AI Based Injection Molding Process for Consistent Product Quality

• Published in: Procedia manufacturing Vol. 28; pp. 102 - 106
• Main Authors: Park, Hong Seok, Phuong, Dang Xuan, Kumar, Saurabh
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 2019
• Subjects: Data Driven; Diagnosis; Injection Molding; Machine learning; Smart manufacturing; Data Driven; Diagnosis; Machine learning; Smart manufacturing; Injection Molding
• ISSN: 2351-9789; 2351-9789
• DOI: 10.1016/j.promfg.2018.12.017

In manufacturing processes, Injection Molding is widely used for producing plastic components with large lot size. So, continuous improvements in product quality consistency is crucial to maintaining a competitive edge in the injection molding industry. Various optimization techniques like ANN, GA, Iterative method, and simulation based are being used for optimization of Injection Molding process and obtaining optimal processing conditions. But still due to variation during molding cycles, quality failure occurs. As many constituents like process, Material, machine together yields product quality. This paper is focused on Real time AI based control of process parameters in injection molding cycle. Process parameters and their interrelationship with quality failure has been studied and later supposed to be used to generate algorithm for compensating the deviation of process parameters. Pressure and temperature sensor assisted monitoring system is used to collect data in real time and based on its comparison with the standard values an interrelationship is formed between parameters and plastic material properties. Algorithm generates new process parameter values to compensate the deviation and machine control follows the same. The entire process is supposed to be smart and automatic after being trained with AI and machine learning techniques. Simulation using Moldflow software and real industry collected data has been used for understanding whole molding process establishing relationship between failure and parameters. An automotive product in real industry is chosen for data acquisition, implementation and validation of entire AI based system.