AI Material Visualization Method Simplifies Design Process

Facebook X LinkedIn Email
CAMBRIDGE, Mass., May 4, 2021 — MIT researchers have developed a technique to quickly determine certain properties of a material, such as stress and strain, based on an image that shows the material’s internal structure. The method addresses the bottleneck that requires engineers to rely on physical laws to understand material qualities, which involve equations that can be arduous and highly complex.

The MIT approach instead uses computer vision and machine learning to deliver insights on those characteristics, and it generates estimates in real time.

Specifically, the researchers used a technique called a generative adversarial neural network. They trained their network on thousands of paired images — one that depicted a material’s internal microstructure subject to mechanical forces, and another that depicted that material’s color-coded stress and strain values. A generative adversarial neural network then took those examples and used principles of game theory to iteratively figure out the distinct relationships between the geometry of a material and its resulting stresses.

“From a picture, the computer is able to predict all those forces: the deformations, the stresses, and so forth,” said Markus Buehler, the McAfee Professor of Engineering, the director of the Laboratory for Atomistic and Molecular Mechanics, and a co-author of a paper describing the research. “That is really the breakthrough — in the conventional way, you would need to code the equations and ask the computer to solve partial differential equations. We just go picture to picture.”

That image-based approach is especially advantageous for complex, composite materials, the researchers said; forces on a material may operate differently at the atomic scale than at the macroscopic scale.

“If you look at an airplane, you might have glue, a metal, and a polymer in between. So, you have all these different faces and different scales that determine the solution,” Buehler said. “If you go the hard way — the Newton way — you have to walk a huge detour to get to the answer.”

Newton is credited with the development of many of the physical laws and equations on which engineers have relied to determine the internal forces of materials.

“Many generations of mathematicians and engineers have written down these equations and then figured out how to solve them on computers,” Buehler said. “But it is still a tough problem. It is very expensive — it can take days, weeks, or even months to run some simulations. So, we thought: Let’s teach an AI to do this problem for you.”

The newly developed network additionally works well at multiple scales, as it uses a series of “convolutions” to process information. It is these convolutions themselves that serve to analyze images at progressively larger scales.

In tests, the network successfully rendered stress and strain values from a series of close-up images of the microstructure of various soft composite materials. The network was even able to capture “singularities,” like cracks developing in a material. In these instances, forces and fields change rapidly across tiny distances.

Once trained, the network runs almost instantaneously on consumer-grade computer processors. That could enable mechanics and inspectors to diagnose potential problems with machinery simply by taking a picture, the researchers reported.

The advance is also poised to reduce the number iterations needed to design products, said Suvranu De, a mechanical engineer at Rensselaer Polytechnic Institute. De was not involved in the research.

The variety of engineering applications De said the work will influence include those in the automotive and aircraft industries, and those that involve natural and engineered biomaterials.

This visualization shows the deep-learning approach in predicting physical fields given different input geometries. The left figure shows a varying geometry of the composite in which the soft material is elongating, and the right figure shows the predicted mechanical field corresponding to the geometry in the left figure. Courtesy of Zhenze Yang, Markus Buehler, et al.
The deep learning approach in predicting physical fields given different input geometries: The left figure shows a varying geometry of the composite in which the soft material is elongating, and the right figure shows the predicted mechanical field corresponding to the geometry in the left figure. Courtesy of Zhenze Yang, Markus Buehler, et al.
“It will also have significant applications in the realm of pure scientific inquiry, as force plays a critical role in a surprisingly wide range of applications from micro/nanoelectronics to the migration and differentiation of cells,” De said.

The technique could give nonexperts access to state-of-the-art materials calculations. Architects or product designers, for example, could test the viability of their ideas before they pass the project to an engineering team.

In the new paper, the researchers worked primarily with composite materials that included both soft and brittle components in a variety of random geometrical arrangements. In future work, they plan to use a wider range of material types.

Funding for the research was provided in part by the Army Research Office and the Office of Naval Research.

The research was published in Science Advances (

Published: May 2021
computer vision
Computer vision enables computers to interpret and make decisions based on visual data, such as images and videos. It involves the development of algorithms, techniques, and systems that enable machines to gain an understanding of the visual world, similar to how humans perceive and interpret visual information. Key aspects and tasks within computer vision include: Image recognition: Identifying and categorizing objects, scenes, or patterns within images. This involves training...
neural network
A computing paradigm that attempts to process information in a manner similar to that of the brain; it differs from artificial intelligence in that it relies not on pre-programming but on the acquisition and evolution of interconnections between nodes. These computational models have shown extensive usage in applications that involve pattern recognition as well as machine learning as the interconnections between nodes continue to compute updated values from previous inputs.
machine learning
Machine learning (ML) is a subset of artificial intelligence (AI) that focuses on the development of algorithms and statistical models that enable computers to improve their performance on a specific task through experience or training. Instead of being explicitly programmed to perform a task, a machine learning system learns from data and examples. The primary goal of machine learning is to develop models that can generalize patterns from data and make predictions or decisions without being...
Materialscomputer visionAIneural networksimulationscomputer simulationsatomic scale measurement of material propertiesphysicsmachine learningmachine learning techniques

We use cookies to improve user experience and analyze our website traffic as stated in our Privacy Policy. By using this website, you agree to the use of cookies unless you have disabled them.