Change Leader: Research Into Dynamic Systems Will Lead to Better Structural Designs

Seth Guthrie, P.E., S.E., is the director of product management of Structural Engineering Products for Bentley Systems. Geoff McDonald is the senior product manager of Offshore Structural Engineering Products for Bentley Systems.

---

The end products of engineering typically are very easy to observe: the buildings, bridges, water systems and other structures that human life increasingly depends on. The tools to build these structures often are easy to point to: the mathematics, engineering and software to design modern structures and then the construction equipment to build them. But the foundations of these omnipresent structures and tools are laid in research and testing that happens behind the scenes and gets little credit outside of academia. Such research is fundamental to improve the safety, sustainability and affordability of infrastructure, and it should receive more of the credit.

Shakedown Competition

A fascinating 2023 competition designed to predict the dynamic shake-table response of a three-story, 40-ton reinforced concrete U-shaped wall will hopefully lead to a better understanding of non-linear dynamic behavior, but it certainly led to a very informative interview with Geoff McDonald and Seth Guthrie. I highly recommend listening to the full 45-minute discussion, as the space here only allows for a brief summary of a complicated topic.

As far as the competition details, an “Engineered Solutions” in the October 2024 issue of Informed Infrastructure, “ADINA Wins International Blind Prediction Contest,” is an excellent summary (bit.ly/4fWwNzR). The rest of this profile will focus on my follow-up interview and what I learned about research related to performance-based analysis and design.

Beyond Finite Element Analysis

McDonald and Guthrie primarily work with structural engineering products and projects, particularly for offshore structures, which are complicated infrastructure built in some of the most-difficult and remote conditions in the world. Building them requires understanding complicated mathematics and modeling related to nonlinear (dynamic) systems. The basis for dynamic modeling is finite element analysis (FEA), which has improved greatly due to modern software and computing power.

“The problem you have to solve is to invert a really, really big matrix, which is hard for humans, but easy for computers,” Guthrie explains. “Now problems that seem unimaginably complex can be solved with a computer, and then the engineer can look at that and understand in a macro way why it’s behaving that way. Why is that thing twisting so much? Why is that thing cracking in this location?”

McDonald rounds out the definition of FEA: “They have an element of a finite length or space that represents some sort of physical property—stiffness, thermal conductivity, fluid flow. It’s the mathematical representation of that system broken down into these small, relatively easy-to-solve bits.”

FEA is the basis for many engineering codes and standards, but going beyond it into performance-based analysis and design is the next step in sustainable design.

Performance-Based Analysis

“We tend to treat steel and concrete material as if it is linear,” notes Guthrie. “It deflects a certain amount under a certain amount of load. It has a constant stress/strain curve until it gets to yield. But once it goes beyond yield, it starts to become plastic, and then the deformation is much greater with just a little bit more stress.”

Unfortunately, to get the most-accurate real-world models of actual structures, engineers need to move beyond linear analysis, especially as “worst-case scenarios” such as 100-year hurricanes, flooding and fire events become more regular. McDonald describes this as “pushover analysis.”

“We take an extreme-level event, and then we increase that load on a structure,” he says. “We model the nonlinear behavior of the structure and take it until it collapses, and we get something called a ‘reserve strength ratio,’ which gives us the ratio of the structure performance against that design code check.”

Guthrie adds that all these improvements and innovations stem from research: “Pretty much everything in the code started as a research exercise for somebody,” he notes.

Avoid Conservative Engineering Pitfalls

“A lot of people in the structural engineering world think of conservatism when it comes to linear systems, and that can be a pitfall for people when they start to think about dynamics,” cautions McDonald. “We’ve seen real-world failures because the dynamic behavior of the structure was wrong, because they were making conservative assumptions about the stiffness or the nonlinear behavior of their system.

“Using nonlinear dynamic analyses can avoid those pitfalls of falling into that conservative mindset,” he adds. “If you see a dynamic nonlinear system, analyze it as one—don’t rely upon a simplification for a linear system. This is where we can really contribute in the space to improving, making it more accessible to people to be able to do those types of analyses and not rely upon those simplifications.