Monitoring System for Industry 4.0: AI-Driven, Data Analysis and Health Maintenance

A special issue of Applied Sciences (ISSN 2076-3417). This special issue belongs to the section "Computing and Artificial Intelligence".

Deadline for manuscript submissions: closed (10 January 2024) | Viewed by 3813

Special Issue Editors

Special Issue Information

Dear Colleagues,

With the development of IoT and “Industry 4.0”, industrial systems become more intelligent and complex, and monitoring systems’ health is very important to guarantee stability, security, and economy. This shift also concerns diverse research areas, e.g., detection of abnormal data, unhealthy status, fault diagnosis, adversarial attacks, robustness analysis, and so on. On the other hand, with the development of sensor systems, large quantities of data have become easily available, bringing challenges to industrial systems’ condition monitoring.

A number of methodologies and algorithms related to data mining, big data analysis, and deep learning have been developed in this research area. However, there are still many challenging problems worth exploring and solving. Therefore, this Research Topic aims to select potential contributions related to advanced theoretical findings, technologies, algorithms, and industrial applications in the monitoring of industrial systems’ health (i.e., condition monitoring).

Subtopics of interest include:
• Theory development on monitoring systems’ health:
-Machine learning;
-Deep learning;
-Data Mining;
-Big data analytics;
-Graph theory.
• Engineering applications related to monitoring systems’ health:
-Data cleaning;
-Abnormal data detection;
-Anomaly detection;
-Condition monitoring;
-Fault diagnosis.
• Anomaly detection in energy-related industrial systems.

Dr. Yusen He
Dr. Huajin Li
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Applied Sciences is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • data-driven modeling
  • data mining
  • big data
  • machine learning
  • deep learning
  • condition monitoring
  • anomaly detection
  • fault diagnosis

Published Papers (5 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

20 pages, 5266 KiB  
Article
Unsupervised Deep Anomaly Detection for Industrial Multivariate Time Series Data
by Wenqiang Liu, Li Yan, Ningning Ma, Gaozhou Wang, Xiaolong Ma, Peishun Liu and Ruichun Tang
Appl. Sci. 2024, 14(2), 774; https://doi.org/10.3390/app14020774 - 16 Jan 2024
Viewed by 670
Abstract
With the rapid development of deep learning, researchers are actively exploring its applications in the field of industrial anomaly detection. Deep learning methods differ significantly from traditional mathematical modeling approaches, eliminating the need for intricate mathematical derivations and offering greater flexibility. Deep learning [...] Read more.
With the rapid development of deep learning, researchers are actively exploring its applications in the field of industrial anomaly detection. Deep learning methods differ significantly from traditional mathematical modeling approaches, eliminating the need for intricate mathematical derivations and offering greater flexibility. Deep learning technologies have demonstrated outstanding performance in anomaly detection problems and gained widespread recognition. However, when dealing with multivariate data anomaly detection problems, deep learning faces challenges such as large-scale data annotation and handling relationships between complex data variables. To address these challenges, this study proposes an innovative and lightweight deep learning model—the Attention-Based Deep Convolutional Autoencoding Prediction Network (AT-DCAEP). The model consists of a characterization network based on convolutional autoencoders and a prediction network based on attention mechanisms. The AT-DCAEP exhibits excellent performance in multivariate time series data anomaly detection without the need for pre-labeling large-scale datasets, making it an efficient unsupervised anomaly detection method. We extensively tested the performance of AT-DCAEP on six publicly available datasets, and the results show that compared to current state-of-the-art methods, AT-DCAEP demonstrates superior performance, achieving the optimal balance between anomaly detection performance and computational cost. Full article
Show Figures

Figure 1

17 pages, 4763 KiB  
Article
Transfer Learning-Based Remaining Useful Life Prediction Method for Lithium-Ion Batteries Considering Individual Differences
by Borui Gu and Zhen Liu
Appl. Sci. 2024, 14(2), 698; https://doi.org/10.3390/app14020698 - 14 Jan 2024
Viewed by 580
Abstract
With the wide utilization of lithium-ion batteries in the fields of electronic devices, electric vehicles, aviation, and aerospace, the prediction of remaining useful life (RUL) for lithium batteries is important. Considering the influence of the environment and manufacturing process, the degradation features differ [...] Read more.
With the wide utilization of lithium-ion batteries in the fields of electronic devices, electric vehicles, aviation, and aerospace, the prediction of remaining useful life (RUL) for lithium batteries is important. Considering the influence of the environment and manufacturing process, the degradation features differ between the historical batteries and the target ones, and such differences are called individual differences. Currently, lithium battery RUL prediction methods generally use the characteristics of a large group of historical samples to represent the target battery. However, these methods may be vulnerable to individual differences between historical batteries and target ones, which leads to poor accuracy. In order to solve the issue, this paper proposes a prediction method based on transfer learning that fully takes individual differences into consideration. It utilizes an extreme learning machine (ELM) twice. In the first stage, the relationship between the capacity degradation rate and the remaining capacity is constructed by an ELM to obtain the adjusting factor. Then, an ELM-based transfer learning method is used to establish the connection between the remaining capacity and the RUL. Finally, the prediction result is adjusted by the adjusting factor obtained in the first stage. Compared with existing typical data-driven models, the proposed method has better accuracy and efficiency. Full article
Show Figures

