Predicting the local solidification time using spherical neural networks

Castings are predestined for the application of structural optimization, but to date, the integration of process simulation into structural optimization is limited due to high computational cost and is therefore often neglected at the beginning of the design process. This leads to the need for surrogate models, which allow a fast and simplified evaluation of design proposals during the optimization in order to improve the integration. This article introduces a novel approach that estimates the solidification time of randomly created geometries solely based on the casting geometry. The approach uses ray-tracing methods to calculate the distance function along preset directions. The estimated solidification time is calculated using a Spherical Convolutional Neural Network (CNN). The training data is obtained by several thousand solidification simulations using the optimization toolkit of a commercial casting simulation software combined with further data augmentation. The model is experimentally validated for five different geometries in the sand casting process.     «

Castings are predestined for the application of structural optimization, but to date, the integration of process simulation into structural optimization is limited due to high computational cost and is therefore often neglected at the beginning of the design process. This leads to the need for surrogate models, which allow a fast and simplified evaluation of design proposals during the optimization in order to improve the integration. This article introduces a novel approach that estimates the s...     »