Generative Ingate Design Assistance for Superior High Pressure Die Casting

Purpose and goal:

To meet the rapid industrial drive for manufacturing both larger and more complex die cast aluminum components, the project will develop a completely new design support for inlet and gating design die casting dies. This will be achieved by using coupled physics-based machine learning and computational fluid dynamics (CFD).

The goals of the project are as follows. 1. Verified method for AI based gating design for high pressure die casting. 2. 25% shorter development time of tools for die casting. 3. 20% less flow and gas related defects. Expected results and effects:

The result of the project is a validated physics-based machine learning model that can be used to predict and optimize process parameters for the design of the injection molding system. The result will primarily be able to be utilized by the entire die casting industry, but will also be applicable within more manufacturing processes as the method is generic.

Through the application of physics-based machine learning, the project will contribute to increased efficiency, quality and sustainability, which in the long run will also lead to strengthened competitiveness. Approach and implementation: The generative design optimization method will be developed through the following main activities.

1. Development and validation of model for die casting fluid mechanical characteristics in OpenFOAM. 2. Integration of ML algorithms for injection molding system optimization. 3. Validation of the method through both virtual and physical demonstrator.

The results will be disseminated outside the project group both through popular science and academic journals/conferences.