Figure 1

17 pages, 8947 KiB  
Article
The Influence of Vertical Seismic Acceleration on the Triggering of Landslides Constrained by Bedding Faults under an Inertial Frame Reference: The Case of the Daguangbao (DGB) Landslide
by Guoping Xiang, Tao Jiang, Qingwen Yang, Shenghua Cui, Ling Zhu, Yuhang He and Huajin Li
Appl. Sci. 2023, 13(23), 12911; https://doi.org/10.3390/app132312911 - 02 Dec 2023
Viewed by 536
Abstract
The Daguangbao (DGB) landslide was the largest landslide that was triggered by the 2008 Wenchuan earthquake with a magnitude of Ms8.0. The sliding surface of this landslide was constrained on a bedding fault 400 m below the ground surface. Seismic records show that [...] Read more.
The Daguangbao (DGB) landslide was the largest landslide that was triggered by the 2008 Wenchuan earthquake with a magnitude of Ms8.0. The sliding surface of this landslide was constrained on a bedding fault 400 m below the ground surface. Seismic records show that the landslide suffered not only from strong horizontal but also vertical ground shaking that was almost equal to the horizontal component. In this study, to reveal the landslide triggering mechanism of the DGB landslide, this study ignores the steep dipping tension fracture section and the leading edge-locking section of the trailing edge of the DGB landslide, and the geological model of the large optical package landslide is generalized into a block model with the bottom controlled slip soft zone as the interface. Based on the improved Newmark method that considers vertical ground motion, the three-way seismic acceleration data and the shear strength parameter of the sliding surface being taken as a variable are used to calculate the cumulative permanent displacement of the slider. Then, by considering the cumulative permanent displacement ratio of vertical seismic acceleration or not and the cumulative permanent displacement ratio value considering the inertial force as the index, the response characteristics of the cumulative permanent displacement of the block-to-vertical ground motion and inertial forces were analyzed. The results show that both the horizontal inertial force and the vertical acceleration significantly increased the permanent displacement. The permanent displacement is 4.9 cm when considering the vertical acceleration, whereas it is only 2.0 cm without taking this into account. The contribution of vertical acceleration is significantly enlarged (87.8–90.7%) by the decreasing of the internal friction angle of the slide surface, while it is less influenced (5–27.4%) by the cohesion. Compared with the lower shear strength parameter of the sliding surface, the contributions of vertical acceleration and inertial force to the permanent displacement are more obvious when the shear strength parameter of the sliding surface is higher. When ϕ > 18°, the D/D* is greater than 1, and the maximum D/D* reaches 7. The fast accumulation event of permanent displacement is triggered in the concentration stage of the seismic energy release. In the DGB landslide area, 50% of the energy is released within 30–50 s, as indicated by the acceleration peaks recorded at the nearest seismic station, Qingping station. It is assumed that the DGB landslide may be triggered at 30–50 s due to half of the seismic energy being released during that time span. Full article
Show Figures

Figure 1

11 pages, 13525 KiB  
Article
Noise-to-Norm Reconstruction for Industrial Anomaly Detection and Localization
by Shiqi Deng, Zhiyu Sun, Ruiyan Zhuang and Jun Gong
Appl. Sci. 2023, 13(22), 12436; https://doi.org/10.3390/app132212436 - 17 Nov 2023
Viewed by 593
Abstract
Anomaly detection has a wide range of applications and is especially important in industrial quality inspection. Currently, many top-performing anomaly detection models rely on feature embedding-based methods. However, these methods do not perform well on datasets with large variations in object locations. Reconstruction-based [...] Read more.
Anomaly detection has a wide range of applications and is especially important in industrial quality inspection. Currently, many top-performing anomaly detection models rely on feature embedding-based methods. However, these methods do not perform well on datasets with large variations in object locations. Reconstruction-based methods use reconstruction errors to detect anomalies without considering positional differences between samples. In this study, a reconstruction-based method using the noise-to-norm paradigm is proposed, which avoids the invariant reconstruction of anomalous regions. Our reconstruction network is based on M-net and incorporates multiscale fusion and residual attention modules to enable end-to-end anomaly detection and localization. Experiments demonstrate that the method is effective in reconstructing anomalous regions into normal patterns and achieving accurate anomaly detection and localization. On the MPDD and VisA datasets, our proposed method achieved more competitive results than the latest methods, and it set a new state-of-the-art standard on the MPDD dataset. Full article
Show Figures

Figure 1

14 pages, 6667 KiB  
Article
Research on a Ship Mooring Motion Suppression Method Based on an Intelligent Active Anti-Roll Platform
by Feng Gao, Yougang Tang, Chuanqi Hu and Xiaolei Xie
Appl. Sci. 2023, 13(13), 7979; https://doi.org/10.3390/app13137979 - 07 Jul 2023
Cited by 1 | Viewed by 899
Abstract
Conventional ship mooring in ports has many shortcomings such as a high safety risk, low efficiency and high labor intensity. In order to explore and develop the theory and key technologies of intelligent automatic mooring systems, this paper takes an intelligent mooring system [...] Read more.
Conventional ship mooring in ports has many shortcomings such as a high safety risk, low efficiency and high labor intensity. In order to explore and develop the theory and key technologies of intelligent automatic mooring systems, this paper takes an intelligent mooring system based on a parallel anti-rolling mechanism as the research and development object. A new mooring method integrating ship hydrodynamics, mechanism kinematics and intelligent algorithms is proposed. Through numerical simulation and comparative analysis of the model, the motion inhibition effect of mooring ships under different working conditions is obtained. The results show that the control strategy and intelligent algorithm of the system can realize the active control of the wharf mooring ships and achieve the goal of improving wharf stability conditions through an intelligent mooring system. Full article
Show Figures

Figure 1

Back to TopTop