Uncertainty Estimation and Failure Prediction of Full-Scale Structures using a Multi-Scale Approach
An Engineering School Interdepartmental Seminar in the Series: "Mechanics of Materials, Solids, and Structures"
Featuring: Jeffrey T. Fong, Ph.D., P.E.
Applied and Computational Mathematics Division
National Institute of Standards & Technology (NIST)
Hosted by: Distinguished Professor Satya Atluri, Department of Mechanical and Aerospace Engineering, UC Irvine
In this talk, I will begin with a fascinating story, as told by David Hand (2008) in a little book entitled “Statistics: A Brief Insight (Sterling).”In that book, Prof. Hand explained how statistics has been transformed by computers in a span of 50 years (1939-89) from a dry Victorian discipline concerned with the manual manipulation of columns of numbers, to a highly sophisticated new technology, which he named “modern statistics,” involving the use of the most advanced of software tools.
To illustrate the sheer power, excitement, and ubiquity of this new discipline for engineers to take advantage of, I will next introduce a multi-scale approach to the solution of the three so-called fundamental problems (FB) in structural design against fatigue failure (Fong, J.T., ASTM STP 675, pp. 3-8, 1979):(FB-1) Will it (the structure) work as designed? (FB-2) How long will it last? (FB-3) Why did it fail prematurely without warning?
Using the concept of “uncertainty” and its quantification, we show how engineer’s design of critical structures against fatigue failure and the estimation of their reliability need to engage modern statistics at six levels of scale: (L-1) Micro. (L-2) Specimen. (L-3) Component. (L-4) Assembly such as a structural panel. (L-5) Subsystem such as a single floor of a high-rise building. (L-6) A complete System such as a full-scale structure.
To illustrate the multi-scale modeling concept, I will introduce 4 sample applications using a public-domain statistical analysis software named DATAPLOT. The four sample problems are: (S-1) Design of aircraft windows and uncertainty estimation of time-to-failure for a specific window geometry and material. (S-2) Uncertainty estimation of fracture toughness of ASTM A533 Grade B-1 steel plate using an 8-factor, 17-run fractional factorial orthogonal design of experiments. (S-3) Uncertainty estimation of weld flaw length measured by ultrasonic testing using a 5-factor, 9-run fractional factorial orthogonal design of experiments. (S-4) Uncertainty estimation of microstructural parameters of U.S. currency paper for fatigue prediction. Significance of the multi-scale approach to high-consequence engineering decision making is discussed.
Biographical Sketch of Speaker:
Since 1966, Dr. Fong has been a Physicist and Project Manager at the Applied and Computational Mathematics Division of the U.S. National Institute of Standards and Technology (NIST) in Gaithersburg, Maryland. He conducts research and provides consulting services to engineers and scientists in both government and industry on computational and statistical modeling, fatigue, fracture, and failure analysis of complex high-consequence engineering systems, and risk-informed structural health monitoring for disaster warning and prevention. A registered professional engineer (P.E.) in the State of New York, Dr. Fong was a design engineer on hydro-, fossil-, and nuclear-power plants for eight years prior to joining NIST. Dr. Fong is a Fellow of ASTM International, Fellow of ASME International, and the recipient of an ASME Pressure Vessel and Piping Medal (1993). A graduate of the University of Hong Kong (B.Sc., eng.) and Columbia University (M.S., eng. mechanics), Dr. Fong earned his Ph.D. in applied mathematics and mechanics at Stanford University in 1966. In 2006, Dr. Fong was appointed Adjunct Professor of Statistics and Structures at the Mechanical Engineering and Mechanics Department of Drexel University, Philadelphia, PA, where he has taught a graduate-level course on “Finite Element Method and Uncertainty Analysis.” Since Jan. 2010, Dr. Fong has taught an online course at Stanford University on “Uncertainty Estimation for Structural Engineering Reliability.”