Digital twin for zinc roaster furnace based on knowledge-guided variable-mass thermodynamics: Modeling and application

Abstract

Roaster furnace is a large-scale equipment in zinc smelting process, which plays an significant role in safety production and enviromental protection. An accurate digital twin of roaster furnace can help to explore the influence of control parameters and serve as a test platform to verify the effectiveness of control strategies. However, the traditional thermodynamics cannot be used for dynamic modeling, and inability to measure key data greatly affects the accuracy of kinetic modeling. Therefore, this paper establishes a novel digital twin of zinc roaster furnace based on knowledge-guided variable-mass thermodynamics. First, based on the integration of mechanism analysis for mass balance and energy balance, a dynamic modeling method is proposed. Then, particle swarm optimization (PSO) algorithm is introduced to optimize the parameters of conversion rates guided by knowledge. Finally, by connecting the dynamic model with distributed control system (DCS) through OPC communication protocol, a digital twin of roaster furnace is constructed. Extensive experiments show that the simulation results of the digital twin roughly agree with the actual industrial data under steady and dynamic working conditions, and the application of the digital twin on the performance analysis of control parameters and testing of control strategies can provide guidance for the optimal control of roaster furnace.