Tutorials

One of the unique features of the PHM conferences is free technical tutorials on various topics in health management taught by industry experts. At PHM15, tutorials will take place on Monday, October 19, and Tuesday, October 20, 2015. As educational events tutorials provide a comprehensive introduction to the state-of-the-art in the tutorial’s topic. Proposed tutorials address the interests of a varied audience: beginners, developers, designers, researchers, practitioners, and decision makers who wish to learn a given aspect of prognostic health management. Tutorials will focus both on theoretical aspects as well as industrial applications of prognostics. These tutorials reach a good balance between the topic coverage and its relevance to the community.

Tutorial Topics

  • Dynamic Model-Based PHM Design and Model-Based Diagnostics
    Presented by: Matt J. Smith, Manager of Electro-Mechanical Systems, Impact Technologies
    (Sikorsky Aircraft Corporation)
  • Prognostics
    Presented by: Kai Goebel, Tech Area Lead, Discovery and Systems Health, NASA
    Ames Research Center
    Presented by: George Vachtsevanos, Ph.D. is a Professor Emeritus with the School of Electrical and Computer Engineering at The Georgia Institute of Technology, and the director of the Intelligent Control Systems Laboratory
  • Feature Engineering for PHM applications
    Presented by: Weizhong Yan, Principal Scientist, Machine Learning Lab, GE Global Research
    Center
  • Cost/Benefit Tradeoffs for the Inclusion of Prognostics and Health Management (PHM) in Systems
    Presented by: Peter Sandborn, Professor, University of Maryland and member, Center for
    Advanced Life Cycle Engineering (CALCE)
Key Conference Dates
Tutorials 19 – 20 Oct 2015

PHM Conference tutorials have been a popular event in the past and the PHM society is proud to continue this service to the community. Topics of interest of these tutorials span fundamentals of PHM (Diagnostics, Prognostics, Health Management, Uncertainty Management, etc.) as well as specialized topics such as Cost-Benefit analysis, Data-Mining, Electronics PHM, Bayesian Filtering for Prognosis, etc. For a more comprehensive list of past tutorials please look at the following links:

Past PHM Tutorials
[PHM 2010] [PHM 2011] [PHM 2012] [PHME 2012] [PHM 2013] [PHME 2014] [PHM 2014]

Tutorials Chairs:
George Vachtsevanos george.vachtsevanos@ece.gatech.edu
Kai Goebel kai.goebel@nasa.gov

Tutorial Details

Tutorial Title: Dynamic Model-Based PHM Design and Model-Based Diagnostics

Matt J. Smith, Manager of Electro-Mechanical Systems, Impact Technologies (Sikorsky Aircraft Corporation)
Abstract: This session is focused on the use of dynamic, physics-based models in the development of PHM technologies. Mathematical, dynamic modeling of engineering systems is a common tool used in modern product design. As available computing resources and analytical tools have increased in capability, the fidelity and sophistication of the resulting models have likewise increased. These models can also be leveraged as a powerful tool to support PHM system design, implementation, and testing. This lecture will present methods that utilize system models as a virtual test bed for PHM technology development. Specific topics include: sensor suite optimization, fault effect propagation characterization, virtual sensor development, and prototype PHM system evaluation. In addition, the topic of model-based diagnostics will be introduced. In this approach, a model of sufficiently high accuracy can be compared to actual system response and the resulting condition information can be implemented directly as a part of the PHM system. Throughout the session, multiple case studies will be presented for applications including: hydraulic pumps, flight controls, and other critical aircraft systems.
Presenter Bio: Matt J. Smith is the Manager of Electro-Mechanical
Systems for Impact Technologies (Sikorsky Aircraft
Corporation). He has worked for Impact since 2005 and has been the technical lead on numerous programs that focused on the development of diagnostic and prognostic capability for a range of engineering systems with a focus on high power drive train components and aircraft hydraulic\electro-mechanical actuators. His primary technical interests include: electro-mechanical system diagnostics and prognostics, vibration based health monitoring, component life usage tracking, fault classifier development and health monitoring for aircraft actuator systems. He received his B.S. and M.S. degrees in Mechanical Engineering from The Pennsylvania State University.

 

Tutorial Title: Prognostics

Kai Goebel, Tech Area Lead, Discovery and Systems Health, NASA Ames Research Center

George Vachtsevanos, Ph.D. is a Professor Emeritus with the School of Electrical and Computer Engineering at The Georgia Institute of Technology, and the director of the Intelligent Control Systems Laboratory

