Machine Condition Diagnosis using Machine Learning: A Journey

Professor Asoke K. Nandi

Brunel University London

Abstract

In earlier times machine condition diagnosis was performed solely by an expert and manually. There were no machine learning algorithms or frameworks available and neither was there any automated process for this purpose. In recent years researchers have come up with plethora of machine learning algorithms or frameworks, some of which are very effective. This transition did not happen overnight or in a short time; it has been, and continues to be, a journey.

This lecture will attempt to delineate the path of journey from a personal perspective, identifying some of the developments of key ideas. It will also highlight current concerns in the wider society about the use of machine learning, which our community need to heed and pay attention to.

Biographical Sketch

received the PhD degree in Physics from the University of Cambridge (Trinity College). He held academic positions in several universities, including Oxford, Imperial College London, Strathclyde, and Liverpool as well as Finland Distinguished Professorship. In 2013 he moved to Brunel University London.

In 1983 Professor Nandi co-discovered the three fundamental particles known as W⁺, W⁻ and Z⁰, providing the evidence for the unification of the electromagnetic and weak forces, for which the Nobel Committee for Physics in 1984 awarded the prize to two of his team leaders for their decisive contributions. His current research interests lie in signal processing and machine learning, with applications to functional magnetic resonance data, gene expression data, communications, and biomedical data. He made fundamental theoretical and algorithmic contributions to many aspects of signal processing and machine learning. He has much expertise in “Big Data”. Professor Nandi has authored over 600 technical publications, including 280 journal papers as well as six books, entitled Image Segmentation: Principles, Techniques, and Applications (Wiley, 2022), Condition Monitoring with Vibration Signals: Compressive Sampling and Learning Algorithms for Rotating Machines (Wiley, 2020), Automatic Modulation Classification: Principles, Algorithms and Applications (Wiley, 2015), Integrative Cluster Analysis in Bioinformatics (Wiley, 2015), Blind Estimation Using Higher-Order Statistics (Springer, 1999), and Automatic Modulation Recognition of Communications Signals (Springer, 1996). The H-index of his publications is 85 (Google Scholar) and ERDOS number is 2.

Professor Nandi is a Fellow of the Royal Academy of Engineering and a Fellow of six other institutions including the IEEE. He received many awards, including the IEEE Heinrich Hertz Award in 2012, the Glory of Bengal Award for his outstanding achievements in scientific research in 2010, the Water Arbitration Prize of the Institution of Mechanical Engineers in 1999, and the Mountbatten Premium of the Institution of Electrical Engineers in 1998. Professor Nandi is an IEEE Distinguished Lecturer (EMBS, 2018-2019).

Keynote Speech 02——Robert X. Gao, Ph.D.

Intelligent Maintenance through Big Data Analytics 

Case Western Reserve University

Abstract

The exponential growth of data as a result of advancement in sensing technologies and computational infrastructure has transformed the state of operation and maintenance of manufacturing machines into a data-driven paradigm that augments empirical knowledge and model-based analysis for machine condition monitoring, fault diagnosis, and remaining service life prognosis. The outcome is improved operational reliability, higher material and energy efficiencies, and reduced waste for sustainable manufacturing.

This keynote reviews the integration of physical science with data science to enable more effective and efficient data collection and visualization across supply chains, predictive maintenance, digital performance management, and intelligent process planning and agile operations.

Biographical Sketch

Robert Gao is the Cady Staley Professor of Engineering and Department Chair of Mechanical and Aerospace Engineering at Case Western Reserve University in Cleveland, Ohio.  Since receiving his Ph.D. from the Technical University of Berlin, Germany in 1991, he has been working on physics-based signal transduction mechanisms for process-embedded sensing, multi-resolution analysis, stochastic modeling, and AI/machine learning based data analytics for improving the observability of cyber-physical systems such as manufacturing machines, with the goal to improve process and product quality control.  

