Professors Awarded NASA Grants to Support Their Research on Aerospace Composite Materials
The Aurora D8 airliner, shown here in this Aurora Flight Sciences illustration, is designed to provide greater fuel efficiency and less noise than conventional planes.
11/18/2021
By Madeline Bodin
Composite materials, typically a combination of a fiber and a moldable matrix — often a polymer — are light and strong. However, as advanced as today’s composite materials are, the space agency’s plans go beyond even the most cutting-edge composite materials currently available.
Faculty from the Francis College of Engineering are using integrated computational materials engineering (ICME) to clear the runway for innovative materials to be used in the spacecraft of the future.
Marianna Maiaru, an assistant professor of mechanical engineering, was recently awarded a three-year, $750,000 grant by NASA to apply the concepts of ICME to the manufacture and design of a vital component of a futuristic airplane, the Aurora Flight Sciences D8 “double bubble.”
Scott Stapleton, an associate professor of mechanical engineering working with Asst. Prof. Farhad Pourkamali Anaraki of the Department of Computer Science, was awarded a three-year, $518,000 grant by NASA to create a machine-learning assisted model of the stresses on composite materials to help facilitate their development and manufacturing.
Working Across Composite Length Scales
Preliminary work done by Maiaru and her research team was presented at the 2022 AIAA SciTech Forum, the world’s largest event for aerospace research and development, and was awarded the prestigious ICME Prize.
“ICME is a new branch of mechanics that should advance the state of theoretical modeling at different scales,” she says. “It has been used with metals over the years, but it has been less used for composites.”
Right now, Maiaru says, developing a new engineering application requires lengthy and expensive testing. The complexity of composite materials makes this testing process particularly difficult. By testing the validity of mathematical models, ICME will allow a shift from time-consuming, real-world tests to speedy virtual modeling.
“This will revolutionize the way we design materials,” she says.
Maiaru is the principal investigator for the NASA grant, leading a team of researchers that includes NASA; Michigan Technological University, which will focus on the project’s quantum mechanics for nanoscale modeling; Collier Aerospace; Aurora Flight Sciences, the aerospace design firm that developed the Aurora D8 that is the focus of the research; and Maiaru’s own research group at UMass Lowell, which will concentrate on multiscale process modeling. They will use HyperX, an aerospace software from Collier Aerospace for structural optimization.
While the applications for the outcome of Maiaru’s research are broad, the funded project will focus on the Y joint in the Aurora D8 concept airplane. This vehicle has an innovative double-width fuselage design that will allow it to fly with smaller and lighter engines, reducing noise and fuel use. Its unusual fuselage shape earned it the nickname “double bubble.”
Maiaru’s project will model the performance of the plane’s Y joint from the molecular scale through the macro, whole-component scale. The Y joint is a key component of the aircraft, made from a composite material and holding together the two cylindrical sides of the fuselage.
Maiaru brings a lifelong interest in aerospace to her research.
“Aerospace has always been my passion,” she says. “From the time I was a kid, I knew I wanted to be an engineer, and flight mechanics fascinated me.”
Maiaru introduced ICME in her Mechanics of Composites class for the first time in spring 2021.
“My graduate students worked in groups to replicate the ICME project deliverables,” says Maiaru.
She says as a result of their excellent work, some students were invited to present in a seminar at NASA’s Glenn Research Center in Cleveland, Ohio, in December 2021.
A Versatile Model
Stapleton and Pourkamali Anaraki’s project focuses on analyzing composite materials, from how they handle stress to what their breaking points are.
For more than a decade, engineers at NASA research centers, such as the Marshall Space Flight Center in Huntsville, Alabama, have been testing crew modules and other structures fabricated from composite materials to see how well they could withstand the extreme conditions of a simulated space environment.
The physics of composite materials is complex.
“When a part breaks, all the action happens at a very small scale at first,” Stapleton says. Cracks start at the scale of fibers, which are typically about six microns in diameter, or about one-tenth of the diameter of a human hair. Computer models of these materials that can show cracks forming at the level of single fibers already exist, but they take massive amounts of computing power and may take weeks to run.
Part of the solution — the part that Stapleton’s and Pourkamali Anaraki’s research addresses — is to create machine learning algorithms that can be trained to replace those small-scale models, the ones that model the behavior of a few fibers within a matrix. These algorithms will be integrated into larger-scale models that work at the scale of components and vehicles.
This research for NASA is part of a career-long focus for Stapleton on the behavior of fibers within composites, particularly how fibers tangle and how those tangles influence the properties of the composite.
“Tangling is such a random process,” Stapleton says. “You have to model thousands of tangles to have a conclusive statement about them.”
It is a form of research that requires vast amounts of computing power. And that, he says, “led to the question: Can I build really fast models?”
Stapleton says the machine learning algorithm that he and Pourkamali Anaraki are developing will not just feed information to larger-scale models, but also receive and use information from the larger component- and vehicle-scale models. It will not only predict the behavior of composites but will also allow other researchers to design better composites based on the results of the modeling.
The goal is to create algorithms and models that not only work for today’s composites but will also be equally effective at modeling new composites that are yet to be invented.
“NASA is looking toward the future,” says Stapleton, “to a materials future we haven’t even thought of yet.”