Three-dimensional finite element-based data-driven thermal model of in-situ automated fiber placement

Abstract

This study proposes a data-driven model for thermal history prediction during in-situ consolidation of thermoplastic composites using Hot Gas Torch (HGT)-assisted Automated Fiber Placement (AFP). Temperature data was generated using a three-dimensional finite element (FE) model that was fitted to and validated with experimentally measured temperatures. Temperature curves for various combinations of heat source velocity and locations through the thickness and width were extracted from the fitted FE model. Three feedforward neural networks were trained to predict the temperature distribution during processing. The work demonstrates the effectiveness and potential of using neural networks for online prediction and optimization.