Abstract: This Tutorial will focus on the concepts and basics of prognostics from condition-based systems health management viewpoint. Participants will be introduced to a prognostic framework that will contrast the differences with other techniques and philosophies of prognostics used in other domains. Examples will be used to illustrate various types of prediction scenarios and what does it take to set up a desired prognostic system. This will include discussions on significance of run-to-failure data, requirements and specifications generation for prognostics, prediction algorithms, post prognostic reasoning, etc. The session, then, will go into the details of setting up a prognostics problem and algorithm development using data-driven and model based approaches including data preprocessing and feature extraction steps. Discussion on prognostic performance evaluation and performance metrics will conclude the technical discussion followed by a general discussion on open research problems and challenges in prognostics. Presenter Bio: Kai Goebel works at NASA Ames Research Center where he is the Area Lead for Discovery and Systems Health and the director of the Prognostics Center of Excellence where he helped to pioneer the science of prognostics Kai received a Ph.D. from the University of California at Berkeley in 1996. He worked at General Electric’s Corporate Research Center in upstate New York where he was also an adjunct professor at Rensselaer Polytechnic Institute. He has been on the dissertation committee of seven Ph.D. students. He is currently guest professor at University of Cincinnati, holds eighteen patents, and has published more than 300 technical papers. Presenter Bio: George Vachtsevanos, Ph.D. is a Professor Emeritus with the School of Electrical and Computer Engineering at The Georgia Institute of Technology, and the director of the Intelligent Control Systems Laboratory where faculty and students are conducting interdisciplinary research in intelligent control, hierarchical/ intelligent control of unmanned aerial vehicles, fault diagnosis and prognosis of complex dynamical systems, vision-based inspection and control of industrial processes and the application of novel signal and imaging methods to neurotechnology related research. Dr. Vachtsevanos has published over 250 technical papers in his area of expertise and serves as a consultant to government agencies and industry.

 

Tutorial Title: Feature Engineering

Dr. Weizhong Yan, Principal Scientist, Machine Learning Lab, GE Global Research
Center
Abstract: Feature engineering is a process of identifying or constructing analytic features or signatures from raw sensor measurements such that the signatures maximize prediction power. Feature engineering is one of the most critical components in model building to achieve high-performance analytic solutions. Historically, feature engineering is performed manually based on domain and engineering knowledge (knowledge-driven), thus it is very problem specific. This caused feature engineering to be a labor-intense effort, taking up to 80% of overall effort in the development cycle of analytics solutions. Recently, as Deep Learning becomes the state-of-the-art machine learning technology, feature learning or representation learning, which directly learns features from data via data-driven deep hierarchical learning. Deep Learning has attracted tremendous research interests in the domains of computer vision, speech recognition, and text mining. In this tutorial, we will give a comprehensive overview of different feature engineering methods, including both traditional knowledge-driven and deep learning based, data-driven methods. More importantly, we will provide indepth details and real-world examples on how these different methods, especially deep learning based feature learning, can be used in developing PHM solutions.
Presenter Bio: Dr. Weizhong Yan, Ph.D., PE, has been with the General Electric Company since 1998. Currently he is a Principal Scientist in the Machine Learning Lab of GE Global Research Center, Niskayuna, NY. His research interests include neural networks (shallow and deep), big data analytics, feature engineering & feature learning, ensemble learning, and time series forecasting. He specializes in applying advanced data-driven analytic techniques to anomaly detection, diagnostics, and prognostics & health management of industrial assets such as jet engines, gas turbines, and oil & gas equipment. He has authored over 70 publications in referred journals and conference proceedings and has filed over 30 US patents. He is an Editor of International Journal of Artificial Intelligence and an Editorial Board member of International Journal of Prognostics and Health Management. He is a Senior Member of IEEE.

 

Tutorial Title: Cost/Benefit Tradeoffs for the Inclusion of Prognostics and Health Management (PHM) in Systems

Peter Sandborn, Professor, University of Maryland and member, Center for
Advanced Life Cycle Engineering (CALCE)
Abstract: Prognostics and Health Management (PHM) provides an opportunity for lowering sustainment costs, improving maintenance decision-making, and providing product usage feedback into the product design and validation process. However, support for PHM is predicated on the articulation of clear business cases that quantify the expected cost and benefits of its implementation. The utility of PHM to inform decision-makers within tight scheduling constraints and under different operational profiles likewise affects the cost avoidance that can be realized. This tutorial will address the life-cycle costs associated with the implementation of PHM into systems, the cost avoidance opportunities due to the incorporation of PHM, the calculation of return on investment (ROI) for the inclusion PHM into systems, the value of PHM in outcome-based contracts, and the use of cost analysis results in making maintenance planning decisions.
Presenter Bio: Karl Reichard has over 25 years of experience in the design and development of advanced measurement, control and monitoring systems. He received the Ph.D., M.S. and B.S. degrees in Electrical Engineering from the Virginia Polytechnic Institute and State University (Virginia Tech). Dr. Reichard is a Research Associate with the Pennsylvania State University Applied Research Laboratory, and an Assistant Professor of Acoustics with the Penn State Graduate Program in Acoustics. His research experience includes the development of embedded and distributed sensing and control systems for machinery and system health monitoring, acoustic surveillance and detection, active noise and vibration control and electro-optics. Dr. Reichard is a member of the Board of Directors of the Prognostics and Health Management Society, and a member of the IEEE and the Acoustical Society of America. He is the author of over 50 papers and articles published in journals and conference proceedings.