Draft Full Paper Due:
May 7,2017 (Extended)

Notification of Draft Paper Acceptance:
May 10,2017 (Updated)

Author Registration Deadline:
May 25,2017(Extended)

Final Manuscript Due:
May 25,2017 (Updated)

Early Bird Registration Deadline:
August 1, 2017


Home > Program


SDPC 2017 Conference Program Matrix is now downloadable by clicking the following link: Program Matrix (Updated on August 9, 2017)



Gas Turbine Component Life Prediction and Life Cycle Management

Workshop Schedule:

Section I – II: 9:00 a.m. – 12:00 p.m.; Section III – V: 2:00 p.m. – 5:00 p.m.

Workshop Program:

Section I: Introduction: Component Damage Modes and Reliability (Koul)
a. Component Damage Modes
b. Reliability Practices
c. FMECA and Effective Time to Failure (ETTF) Concept

Section II: Design Life Prediction and Modeling Techniques (Koul and Banerjee)
a. Design Practices (YS, DBTT, HCF, LCF, Creep)
b. Creep Life Prediction Techniques including a Case Study
c. LCF Life Prediction Techniques and a Case Study
d. TMF Life Prediction and a Case Study
e. HCF Life Prediction and a Case Study

Section III: Damage Tolerance and ENSIP (Koul and Banerjee)
a. Fatigue Crack Growth Rate based Damage Tolerance and a Case Study
b. Creep Crack Growth Rate based Damage Tolerance and a Case Study
c. ENSIP (MIL-HDBK-1783)

Section IV: Emerging Prognostics (Predictive Maintenance) Technologies (Banerjee)
a. Why Prognostics
b. Prognostics Process

Section V: Demonstration of XactLIFE-GT PHM for Predictive Maintenance (Banerjee)
a. Demonstration of the XactLIFE software operation
b. Test Cases


Ashok Koul, Ph.D, P.Eng., FASM
President and CEO, Life Prediction Technologies Inc., Canada

Dr. Koul served as a Chief Scientist with the National Research Council of Canada for over 20 years, leading efforts in bleeding edge research related to high temperature structural materials for turbines. He has more than 115 publications in journals and conference proceedings, two patents, and published “High Temperature Structural Materials and Protective Coatings for Gas turbines” in 1994. He was elected a fellow of the ASM International in 1994 for his contributions in the fields of superalloys and damage tolerance. He is a registered professional engineer in Ontario, Canada, a panel member of the US based IGTI and a board member of Met and MatTrans Journal.


Avisekh Banerjee, Ph.D, P.Eng.,   
Senior Manager, Life Prediction Technologies Inc., Canada

Dr. Banerjee is the Systems Development and Services Manager at Life Prediction Technologies Inc. He manages the development of diagnostics and prognostics tools for turbo-machinery and avionics at LPTi. His broad areas of research interest are performing physics based prognostics case studies, ENSIP, data trending for failure prediction, development of parts life tracking systems and the development of PHM framework. He works extensively with end users requiring prognostics services. Dr. Banerjee is a registered professional engineer in Ontario, Canada.


Life Prediction Technologies Inc. (LPTi), Canada (www.lifepredictiontech.com):

LPTi specializes in prognostics and health management (PHM) services for the life cycle management of gas turbines using its patented XactLIFE system. The system accurately predicts structural damage in components (well before actual damage develops) and detects existing structural issues through sensor-based diagnostics, non-destructive inspection or destructive metallurgical testing. LPTi has provided services to various clientele in the military and power generation sectors with significant success. LPTi’s high predictive accuracy is attributed to advanced material physics-based mathematical models that consider the effects of usage and materials processing variability into the microstructure of the component materials. Hence, XactLIFE provides quantitative predictive maintenance strategies for direct cost savings of 40% by optimizing parts replacement, refurbishment, and life extension while simultaneously maintaining safety and reliability of individual components. Cost savings of over US $40 Million were achieved for a land-based turbine fleet through deferred capital expenditures and reduced maintenance. LPTi has also significantly reduced the design life cycle time for emerging OEMs by offsetting prototype testing with simulation.
Engine owners, who maintain their fleets or seek independent opinions on OEM’s recommendations, benefit from LPTi's prognostics services.  LPTi has a strategic partnership with the NRC on material testing and has R&D collaborations with Carleton University, University of Ottawa, and ÉTS, Quebec.




