Optimizing Parameters for Microcellular Foaming and In‐Mold Decoration of Talc/ <scp>PP</scp> Composites Using <scp>PSOBP</scp> Neural Network

ABSTRACT The growing demand for lightweight and high‐performance materials in the automotive industry has driven the development of microcellular foaming technology, which enables weight reduction and improved material properties. However, injection molding processes often result in warpage, limiting large‐scale applications. This study investigates the effects of varying talc contents (0%, 2%, 4%, 6%, and 8%) in polypropylene (PP) composites on foaming behavior, mechanical properties, and warpage. Results reveal that 4% talc provides the optimal balance between mechanical performance and dimensional stability. Based on these findings, a particle swarm optimization with backpropagation (PSOBP) neural network was used to optimize key parameters, including mold temperature, filling pressure, and cooling time. By analyzing 200 parameter combinations via Moldex3D simulations, the optimal process achieved a 2.1% weight reduction and a 7.2% warpage reduction. Experimental trials validated the simulation results, confirming the method's reliability and industrial feasibility. This study provides a practical framework for optimizing manufacturing efficiency, dimensional accuracy, and lightweighting of automotive components.