Predictive Control Of Reheating Furnace Based On Differential Evolution Algorithm

With the rapid development of modern steel industry and the rising energy crisis, the energy conservation and its effective use have become the new problems and new focus in the steel production process. Reheating furnace is not only an important equipment in metallurgical industry production, but also one of the largest energy-consuming equipment in the iron and steel industry. It is quite significant for the entire steel industry to deal with how to improve the efficiency of the reheating furnace, how to reduce energy consumption and how to enhance the competitiveness.The reheating furnace is a typical complex industrial control object with multivariable, time-varying, nonlinear, strong coupling, large inertia and time-delay characteristics, which the heating process is affected by a variety of factors during the production process. Thus, the optimal control of the heating furnace is a complex control and optimized problem which is difficult to achieve good optimal control by conventional techniques. In order to achieve optimal temperature control of reheating furnace, from the practical production of the billet reheating furnace and the complex reality of modern industry, combined with the actual controlled applications of reheating furnace and present research, as the main object of study, the temperature of the reheating furnace is analyzed as follows.(1) The predictive control of reheating furnace. Reheating furnace is a complex and nonlinear system which is difficult to describe by precise mathematical model. The feature of generalized predictive control is not strongly dependent on the mathematical model and it is well applied in the actual industrial production process. In order to overcome model prediction error caused by the uncertainty disturbances of the system and obtain a more precise predictive temperature value of the reheating furnace, in this paper, generalized predictive control algorithm model of reheating furnace is established and has achieved good experimental results.(2) Parameter tuning of predictive control. As the controlled temperature of reheating furnace is directly affected by the changes gas flow, in order to achieve precise controlled temperature, it is an important way to ensure its optimization. In this paper, the parameter tuning of PID controller based on differential evolution algorithm method is proposed to optimize of the control variables to achieve optimal control of the size of the input.(3) Modeling and simulation experiment. Good results have been achieved through high-level programming language C# and MATLAB after the simulation experiments to the above algorithm model.(4) The validation of experimental results. In order to verify generalized predictive control strategy based on differential evolution algorithm is with good feasibility and effectiveness of the control of reheating furnace temperature, the controlling strategies were compared with traditional classical experiments. The results show that these approaches proposed in this paper are more excellent.