Approach of Dynamic Simulation and Optimization for Coevolutionary Intelligent Manufacturing Cell in Digital Twin

In real manufacturing industry, the timely identification of weaknesses and precise determination of optimization areas are crucial for efficient workshop production. Traditional plant simulation methods have been widely employed to support decision-making, but the accuracy of these decisions heavily depends on the reliability of the simulation results. Static simulations, which rely on manually set parameters often fall short of meeting these demands. To address this, dynamic simulations incorporating real-time data are essential for generating accurate and actionable insights. A worth-trying approach is proposed therefore in this paper by establishing a plant simulation environment in Digital Twin(DT) utilizing a high-fidelity real-time coevolution model for intelligent manufacturing cells. To get the coevolutionary models in DT system, data modeling and integration, virtual-real event driven method are illustrated. This approach provides a true-to-life workshop simulation environment. By employing variable and customizable optimization strategies, diverse simulation outcomes can be generated, aiding in the selection of the most effective optimization strategies. The development of a coevolutionary DT for a manufacturing cell within an enterprise is presented as a case study, demonstrating the effectiveness of dynamic and combo-strategy production simulations.