Optimization Control Of Reheating Process In Rotary Reheating Furnace

The main task of the rotary reheating furnace is to provide the qualified billets in heating quality for the rolling process, which meet the requirements of rolling technology. As the key link and high energy consumption equipment in hot rolling production line of seamless steel tube plant, its control level can directly affect the quality of steel tube, productions, cost of the whole system, and so on. The rotary reheating furnace is a complex industrial process controlled object. There are a lot of external disturbance factors, and it can't directly and accurately real-time measures the billet temperature distribution in heating process. Because the traditional control strategy is hard to adapt to the complex and changeable thermal condition of the heating furnace, it can only rely on operators with experience to set the temperature of each control zone in order to control the billet temperature in the furnace. Such method often makes the control accuracy decline and energy consumption increased. In order to improve the control level of the furnace, all kinds of optimization control method come about at serial, but such models need to create a lot of approximation and assumptions, and the precision of models are not enough, also it is difficult to improve the control performance. In order to overcome the existing problems, optimization control method of the rotary reheating furnace based on modified PSO is presented. Establish temperature prediction model of the round billet and search the optimal temperature settings. It can reduce billet temperature deviation at the exit, difference of temperature in cross section and energy consumption.It introduces the rotary reheating furnace configuration, characteristics of the technological process and control requirements. It summaries the control project of the furnace and analyzes the process control system of the rotary reheating furnace. As to the existent problems, the following aspects were studied in this paper.First, for the problem that temperature of the round billet are known, temperature prediction model of the round billet surface at the exit based on RBF neural network are established. The simulation result shows the error of the model is less than 20℃, which meets the requirements of the heating process.Second, the objective function constituted by the minimum value of temperature deviation of billet surface at the exit and energy consumption in this paper. It adopts a modified PSO algorithm to search the optimal temperature settings. It can solve the blindness and randomness of temperature settings. It adopts mutation strategy to improve premature convergence problem resulting from population degradation and the search precision.Third, simulate billet temperature field through the finite element software of ANSYS in this paper. Simulation results show that optimized energy consumption decline, and difference of temperature in cross section decrease. It verifies the effectiveness of the optimization.