Special Sessions
#1 Scientific Machine Learning for the Prognostics and Health Management of Energy Systems
Energy systems play a vital role in the functioning of industrial operations. Developing effective prognostics and health management (PHM) within energy systems are of utmost importance in ensuring their reliable and efficient operation, reducing downtime, and improving productivity. In the era of big data, decision support systems have progressed towards a high level of automation and intelligence in energy systems. Machine learning renders promising tools to extract useful knowledge from monitoring data to support the PHM. Despite significant progress, most of existing machine learning-powered decision support systems are purely data-driven and are often flawed under the widely used close-world assumption, i.e., the independently identically distribution (iid) between training and testing data. This hypothesis has accentuated many problems, such as the poor generalization to out-of-distribution (OOD) data. To address these challenges, it is meaningful to develop novel scientific machine learning techniques for scalable, domain-aware, robust, reliable, interpretable, and trustworthy data analysis and decision-making in prognostics and health management.
This special session is interested in collecting articles on the latest research progress and achievements of scientific machine learning for the prognostics and health management of energy systems. Potential topics include but are not limited to the following:
● Domain-aware and physics-Informed machine learning for PHM
● Interpretable/explainable machine learning for PHM
● Uncertainty quantification in machine learning towards reliable and trustworthy PHM in safety-critical applications.
● Out-of-distribution detection in intelligent fault diagnosis
● Machine learning with big data but small labels for PHM
● Machine learning with lower quality (noisy, heterogeneous, unstructured) data for PHM
Organizers:
Prof. Te Han, Beijing Institute of Technology, China, Email: hant15@tsinghua.org.cn
Prof. Haidong Shao, Hunan University, China, Email: hdshao@hnu.edu.cn
Dr. Ke Zhao, Chang’an University, China, Email: zhaoke@chd.edu.cn
#2 Magnetic flux Leakage detection theory and its applications for PHM
Rotating machines with transmission components (such as pumps, bearings, and gears) can naturally generate signals that can be mined for fault diagnosis; however, static equipment relies on certain excitation methods to forcedly generate signals for its fault diagnosis. The magnetic flux leakage (MFL) detection effect is one of the excitation methods, and it has been widely used in engineering practice. The challenge of this non-destructive testing method is how to well establish the link between the MFL signals and the fault of the static equipment. This special session focuses on the MFL detection technology and its application in fault diagnosis, with possible topics including but not limited to:
● Excitation module design and MFL sensor design
● MFL simulation and modelling
● Noise analysis and denoising method for MFL signals
● Quantitative or qualitative fault diagnosis methods based on the MFL effect
● Application of MFL detection
Organizers:
Prof. Zhiliang Liu, University of Electronic Science and Technology of China, China, China, Email: Zhiliang_Liu@uestc.edu.cn
Liyuan Ren, Institute of Applied Physics and Computational Mathematics, China，Email: migogoren@outlook.com
#3 Cross-domain fault diagnosis methods for rotating machines
Modern machines are becoming more automatic and efficient, which has increasingly high requirements for the reliability and quality. To ensure the reliable operation of rotating machines, it has always been an issue of significance to comprehensively and accurately diagnose the latent faults of the machinery. In recent years, transfer learning has been widely used in various fields of rotating machinery fault diagnosis due to its ability to learn cross domain features. This special session is interested in articles on the latest research progress and achievements of cross domain fault diagnosis methods for rotating machines. Potential topics include but are not limited to the following:
● Advanced signal processing techniques for cross domain feature extraction;
● Transfer learning–based intelligent fault diagnosis of machines;
● Fault detection of machines under varying speed conditions;
● Advanced sensing and monitoring techniques under variable working condition;
● Fault diagnosis methods based on cross-domain adaptive;
● Sensor fusion techniques under different domain data.
● Applications of domain generalization techniques for fault classification of bearing and gear.
Organizers:
Prof. Xingxing Jiang, Soochow University, China, Email: jiangxx@suda.edu.cn
Prof. Jinrui Wang, Shandong University of Science and Technology, China, Email: wangjinrui@sdust.edu.cn
Prof. Xiaoli Zhao, Nanjing University of Science and Technology, China, Email: xlzhao@njust.edu.cn
#4 Prognostics Health Management and Reliability
The field of Prognostics Health Management (PHM) has become increasingly important in recent years, particularly in the context of Industry 4.0. PHM involves detecting signs of equipment failure, modeling equipment aging and predicting its useful life, and understanding equipment health in order to make appropriate decisions. This technology has been applied in a wide range of real-life systems, including the Internet of Things (IoT) and power transmission systems. Reliability, which measures the probability of successful system operation, is a critical aspect of PHM and is of great interest to system engineers and researchers working to assess and enhance reliability in various systems. This special session at PHM 2023 will bring together top researchers and scholars in the field of Prognostics Health Management Reliability for Modern Systems in Real-life Applications.