Tutorial 1: Performance Prediction of Nonlinear Degrading Systems
Instructor: Professor Fai Ma, University of California at Berkeley, USA

Tutorial Abstract:
All structures exhibit nonlinear behavior and degrade when acted upon by cyclic loads associated with earthquakes, high winds, and sea waves. If the restoring force is plotted against the structural deformation, degradation manifests itself in the evolution of hysteresis loops. However, a fundamental theory of the evolution of hysteresis loops has not been developed.
In the absence of a model of hysteretic evolution or degradation, cyclic tests of structural joints and connections were routinely conducted around the world in the past thirty years. These tests have generated a substantial amount of experimental data on load-displacement hysteretic traces for wood, steel, and concrete structures. In the same period, generalization of the Bouc-Wen differential model of hysteresis permits curve-fitting of practically any hysteretic trace with a suitable choice of its thirteen control parameters. Using system identification techniques, it appears highly feasible to utilize the generalized differential model of hysteresis and the extensive database of experimental hysteretic traces to deduce a working model for degrading structures. A fundamental objective of this research project is to do just that.
Three specific tasks will be addressed in this presentation. First, a robust identification algorithm will be devised to generate models of degradation of a structure from its experimental load-displacement traces. This algorithm will be based upon the generalized differential model and the theory of genetic evolution, streamlined through sensitivity analysis. Second, it will be verified by experimentation that a model of degradation obtained by identification can be used to predict the future performance of a structure. Third, a procedure will be suggested to decompose a complex structure into a number of elementary joints and connections. Through such decomposition, the relationship between degradation of a complex structure and the degradation of its constituent joints and connections will be explored.
The significance of this project cannot be over-emphasized. Through brute-force identification of hysteretic evolution or degradation, it becomes possible to assess, for the first time in analysis, the performance of a real-life structure that has previously been damaged. There is not any other method that can predict the response of a nonlinear degrading structure well beyond its linear range.

Presenter’s Biography:
Dr. Fai Ma is Professor of Applied Mechanics in the Department of Mechanical Engineering, University of California at Berkeley. He received his B.S. degree from the University of Hong Kong in 1977 and his Ph.D. degree from the California Institute of Technology in 1981. From 1981 to 1986, he was a research engineer with Weidlinger Associates, the IBM Thomas J. Watson Research Center, and the Standard Oil Company. He is the recipient of several awards, which include a Presidential Young Investigator Award from the National Science Foundation, an Alexander von Humboldt Fellowship, and a Fulbright Senior Scholar Award. He is the author or co-author of more than 180 technical publications in the areas of vibration, system uncertainties, and stochastic simulation. He often serves as a consultant to industry and is a fellow of the American Society of Mechanical Engineers.

Tutorial 2: Intelligent Health Monitoring of Rotating Machines
Instructor: Professor Nishchal K. Verma, IIT Kanpur, India

Tutorial Abstract:
Intelligent health monitoring of machines for early recognition of faults saves industry from heavy losses occurring due to untimely machine breakdowns. The goal of intelligent health monitoring is achieved with the help of Computational Intelligence. Computational Intelligence is a set of biological and linguistic tools that provide lot of freedom in efficiently addressing complex and challenging real world problems. In this tutorial, an intelligent framework to build effective data driven models for health monitoring will be presented. The tutorial will have three parts. In the first part, the tutorial will explain how Computational Intelligence techniques can be used for performing the primary operations of data pre-processing, feature selection and classification. In the second part, Sensitive Position Analysis will be discussed. Optimizing the number of sensors and finding optimal locations for placing sensors is a major concern for many health monitoring applications, especially with respect to reliability of diagnostic outcomes and cost efficiency. This tutorial will also give a brief history as to how this problem was first tackled with statistical analysis and later through Computational Intelligence techniques. The final discussion would be on making these technologies portable. Portable diagnostics is important because it can significantly cut down labor costs and also allow diagnosis in areas which are unreachable by humans. The tutorial will also illustrate how the entire technology as described earlier, was implemented for rotating machines such as Air compressor monitoring and Drill bit monitoring using smart phones, tablets, sensors, and also discuss the challenges faced in this endeavor.

Presenter’s Biography:
Dr. Nishchal K. Verma is an Associate Professor with Department of Electrical Engineering, IIT Kanpur, India and recipient of Devendra Shukla Young Faculty Research Fellowship from IIT Kanpur for 2013-16. He is an IETE Fellow, IEEE Senior Member, and was the Founding Chairman of IEEE UP Section Computational Intelligence Society Chapter from 2013 to 2015. His research interests include Deep Learning, Computational Intelligence, Big data, Internet of Things, Intelligent Data Mining Algorithms, Diagnosis and Prognosis for Health Management, Computer Vision, and Cyber Physical Systems. He has authored/co-authored more than 160 research papers in reputed national and international journals and conferences. Dr. Verma is the Editor of IETE Technical Review, an Associate Editor of the IEEE Computational Intelligence Magazine, Transactions of the Institute of Measurement and Control, and Editorial Board Member for several reputed national and international journals and conferences.

