Predictive ANN Modeling and Optimization of Injection Molding Parameters to Minimize Warpage in Polypropylene Rectangular Parts

Abstract

Injection molding is a fundamental process for transforming plastics into various industrial components. Among the critical aspects studied in this process, volumetric contraction and warpage of plastic parts are of particular importance. Achieving precise control over warpage is crucial for ensuring the production of high-quality components. This research explores optimizing injection process parameters to minimize volumetric contraction and warpage in rectangular polypropylene (PP) parts. The study employs experimental analysis, MoldFlow simulation, and Artificial Neural Network (ANN) modeling. MoldFlow simulation software provides valuable data on warpage, serving as input for the ANN model. Based on the Backpropagation Neural Network algorithm, the optimized ANN model accurately predicts warpage by considering factors such as part thickness, flow path distance, and flow path tangent. The study highlights the importance of accurately setting injection parameters to achieve optimal warpage results. The BPNN-based approach offers a faster and more efficient alternative to computer-aided engineering (CAE) processes for studying warpage.