The Prognostics Health Management Reliability has been acknowledged as an important research area that has received intensive concerns in the past few decades due to its universalization and significance in various kinds of fields. In recent years, we have seen an increasing interest in Prognostics Health Management Reliability for Modern Systems in Real-life Application. The Modern Systems perform an important and critical vehicle in numerous fields including: (1) Industry 4.0, (2) IoT, (3) Cloud Computing, (4) Block Chain, (5) Smart Grid, (6) Social Network
The special session will provide a platform for researchers and practitioners to exchange their ideas and findings, discuss the latest developments, and explore future directions in the field of Prognostics Health Management Reliability for Modern Systems in Real-life Application. Topics of interest include, but are not limited to:
● Data-driven methods for fault diagnosis and prognosis
● Uncertainty quantification and management in PHM
● Reliability and risk analysis of complex systems
● Health management and maintenance optimization
● PHM applications in manufacturing and production systems
● PHM applications in transportation systems
● PHM applications in energy systems
● PHM applications in medical devices and healthcare systems
● PHM applications in aerospace and defense systems
● PHM applications in information and communication systems
Organizers:
Chair Professor Wei-Chang Yeh, National Tsing Hua University, email: yeh@ieee.org
#5 Development of vibration based data driven nonlinear measures for dynamic health monitoring of civil and mechanical structures
With the advancement of technology, vibration based data driven health monitoring of civil and mechanical infrastructure has become a centre of attraction for the researchers. Due to the presence of unwanted shock, vibration and noise component, collected vibration signals are often associated with high degree of non-linearity. In this context, the most important aspect of a data driven health monitoring operation is the construction of nonlinear measures. Considering the aforementioned interest among the researchers, this special session aims to provide a platform to present high-quality original research on the latest developments of vibration based data driven health monitoring indices and its application to the prognostic and health management of civil and mechanical structures. Potential topics include but are not limited to the following:
● Health monitoring
● Data driven predictive maintenance
● Fault diagnosis and prognosis
● Condition monitoring
● Degradation assessment
● Early fault detection
● Nonlinear measure
● Condition based maintenance
Organizers:
Dr. Khandaker Noman, Northwestern Polytechnical University
Dr. Yongbo Li, Northwestern Polytechnical University
Dr. Tao Liu, Northwestern Polytechnical University
#6 Fault diagnosis and Intelligent Prognosis for Mechatronics System
The mechatronics systems are widely used in the complex engineering equipment. The research on the PHM of the mechanics systems the key component of the overall PHM system of the equipment. Fault diagnosis and intelligent prognosis is one of the critical enable techniques of PHM. This session focuses on the recent original research progress in the innovative methods, intelligent techniques and application cases of fault diagnosis and intelligent prognosis for the specific mechatronics systems.
Organizers:
Dr. Cheng Zhe, chengzhe@nudt.edu.cn
Dr. Bai Guanghan, baiguanghan@nudt.edu.cn
#7 Advanced Methods boosted PHM and Applications in Electro-Mechanical Systems
Prognostics and health management has been rapidly developed in recent years. The developments benefit from artificial intelligence algorithms, advanced sensing techniques, digital twin among others. Many developments and efforts have already been devoted to PHM for mechanical systems and electro-mechanical systems, such as industrial robots, wind turbines, aerospace equipment etc. However, in consideration of some special issues and difficulties about the equipment working conditions or the collected historical data, the development of effective PHM methods are challengeable. This special session focuses on advanced methods, innovative ideas and successful applications for electro-mechanical systems health condition monitoring and fault prognostics. The topics for this special session include but are not limited to the following:
● Methods for data quality improvement or data augmentation.
● Multi-feature fusion techniques
● Data-model interactive modeling
● Knowledge-driven diagnostic/prognostic algorithms
● Physics-informed degradation modeling
● Fault detection or prognosis considering complex issues
Organizers:
Lei Xiao, Associate Professor, Donghua University, China, Email: leixiao211@dhu.edu.cn
Diyin Tang, Associate Professor, Beihang University, China, Email: tangdiyin@buaa.edu.cn
#8 PHM based on Semantic Computing
Semantic computing plays an important role in equipment health management industrial software. Health management industrial software requires the participation of semantic technology from data acquisition, data processing, fault diagnosis and fault algorithm, knowledge search, health management question answering system, health management question answering robot, etc. Topics for submission in this session include but are not limited to:
● Health management content analysis based on semantics and rules.