 Professor Gao is a Fellow of the ASME, SME, IEEE, CIRP, and a Distinguished Fellow of the International Institute of Acoustics and Vibration (IIAV). He has published over 400 technical papers, including 200 journal articles, three books, and holds 13 patents. He has received several professional awards, including the ASME Milton C. Shaw Manufacturing Research Medal (2023), ASME Blackall Machine Tool and Gage Award (2018), SME Eli Whitney Productivity Award (2019), IEEE Instrumentation and Measurement Society Technical Award (2013), IEEE Best Application in Instrumental and Measurement Award (2019), Hideo Hanafusa Outstanding Investigator Award (2018), and several Best Paper awards. Prof. Gao is the Chair of the Scientific Committee of the North American Manufacturing Research Institute (NAMRI/SME) and Chair of the Collaborative Working Group on AI in Manufacturing (CWG-AI) of CIRP. He has served as an Associate Editor for several journals, and is currently a Senior Editor for the IEEE/ASME Transactions on Mechatronics.

Keynote Speech 03——Enrico Zio

Knowledge, Information and Data (KID)-Informed Operation and Maintenance (IOM)

Enrico Zio

PSL University, France

Abstract

The advancement of monitoring capabilities and other technologies have enabled the abundant collection of knowledge, information and data (KID) on equipment operation and maintenance. Such KID can inform the assessment and prediction of equipment state, whose outcomes can be used for intelligent operation and maintenance (IOM).

In this keynote lecture, we look at the grown ability of elaborating equipment KID by machine learning algorithms to mine out information relevant to the assessment and prediction of the equipment functional state, and reflect on how to use these capabilities for IOM.

Biographical Sketch 

Enrico Zio received the MSc degree in nuclear engineering from Politecnico di Milano in 1991 and in mechanical engineering from UCLA in 1995, and the Ph.D. degree in nuclear engineering from Politecnico di Milano and in probabilistic risk assessment at MIT in 1996 and 1998, respectively. He is currently full professor at the Centre for research on Risk and Crises (CRC) of Ecole de Mines, Paris, PSL University, France, full professor and President of the Alumni Association at Politecnico di Milano, Italy, distinguished guest professor at Tsinghua University, Beijing, China, adjunct professor at City University of Hong Kong, Beihang University and Wuhan University, China and Co-Director of the Center for REliability and Safety of Critical Infrastructures (CRESCI) and the sino-french laboratory of Risk Science and Engineering (RISE), at Beihang University, Beijing, China.

He is IEEE and Sigma Xi Distinguished Lecturer.

In 2020, he has been awarded the prestigious Humboldt Research Award from the Alexander von Humboldt Foundation in Germany (https://www.humboldt-foundation.de/web/home.html), one the world's most prestigious research awards across all scientific disciplines. The Award is given to outstandingly qualified researchers and future leaders from science-related fields (but very seldom awarded to engineers!). The Award is granted in recognition of a researcher's entire achievements to date, to academics whose fundamental discoveries, new theories, or insights have had a significant impact on their own discipline and who are expected to continue producing cutting-edge achievements in the future. Professor Zio has been selected for the Award in light of being a World leading scientist in Risk and Resilience Assessment, Safety Analysis and Reliability Engineering of complex systems and infrastructures, in particular for energy applications. He has been one of the pioneers in using artificial intelligence (such as neural networks) and genetic algorithms in reliability engineering and risk assessment, solving key problems related to the safety and reliability of critical systems such as those used in the nuclear, oil and gas, transportation industries. He has promoted the use of computational modeling within various international initiatives.

His Google Scholar H-index is 93 and he is in the top 2% of the World scientists, according to Stanford ranking.

In 2021, he has been appointed as:

• • Member of the Board Committee of the International Joint Research Center for Resilient Infrastructure (ICRI)

• • 4TU.Resilience Ambassador by the 4TU Centre for Resilience Engineering and its backbone – the 4TU- programme DeSIRE (Designing Systems for informed Resilience Engineering), a strategic capacity building research programme of the four Dutch Technical Universities
• • Fellow of the of the Prognostics & Health Management Society a world recognized scientist in the area of reliability centered, condition based and predictive maintenance.

In 2023, he has been appointed as Scientific Director of Research and Development of Datrix AI Solutions group.

His research focuses on the modeling of the failure-repair-maintenance behavior of components and complex systems, for the analysis of their reliability, maintainability, prognostics, safety, vulnerability, resilience and security characteristics, and on the development and use of Monte Carlo simulation methods, artificial intelligence techniques and optimization heuristics. He is author and co-author of seven books and more than 500 papers on international journals, Chairman and Co-Chairman of several international Conferences, associate editor of several international journals and referee of more than 20.