Physics-Informed and AI-Driven Forming for Sustainable Manufacturing

Aim and Objectives

The PINNForm project aims to develop a new method for rapid, flexible, and data-driven manufacturing of metal components. The main goal is to create an AI-based, physics-informed process for metal forming that is both faster and more sustainable and flexible than traditional methods. The project seeks to strengthen the Swedish industry's transition to more sustainable processes and increase its resilience.

Project Challenge

Today's industrial forming processes are often slow, expensive, and resource intensive. They require extensive manual expertise, physical testing, and long lead times, which creates both economic and environmental costs. There is a growing need for locally adaptable and resilient manufacturing, especially in light of global challenges such as climate change, geopolitical tensions, and the need for rapid transition. The project also addresses cybersecurity by reducing reliance on global cloud services for AI model training.

Our Solution

The project combines physics-informed AI and machine learning (ML) with a unique origami-inspired technique. By integrating the laws of physics directly into the AI models, a digital twin is created that can simulate and optimize the manufacturing process with high precision. This method reduces the need for large amounts of data and extensive test series, making the process significantly faster and more cost-effective. The generative AI model, which is trained at MIMER, Sweden's AI gigafactory in Linköping, functions as intelligent support for design and manufacturing. As a tangible demonstrator, the method will be used to develop light and strong drone components (a "Frame Exoskeleton" for electric drones) for forestry and emergency preparedness.

Project Impact

The PINNForm project has great potential to create a paradigm shift in metal forming. By optimizing design and production parameters, waste material and rejects are drastically reduced, which contributes to more sustainable manufacturing and increased circularity. Rapid adaptability strengthens the industry's resilience and ability to handle disruptions. In the long term, technology can also be used to predict repair methods, which reduces the need for new production. By decreasing the reliance on manual expertise and physical strength, the project also contributes to a more gender-equal industry, in line with Agenda 2030. Technology is scalable and can be applied in many other sectors, such as the automotive, aerospace, space, defense, and energy sectors.