● Health management knowledge base construction based on semantic technologies.
● Ontology integration for PHM knowledge.
● Semantic search engines and industrial software for PHM.
● Semantic Health management Q/A robot.
● Use of Semantics in PHM Applications for aircraft, engine and other complex equipment.
Organizers:
Guigang Zhang, Institute of Automation, Chinese Academy of Sciences, China. Email: guigang.zhang@ia.ac.cn.
Tengfei Zhang, Institute of Automation, Chinese Academy of Sciences, China. Email: tengfei.zhang@ia.ac.cn.
Guanghao Ren, Institute of Automation, Chinese Academy of Sciences, China. Email: guanghao.ren@ia.ac.cn.
Yun Wang, Institute of Automation, Chinese Academy of Sciences, China. Email: y.wang@ia.ac.cn.
Hua Ming, Institute of Automation, Chinese Academy of Sciences, China. Email: hua.ming@ia.ac.cn.
Nan Yang, Institute of Automation, Chinese Academy of Sciences, China. Email: nan.yang@ia.ac.cn.
#9 Prognostics and Health Management Applications using Multisource Data
The increasing complexity of modern industrial systems presents challenges for single sensor-based monitoring systems in ensuring optimal performance and preventing system failures. As a result, researchers have been developing advanced signal processing techniques that can handle multiple types of data from different sensors, providing accurate and timely information about the system's health. Multisource data-driven health management methods offer a promising solution to this challenge. These methods leverage data from various sensors, such as vibration, temperature, and pressure sensors, to provide a comprehensive understanding of the system's health status, enabling more effective predictive maintenance and reducing downtime and maintenance costs.
To solicit the latest advances in multisource data-driven health management applications in various industrial systems, this special issue invites submissions of both review and research papers on this topic. Topics of interest include, but are not limited to:
● Multiple signals processing approaches
● Condition state recognition under multivariate data
● Risk assessments for industrial systems using multiple signals
● Reliability prediction for industrial systems using multiple signals
● Remaining useful life prediction using multiple signals
● Fault diagnosis for industrial systems using multiple signals
● Fault prediction for industrial systems using multiple signals
● Condition-based maintenance schemes using multiple signals
Organizers:
Prof. Chaoqun Duan, Shanghai University, China, Email: chaoqun@shu.edu.cn
Dr. Shabbir Ahmed, Stanford University, USA, Email: ahmed07@stanford.edu
#10 Reliable Spectrum Sensing Technology
The growing demand for wireless applications imposes significant constraints on using limited and valuable radio spectrum resources. In order to minimize interference to authorized users and provide access to as many cognitive nodes as possible, spectrum-aware technologies are required to have high detection accuracy and detection efficiency. How to obtain reliable spectrum-aware technology has been a problem of great importance. This special session aims to solicit the latest research progress and results in spectrum sensing. Potential topics include, but are not limited to, the following:
● Cognitive radio
● Radio monitoring
● Spectrum-awareness
● Single-node spectrum awareness
● Dynamic spectrum management
● Collaborative Spectrum Awareness
● Machine Learning
Organizers:
Prof. Ying Li, Yantai University, China, Email: liying@ytu.edu.cn
Prof. Zhongxun Wang, Yantai University, China, Email: ytwzx3@126.com
Prof. Zhuoran Cai, Yantai University, China, Email: caizhuoran@ytu.edu.cn
#11 Prognostics and Health Management for the Power Devices and Power Conversion System
Power conversion system (PCS), such as electric drive and power supply, is the “Heart” of intelligent and electric equipment, such as aircraft, high-speed train and automobile and so on. And the power device is the most widely used and has the highest failure rate in power electronics devices. There are some defects about the traditional reliability assurance methods based on pre-life prediction and post-failure analysis, such as high cost, low efficiency and poor accuracy. Prognostics and health management (PHM) technology is based on failure physics, which is used to predict and evaluate the reliability of a product in a real environment, and major accidents may be avoided.
