Rule-Based Particle Swarm Algorithm for Real-Time Optimization of Cement Clinker Burning Process

The cement clinker burning process consumes a significant amount of coal. To reduce coal consumption, the realtime optimization (RTO) system has become a critical focus of research in the optimization and control of cement production processes. The real-time optimization system leverages real-time data and accurate performance indicator models to dynamically adjust key control variables in the production process. This paper focuses on the real-time optimization of cement clinker burning process. Since the optimization process relies on accurate performance indicator models of performance indicators, this paper first introduces the production process of the cement clinker burning system and categorizes the key variables involved. On this basis, a detailed description of the modeling methods of performance indicators in the real-time optimization system is presented. After verifying the accuracy of the performance indicator models, this paper describes the optimization problem model within the real-time optimization system and proposes an improved version of the PSO algorithm, i.e., the rule-based particle swarm (RPSO) method to enhance the speed of convergence to the feasible region. Numerical experiments demonstrate that the proposed RPSO algorithm not only reduces production costs more effectively but also improves the operational stability of the real-time optimization system.