Tutorial 3: Gearbox Fault Detection and Diagnosis
Instructor: Professor Chris Mechefske, Queen’s University, Canada

Tutorial Abstract:
Up to half of all operating costs in most industrial facilities are a result of maintenance. A range of different maintenance strategies may be applied to machines in order to assure optimum performance over the operational life at minimum overall cost. Which maintenance strategy is applied in a particular case depends on a host of factors. Typically, because gearboxes are relatively complex, expensive to repair or replace, carry high loads and are often used in situations that are critical to production, a condition based maintenance approach is employed. Condition based maintenance involves monitoring physical parameters over the course of machinery life and using these parameters to detect, diagnose and sometimes predict failure in a machine. Current techniques used for condition monitoring and fault detection applied to gearboxes are based on the analysis of vibration measurements, acoustic emission signals, and/or oil quality and wear particle assessments. These methods will be described, including their advantages and disadvantages. Particular attention will be paid to vibration based gear dynamic modeling and vibration signal analysis techniques. A review of some recently developed vibration signal analysis methods for enhanced gearbox fault detection and diagnosis will also be included.

Presenter’s Biography:
Dr. Chris Mechefske is a full Professor in the Department of Mechanical and Materials Engineering at Queen’s University in Kingston, Ontario, Canada. His research interests include vibration based machine condition monitoring and fault diagnostics, maintenance and reliability, machine dynamic analysis, and vibration and noise reduction. He is a member of the editorial board of the Journal of Condition Monitoring and Diagnostic Engineering Management; Canadian Advisory Council, ISO Technical Committee 108, Sub-Committee 5; American Society of Mechanical Engineers; Canadian Machinery Vibration Association (past president 2003-2005); the International Institute of Acoustics and Vibration (Director 2007-2009); and a Fellow of the Canadian Society of Mechanical Engineers.

Tutorial 4:  Fault Diagnosis of Induction Motor Using Machine Learning Techniques
Instructor: Professor Jaya Sil, Indian Institute of Engineering Science and Technology, India

Tutorial Abstract:
Real world data are often imprecise, inexact and redundant which limits applicability of the conventional methods in decision making or diagnosing the systems. Curse of dimensionality creates obstacle in training and run time phases of machine learning techniques applied to solve real world problems, where the exact parameters of relations are not necessarily known. During system modeling many attributes are used to ensure presence of all the necessary information without evaluating significance of the attributes. All attributes are not equally important and often redundant, therefore increasing complexity of the system. The goal of this research is to analyzing the data-sets from different perspective and summarizing it into useful information or knowledge. Large dimensional data consists of redundant as well as unique information, which are extracted by analyzing the data using machine learning algorithms. Different computational tools such as Rough-set theory, Fuzzy-set theory, Fuzzy-rough set and genetic algorithm are utilized for developing novel knowledge extraction algorithms. Machine learning is becoming a well adopted tool in the area of condition monitoring of electrical machines in the recent years. This novel approach of fault classification dominates traditional methods as it encompasses the wide range of behavioral operation and does not need any prior information about the induction motor parameters. The advancement in digital technology motivates researchers to develop an efficient memory based data driven approach for fault diagnosis. Discovering useful knowledge from motor current or vibration in the form of relevant features is a vital step to develop any fault diagnosis algorithm.

Presenter’s Biography:
Dr. Jaya Sil is attached with the Department of Computer Science and Technology in the Indian Institute of Engineering Science and Technology, Shibpur as a Professor since 2003. She passed out BE in Electronics and Tele Communication Engg from B.E. College under Calcutta University, India on 1984 and ME (Tele) from Jadavpur University, Kolkata, India on 1986. Prof. Jaya Sil obtained her Ph.D (Engg) degree from Jadavpur University, Kolkata on 1996 in the topic Artificial Intelligence. She started her teaching career from B.E. College, Howrah, India in the department of Computer Science and Technology as a lecturer on 1987. Prof. Sil worked as Postdoctoral Fellow in Nanyang Technological University, Singapore during 2002-2003. Prof. Sil visited Bioinformatics Lab in Husar, Heidelberg, Germany for collaborative research. INSA Senior scientist fellowship has been awarded to her and she visited Wraclaw University of Technology, Poland in 2012. Prof. Sil also delivered tutorial, invited talk, presenting papers and chairing sessions in different International conferences in abroad and India. Prof. Sil has published more than 50 research papers in refereed journals, more than 150 International conference papers and working in the field of Bioinformatics, Machine learning and Image Processing along with applications in different Engineering fields. She published books and several book chapters and acted as reviewers in IEEE, Elsevier and Springer Journals. Prof. Sil acts as reviewer in IEEE Transaction, Elsevier journal and Springer publications.