While many companies have robust R&D facilities trying to solve problems, higher education provides significant contributions, performing 13% of all U.S. research and development in 2017, according to the National Center for Science and Engineering Statistics, part of the National Science Foundation. In inflation-adjusted dollars, total academic R&D has grown every year since 1975, and in 2018, academic institutions performed $79.4 billion in R&D.
One of the segments benefitting from university research is composites. This year’s annual university research and development article highlights five projects in crucial areas, including composite joining, recycling and computer aided process planning.
Drilling Holes Without Damaging Fibers
Project: Continuous Fiber Fastener Holes
School: North Carolina Agricultural and Technical State University
Location: Greensboro, N.C.
Principal Investigators: Ajit Kelkar and Vishwas Jadhav
Researchers at North Carolina A&T State University have developed a technique to create fastener holes in continuous carbon fiber composites without disrupting the fibers. Drilled holes are one of the primary causes of delamination in CFRP components. Drilling cuts the continuous fibers, reducing their strength and stiffness. “We are weakening the overall strength and stiffness of the laminate when we drill the holes, and this happens due to breakage of continuous fibers,” says Vishwas Jadhav, a graduate research assistant in the Joint School of Nanoscience and Engineering.
Jadhav discussed this long-standing problem with his advisor Ajit Kelkar, a professor in the Department of Mechanical Engineering. Kelkar posed a simple question: “Can we make the holes without cutting the fiber?” Inspired, Jadhav devised a method to insert removable steel pins between continuous carbon fibers during fabrication.
“The idea is similar to when we attach a button to textile fabric,” says Jadhav. “The inserted needle rarely breaks the fiber of the cloth.” When the steel pins are taken out of the carbon fabric, there is a circular fastener hole with intact continuous fibers – something that drilling is unable to achieve.
During phase one of the project, which began in 2019, researchers fabricated 2.6 mm continuous CFRP panels using 12 layers of 0/90 plain weave carbon fabric supplied by Fibre Glast Developments Corp. Working one layer at a time, Jadhav measured and marked the hole locations, used tweezers to separate the fibers at each location and placed a ¼-inch steel pin between the fibers. The panels were vacuum infused with Hexion’s EPON™ 862 epoxy resin and EPIKURE™ Curing Agent W, then cured in an autoclave. Afterward, the pins were removed, leaving ¼-inch holes.
Aligning the holes throughout the 12 layers was tricky. “The first two or three tries, we had some misalignment,” Jadhav says. “But then we got good panels.” Open hole compression testing revealed that the method increased compressive strength 25% to 30% in the adjacent area compared to panels with drilled or water jet cut holes.
In the second phase, the team experimented with a quasi-isotropic lay-up. They used 40 layers of 190 gsm 0/-45 Chomarat non-crimp (NCF) continuous carbon fiber fabric to create the 2.6 mm panels. The team alternated 20 layers of the 0/-45 fabric with 20 layers of the same fabric rotated to 45/90. As before, Jadhav marked the hole locations, carefully separated the fibers and inserted the ¼-inch pin one layer at a time. This time, he also cut the fabric stitches before separating the fibers. It was a time-consuming process. “We had to be careful not to cut the fiber,” he recalls.
Once completed, the panels were tested for compressive strength, static tension and tension-tension fatigue. The results again showed significant improvements in mechanical properties over panels with drilled or water jet cut holes. Microscopic images confirmed that the fibers remained intact.
Jadhav says that future research may focus on creating different hole diameters and automating the process. For now, the team has applied for a patent and is gauging industry interest. Jadhav believes the new technique could benefit many industries, particularly aerospace.
“The future is composites,” he says. “If this technique is used in aerospace whenever there is a joining of two structures, it will reduce delamination problems and help to enhance the life of the products.”
Harnessing the Power of Big Data
Projects: Machine Learning Algorithms
School: University of Washington
Location: Seattle
Principal Investigators: Steve Brunton and Ashis Banerjee
The machine learning industry is expected to hit $9 billion by 2022, up from $1 billion just five years earlier, according to AI Multiple, a technology industry analyst firm. It’s no wonder that schools such as the University of Washington are investing in machine learning research.
The Boeing Advanced Research Center (BARC) in the College of Engineering at the University of Washington hosts joint research projects in which Boeing-employed affiliate instructors work side by side with faculty and students on a range of projects, including ones related to machine learning. Researchers are mining vast quantities of data to find patterns that can be used to accelerate aircraft production rates, streamline design and reduce costs.
One BARC team is analyzing big data to help standardize aircraft brackets. Modern aircraft have thousands of custom brackets. “Brackets are everywhere, from critical structural components to those used to guide conduits and other electric and airflow systems throughout the aircraft,” says Steve Brunton, professor of mechanical engineering and adjunct professor of applied mathematics and computer science at the University of Washington.
Brunton, doctoral student Emily Clark, her co-advisor Nathan Kutz and Boeing project engineer Angelie Vincent spent three years developing and deploying algorithms to sift through data from Boeing and its subcontractors and identify brackets that are similar enough to be standardized. Brunton says that defining a similarity metric to reveal how close two brackets are in design and functionality was surprisingly challenging. While the results are proprietary, Brunton says their work demonstrates that many brackets can be standardized, which will reduce manufacturing and inventory costs and streamline the design process.
Another machine learning research team developed an automatic process for detecting tow boundaries in parts fabricated with automated fiber placement (AFP), which is crucial to ensuring acceptable structural properties. Currently, most composite aircraft parts, including fuselage sections, fairings and empennage components, are visually inspected with the aid of laser projection. Manual tow boundary inspections are extremely time-consuming. For instance, it might take several days to inspect an entire composite wing skin. Ashis Banerjee, associate professor of industrial and systems engineering and mechanical engineering, says that for a part with 100,000 tow ends, even state-of-the-art semi-automated inspection requires three to six hours of additional manual inspection.
Banerjee is leading a BARC team that’s using machine learning to develop an automated tow boundary detection system that would minimize the need for extra manual inspection. Collaborators include two doctoral students at the university, Wei Guo and Ekta Samani, and Agnes Blom-Schieber, technical fellow at Boeing Commercial Airplanes – Product Development/Structures Division. For two years, the researchers gathered thousands of tow boundary inspection images, but they found it difficult to generate a sufficiently large training dataset for their machine learning model.
“While we knew that the inspection images would be challenging to analyze, we were still surprised at the extent of the challenge due to variations in illuminating conditions, material processing parameters and ply geometry,” says Banerjee. “In fact, it was virtually impossible for non-experts like us to discern the tow boundaries visually in many images. Even traditional edge detection-based image processing methods failed miserably.”
So, the team turned to domain experts at Boeing to examine each image and label the tow boundaries. It was a painstaking process that resulted in a training set of approximately 3,500 images. While the data set was smaller than the team would have liked, it was enough to demonstrate that the automated inspection method reduced post-processing manual inspection by 90%.
Boeing is considering using a scaled-up version for its production lines, and a joint Boeing-UW patent has been filed. Banerjee says, “This research has the potential to save hours of manual inspection work for every large-scale aerospace component part.”
Making New Composites from Obsolete Ones
Project: Pyrolysis-Based Recycling Technology
School: The University of Tennessee, Knoxville
Location: Knoxville, Tenn.
Principal Investigator: Ryan Ginder