This special session is interested in articles on the latest research progress and achievements of failure mechanism/physical models and data-driven methods for power conversion systems and devices. Potential topics include but are not limited to the following:
● Health assessment/life prediction based on failure mechanism/physical model
● Fault diagnosis/prediction of power conversion system based on data-driven methods
● PHM for Power devices（IGBT,MOSFET et.al）
● PHM for Capacitors (electrolytic capacitors, film capacitors et.al)
● PHM for Power conversion systems (DC-DC,DC-AC,AC-DC et.al)
Organizers:
Yiqiang Chen, China Electronic Product Reliability and Environmental Testing Research Institute , National Key Laboratory of Science and Technology on Reliability Physics and Application of Electronic Component, Email: yiqiang-chen@hotmail.com
Linghui Meng, China Electronic Product Reliability and Environmental Testing Research Institute , National Key Laboratory of Science and Technology on Reliability Physics and Application of Electronic Component, Email: menglinghui@ceprei.com
#12 Reliability estimation and Fault Detection for Industrial Robots
Reliability estimation and fault detection for industrial robots is essential for maintenance decision making and contingency mitigation of the automatic production line. The fault or reliability of industrial robots can be predicted or estimated by vibration signal, current signal, attitude data, positioning data, etc. This invited session aims to bring experienced researchers in this field together to share ideas, results, and applications of reliability estimation and fault detection for the industrial robot.
The topics for this special session include but are not limited to the following:
● Electro-mechanical coupling modeling for the joint of robots
● Fault detection and prognostics for the robots based on motor current signal, attitude data, positioning data, etc.
● Remaining useful life prediction with hybrid model for the reducers of robots
Organizers:
Bai Guanghan，National University of Defense Technology, China, Email: baiguanghan@nudt.edu.cn
Wang Jia, Hebei University of Technology, China, Email: wangjia@hebut.edu.cn
Wu Jinhui, Hebei University of Technology, China, Email: wujinhui@hebut.edu.cn
#13 Methods and applications in PHM considering random effects
In academic research and engineering practice of reliability and PHM, random effects are often significant factors affecting the health states of products, and many studies on degradation or failure modeling use random effects to depict unit heterogeneity, which shows that random effects play a non-negligible role in PHM. Typical examples include mixed effects models, Wiener degradation processes considering random effects, maintenance decision optimization considering random effects, etc. This special session is interested in articles on the latest research progress and achievements of methods and applications in PHM considering random effects, innovations incorporating engineering problems are particularly welcomed. Potential topics include but are not limited to the following:
● Innovative applications of mixed effects models in reliability
● Wiener degradation processes with random effects
● Proportional hazards models with random effects
● Multi-layer nested models with random effects
● Remaining useful life prediction based on models with random effects
● Maintenance decision making and optimization considering random effects
Organizers:
Prof. Zengqiang Jiang, Beijing Jiaotong University, China, zqjiang@bjtu.edu.cn
Prof. Qi Li, Beijing Jiaotong University, China, liqi@bjtu.edu.cn
Prof. Mingcheng E, Beijing Jiaotong University, China, emch@bjtu.edu.cn
#14 Health Monitoring for Composite Material Structural
Composite materials have high specific strength, high specific modulus, fatigue resistance, as well as good designability, thermal resistance, corrosion resistance, and radar absorption and suppression capabilities. They are widely used in aerospace, mechanical, shipbuilding, power and other fields. The effects of stress and strain during service process will lead to matrix damage, fiber disadhesion or fracture, interlayer cracking and other internal damages of the composite materials, which will continue to expand and eventually lead to the destruction of the composite structure. Composite material structural health monitoring can accurately monitor the deformation and damage to provide warnings for failure. It is of great academic significance and engineering value to carry out composite health monitoring research to ensure the safe use of composite materials. This special session mainly focuses on monitoring, sensing, and damage mechanisms of composite material structures, with possible topics including but not limited to:
● Sensor fabrication technology for composite materials
● Integrated technology of sensors and composite materials
● Damage mechanism of composite materials
● Composite material damage and failure characterization technology
● Uncertainty analysis for composite materials
● Health analysis of composite materials based on intelligent algorithms
Organizers:
Assoc. Prof. Guijie Li, Dalian University of Technology, China, Email: ligj@dlut.edu.cn
Prof. Feng Zhang, Northwestern Polytechnical University, China, Email: nwpuwindy@nwpu.edu.cn
Dr. Zheng Wei, The Research Institute for Special Structures of Aeronautical Composite AVIC, China, Email: Danielwei1994@163.com
Assoc. Prof. Wei Li, Nanjing University of Aeronautics and Astronautics, China, Email: liweiair@nuaa.edu.cn
#15 Multi-physics based vibration and reliability analysis
Vibration has a significant impact on the reliability of mechanical equipment. An uncomfortable truth about modern mechanical systems, such as aircraft and marine ship, is that the vibration is excited by multi-field coupled loads. The more reliable and robust design of equipment involves vibration monitoring and analysis. Here, the vibration analysis is multi-physics-based interdisciplinary research, which brings new challenges to the reliability design for mechanical equipment. This session covers the aspects of experimental and computational vibration analysis methods for reliability enhancement, design and optimization of mechanical systems. These topics include, but not limited to, multi-physics-based vibration analysis models, emerging techniques for the design of mechanical systems, uncertain dynamic analysis and robust design. Presentations on theoretical and experimental work, as well as case studies, are welcome.
