiently mine data relationships by taking graph data with topological structure as input, making them highly effective for feature representation in non-Euclidean space. In this article, to make good use of the advantages of CNN and GCN, we propose a graph attentional convolutional neural network (GACNN) for effective intelligent fault diagnosis, which includes two subnetworks of fully CNN and GCN to extract the multilevel features information, and uses Efficient Channel Attention (ECA) attention mechanism to reduce information loss. Extensive experiments on three datasets show that our framework improves the representation ability of features and fault diagnosis performance, and achieves competitive accuracy against other approaches. And the results show that GACNN can achieve superior performance even under a strong background noise environment.

Explainable AI for Bearing Fault Prognosis Using Deep Learning Techniques

Predicting bearing failures is a vital component of machine health monitoring since bearings are essential parts of rotary machines, particularly large motor machines. In addition, determining the degree of bearing degeneration will aid firms in scheduling maintenance. Maintenance engineers may be gradually supplanted by an automated detection technique in identifying motor issues as improvements in the extraction of useful information from vibration signals are made. State-of-the-art deep learning approaches, in particular, have made a considerable contribution to automatic defect identification. Under variable shaft speed, this research presents a novel approach for identifying bearing defects and their amount of degradation. In the proposed approach, vibration signals are represented by spectrograms, and deep learning methods are applied via pre-processing with the short-time Fourier transform (STFT). A convolutional neural network (CNN), VGG16, is then used to extract features and classify health status. After this, RUL prediction is carried out with the use of regression. Explainable AI using LIME was used to identify the part of the image used by the CNN algorithm to give the output. Our proposed method was able to achieve very high accuracy and robustness for bearing faults, according to numerous experiments.

An enhanced CNN-LSTM remaining useful life prediction model for aircraft engine with attention mechanism

Remaining useful life (RUL) prediction is one of the key technologies of aircraft prognosis and health management (PHM) which could provide better maintenance decisions. In order to improve the accuracy of aircraft engine RUL prediction under real flight conditions and better meet the needs of PHM system, we put forward an improved CNN-LSTM model based on the convolutional block attention module (CBAM). First, the features of aircraft engine operation data are extracted by multi-layer CNN network, and then the attention mechanism is processed by CBAM in channel and spatial dimensions to find key variables related to RUL. Finally, the hidden relationship between features and service time is learned by LSTM and the predicted RUL is output. Experiments were conducted using C-MPASS dataset. Experimental results indicate that our prediction model has feasibility. Compared with other state-of-the-art methods, the RMSE of our method decreased by 17.4%, and the score of the prediction model was improved by 25.9%.

An Augmented Reality-Assisted Prognostics and Health Management System Based on Deep Learning for IoT-Enabled Manufacturing

With increasingly advanced Internet of Things (IoT) technology, the composition of workshop equipment has become more and more complex. Based on this, the rate of system performance degradation and the probability of fault have both increased. Owing to this, not only has the difficulty of constructing the remaining useful life (RUL) model increased but also the improvement in speed of maintenance personnel cannot keep up with the speed of equipment replacement. Therefore, an augmented reality (AR)-assisted prognostics and health management system based on deep learning for IoT-enabled manufacturing is proposed in this paper. Firstly, the feature extraction model based on Convolutional Neural Network-Particle Swarm Optimization (PSO-CNN) is proposed with the purpose of excavating the internal associations in large amounts of production data. Based on this, the high-accuracy RUL prediction is accomplished by Gate Recurrent Unit (GRU)-attention, which can capture the long-term and short-term dependencies of time series and successfully solve the gradient disappearance problem of RNN. Moreover, more attention will be paid to important content with the help of the attention mechanism. Additionally, high-efficiency maintenance guidance and visible instructions can be accomplished by AR. On top of this, the remote expert can offer help when maintenance personnel encounters tough problems. Finally, a real case was implemented in a typical IoT-enabled workshop, which validated the effectiveness of the proposed approach.

Systematic Literature Review on Visual Analytics of Predictive Maintenance in the Manufacturing Industry

The widespread adoption of cyber-physical systems and other cutting-edge digital technology in manufacturing industry production facilities may motivate stakeholders to embrace the idea of Industry 4.0. Some industrial companies already have different sensors installed on their machines; however, without proper analysis, the data collected is not useful. This systematic review's main goal is to synthesize the existing evidence on the application of predictive maintenance (PdM) with visual aids and to identify the key knowledge gaps in areas including utilities, power generation, industry, and energy consumption. After a thorough search and evaluation for relevancy, 37 documents were identified. Moreover, we identified the visual analytics of PdM, including anomaly detection, planning/scheduling, exploratory data analysis (EDA), and explainable artificial intelligence (XAI). The findings revealed that anomaly detection was a major domain in PdM-related works. We conclude that most of the literature lacks depth in terms of an overall framework that combines data-driven and knowledge-driven techniques of PdM in the manufacturing industry. Some works that utilized both techniques indicated promising results, but there is insufficient research on involving maintenance personnel's feedback in the latter stage of PdM architecture. Thus, there are still pertinent issues that need to be investigated, and limitations that need to be overcome before PdM is deployed with minimal human involvement.

Multi-Sensor GA-BP Algorithm Based Gearbox Fault Diagnosis

To address the problem of the low recognition rate of time-frequency domain methods gearbox fault identification, a method featuring decision-level fusion of DS evidence theory and GA-BP algorithm was proposed in the present study. Firstly, the fault data of each state of the gearbox was classified, based on which the time-frequency domain features were extracted and 19 significant features have been selected. Secondly, the accuracy of the traditional BP algorithm was compared with that of the GA-BP algorithm. On this basis, it has been concluded that the GA-BP algorithm is highly accurate, and the local diagnostic results obtained by the GA-BP algorithm have been used as the basic probability. Finally, the DS evidence theory is currently used to fuses with the GA. In addition, the final fault identification of the gearbox can be achieved by using the DS evidence theory and the multi-sensor local diagnosis results obtained by the GA-BP algorithm for decision fusion. The results of the simulations and experiments showed that the method proposed has improved accuracy over a single algorithm for fault identification of gearboxes, respectively.

Rolling Bearing Fault Diagnosis Based on Deep Learning and Autoencoder Information Fusion

The multisource information fusion technique is currently one of the common methods for rolling bearing fault diagnosis. However, the current research rarely fuses information from the data of different sensors. At the same time, the dispersion itself in the VAE method has asymmetric characteristics, which can enhance the robustness of the system. Therefore, in this paper, the information fusion method of the variational autoencoder (VAE) and random forest (RF) methods are targeted for subsequent lifetime evolution analysis. This fusion method achieves, for the first time, the simultaneous monitoring of acceleration signals, weak magnetic signals and temperature signals of rolling bearings, thus improving the fault diagnosis capability and laying the foundation for subsequent life evolution analysis and the study of the fault–slip correlation. Drawing on the experimental procedure of the CWRU’s rolling bearing dataset, the proposed VAERF technique was evaluated by conducting inner ring fault diagnosis experiments on the experimental platform of the self-research project. The proposed method exhibits the best performance compared to other point-to-point algorithms, achieving a classification rate of 98.19%. The comparison results further demonstrate that the deep learning fusion of weak magnetic and vibration signals can improve the fault diagnosis of rolling bearings.