(372al) Pinn-Based Modular Digital Twin Development for Sustainable Electroplating Manufacturing
AIChE Annual Meeting
2024
2024 AIChE Annual Meeting
Computing and Systems Technology Division
10B: Interactive Session: Systems and Process Control
Tuesday, October 29, 2024 - 3:30pm to 5:00pm
We introduce a novel methodology for developing plant-wide MDTs, offering unprecedented flexibility in process management and scalabilityâessential for tailoring to various operational scenarios within the electroplating systems. By embedding physical laws directly into the neural network learning process, our PINN-based MDTs provide a more accurate and comprehensive analysis of plant design and operations. This innovation not only improves the real-time monitoring and control of electroplating processes but also facilitates a deeper understanding of the complex interactions within these systems. As a result, our MDT system can dynamically assess sustainability performance, identifying opportunities for resource optimization, waste reduction, and energy savings with high precision; thereby significantly improving decision-making for sustainability performance enhancement. The case studies illustrate the superiority of dynamic sustainability assessments conducted with our MDT system over traditional static methods, highlighting significant improvements in resource efficiency, waste reduction, and economic viability. This research not only contributes a groundbreaking tool for the electroplating industry but also sets a new standard for the application of DT technology and AI in manufacturing for sustainability.