● Multi-physics-based vibration analysis model
● Multi-physics-based reliability models
● Uncertain dynamic analysis and robust design
● Data-driven methods for vibarion reliability analysis
● Multi-source information fusion of complex systems
Organizers:
Dr. Zhenguo Zhang, Shanghai Jiaotong University, China. E-mail: zzgjtx@sjtu.edu.cn
Dr. Xihui (Larry) Liang, University of Manitoba, Canada. Email: xihui.liang@umanitoba.ca
#16 Nonstationary Signal Processing for the Machinery Diagnostics and Prognostics
Industrial machinery undergoes inevitable health degradation, which affects its performance and structural integrity. Timely diagnosis and prognosis of the degradation symptoms are essential to support predictive maintenance decision-making and to guarantee industrial safety and productivity. Variable operating conditions are often encountered in industrial machinery. For instance, wind turbine gearboxes operate under random speed and load due to the randomness of wind speed and direction, while train traction gearboxes operate under variable speed and load when the train passes through high-curvature areas. The variable operating conditions may accelerate the degradation process of machinery and manifest in the condition monitoring data, which impede the diagnosis and prognosis of the machinery. Developing methods to effectively and efficiently process the nonstationary condition monitoring data to achieve accurate and reliable fault diagnosis and prognosis under variable operating conditions has drawn much interest in the past decade.
The intention of this special session is to present works dealing mainly (but not exclusively) with state-of-the-art solutions of signal processing, dynamics modelling, and artificial intelligence for machinery diagnostics and prognostics.
Organizers:
Prof. Yuejian Chen, Tongji University, China. E-mail: yuejianchen@tongji.edu.cn
Prof. Gang Niu, Tongji University, China. Email: gniu@tongji.edu.cn
#17 Intelligent gearbox fault diagnosis under limited samples
Gearboxes are the key components of various machines, such as wind turbines, aircraft, high-speed rail, automobile, ship, et al. Therefore, it is significant for monitoring the health status of gearboxes. In recent years, the intelligent gearbox fault diagnosis has received a lot of attention. A variety of methods, such as K nearest neighbors, support vector machine, random forest, manifold learning, deep learning, transfer learning, meta learning, et al., have been proposed to achieve the intelligent fault diagnosis. However, it is difficult to obtain the sufficient gearbox fault samples in the actual engineering, thus the intelligent fault diagnosis of gearboxes under limited samples still faces some great challenges. In order to facilitate the development of intelligent fault diagnosis approaches for the actual gearboxes, we hope to organize a special issue titled as “Intelligent gearbox fault diagnosis under limited samples” in PHM-Hangzhou conference. This special session hopes to attract these studies including data-model combined fault diagnosis, fault diagnosis under variable working condition, machine learning-based fault diagnosis, fault transfer diagnosis, domain generalization fault diagnosis, federated gearbox fault diagnosis, et al.
Topics of this session mainly include:
● data-model combined gearbox fault diagnosis
● gearbox fault diagnosis under variable working condition
● gearbox fault diagnosis under noisy labels
● intelligent gearbox fault diagnosis by machine learning
● intelligent gearbox fault diagnosis by meta learning
● intelligent gearbox fault diagnosis by zero-shot learning
● domain adaption approaches for gearbox fault transfer diagnosis
● domain generalization networks for gearbox fault diagnosis
● federated gearbox fault diagnosis methods
● Other advances in intelligent gearbox fault diagnosis
Organizers:
Yi Qin, Chongqing University, China. E-mail: qy_808@cqu.edu.cn
Xiang Li, Xi’an Jiaotong University, China. Email: lixiang@xjtu.edu.cn
Yi Wang, Chongqing University, China. Email: wyyc@cqu.edu.cn