A proposed methodology to develop digital twin framework for plasma processing

Abstract

Plasma processing presents complex behavioural characteristics that are created by different generation and processing methods. In this context, the digital twin (DT) method offers a visualisation and simulation techniques that enable better understanding of plasma complexities, control, and precision. With the rise of Industry 4.0, DT applications have gained increased attention across engineering sectors. However, despite growing interest in this digital technique, DT characteristics and basic principles vary within the literature, leading to confusion about the architectures and capabilities of this method. This study proposes a comprehensive DT framework tailored for plasma processing aimed at bridging the gap in current literature. Two frameworks are introduced. The first introduces a modular approach at a sub-component-level, incorporating the elements required for successful DT integration in practice. The second focuses on a DT method that offer data acquisition and machine learning integration for parameter optimisation at the plasma process-level, presenting a foundation for future real-time developments. These frameworks are built upon successful DT practices in advanced manufacturing technologies and offer a flexible premise adaptable to various plasma processing techniques. However, integration of enabling technology remains the major limitation, as case studies remain conceptual, therefore requiring future physical implementation, specifically for a goal-oriented DT.