AI model shortens the development time of new materials

Time-consuming testing and computer simulations are bottlenecks in the design of new materials. A thesis from the University of Gothenburg aims to develop an AI model that can efficiently determine the durability and strength of woven composite materials.

Related posts

Study of Buddhist Monks Finds Meditation Alters Brain Activity

February 11, 2026

ICE Is Crashing the US Court System in Minnesota

February 11, 2026

Whether it is a floorball stick or a wind turbine blade to be constructed—often different composite materials are used. Composite means mixing several different materials, e.g., carbon fiber and polymers, to achieve the desired balance between different properties such as weight, durability and flexibility of the product.

However, designing new high-quality composite materials takes a long time. Developers traditionally rely on physical tests and detailed computer simulations, adjusting the design after each (failed) attempt.

Large computational resources

“This is particularly difficult when the composite is created as a woven textile fiber material, where the fibers are wrapped around each other and behave differently depending on the forces the material is subjected to,” says Ehsan Ghane, a Ph.D. student at the Department of Physics at the University of Gothenburg.

Mixing materials in a composite fabric is a challenge. Researchers may have a good understanding of the strength and other properties of individual materials, but what happens when they are mixed in a fabric composite is harder to predict.

Computers can already simulate realistic microstructures based on the interaction and influence of the materials involved at several different scales, from microstructure to macrostructure. The simulations of woven composite materials still require large computational resources.

“Neural networks, i.e. a particular family of AI algorithms, exist as an alternative to the extensive computations. However, these networks need large amounts of training data and have difficulty extrapolating results, says Ehsan Ghane. I have developed a generalized AI model that does not require as much data.”

Integrate material laws

Ehsan Ghane’s model for developing sustainable composite materials has been published and can be used now. By feeding in existing data, both from simulations and tests for the constituent materials in the composite, the model is able to predict the durability of the new composite material.

“In addition, I have investigated methods to directly integrate material laws into the AI model. This allows extrapolations outside the input data on which the model was trained. It also makes it easier to understand the order in which a material deforms, which can be important if you want to predict the material behavior in the long term.”