Limitations in Structural Identification: Case Studies

NDE/NDT for Structural Materials Technology for Highways and Bridges
August 21-24, 2012
New York, NY

Jeffrey Weidner, John Prader, Franklin Moon, and A. Emin Aktan

Structural Identification (St-Id) aims to infer un-measurable but desired responses/attributes of a structure by correlating measured responses with a simulation model. Stated in a slightly different way, St-Id attempts to correlate one or more analytical models with measured data and then uses the correlated models to both estimate current attributes and forecast future performance [1]. St-Id is a valuable tool for aiding in decision-making because when implemented properly, St-Id can estimate key metrics such as live load capacity, seismic vulnerability, etc. that cannot be directly measured or reliably estimated based on heuristics. While this approach is quite promising, there are many potential pitfalls that not only limit its effectiveness, but can also potentially result in highly misleading and unreliable results. This is an unfortunate but undeniable part of the history of St-Id, and has fueled skepticism related to the value of this paradigm.

In contrast to the chosen title, the over-arching goal of this paper is to present a successful case study where St-Id has been applied to a real, operating structure. An honest discussion of some of issues that surfaced during this effort is presented, including how these issues were either directly addressed, mitigated to some extent or circumvented entirely. First a discussion of general limitations and constraints to St-Id is presented, eliminating the need for any repeated discussion in the context of the individual case study. The case study is a “typical” steel stringer bridge, where the focus is on the relationship between response modalities and on practical complications resulting from limited traffic control.

Access the full article here