Combustion Optimal Control And Warning System Development For Continuous Annealing Furnace

The annealing furnace section of Baosteel1550continuousannealing line is assigned with333"W"-type radiant tube burners. Due tocombustion ability/state variance as well as lack of proper combustionwarning system, these burners have encountered many radiant tube orcombustion flue burnout events. What’s worse, the system’s operationbecomes more complicated each day: the number of annealing steel typeshas increased from23to over50and the number of annealing curvesfrom12to19, coupled with many changes in specifications andannealing curves, the defects of the mathematical model used for initialdesigned heating chamber combustion control have been exposed. Theoriginal PID combustion controller cannot response to variouscharacteristics including time variation, nonlinearity and uncertainty,therefore, they can hardly deal with such production problems as stripheating temperature control. This article focuses on improving the controlsystem of annealing furnaces in the following respects:（1）The radiant tube heating warning model, established viatheoretical analysis and industrial tests, can be used for qualitativeanalysis of gas combustion status as well as analyzing the technicalparameters that influences the inner furnace’s heat transfer, therebyestablishing an organic connection for the annealing furnace amongthermal load distribution, combustion system configuration, and theoptimal design of waste heat recovery device. The successfuldevelopment of warning system for annealing furnace’s radiant tubecombustion control solves the technical problem of lacking on-linemonitoring over the cold-rolling continuous annealing furnace andprovides technical support for quality improvement of cold-rollingproducts and for the optimization of annealing furnace’s technical and economic index.（2）The mathematical model of annealing furnace combustion and theoptimized dynamic setting strategy for furnace temperature control aredeveloped. It is found that strip steel heating in the radiant tube furnace isinfluenced by the geometric sizes of furnace chamber, radiant tube, stripsteel and other inner equipments. According to industrial tests, thecomprehensive radiant coefficients of various combustion zones arecalculated via imaginary plane method. The temperature is assessed usingthe infrared radiation thermometer in the outlets of heating andaverage-temperature parts, the heat transfer mathematical model of stripsteel’s heating process is solved using recursive least squares method,both of which are used for dynamic identification of furnace temperatureand conveying speed. The established identification control equationfacilitates the rapid convergence of model’s solving process with amaximum offline error by only0.588%, and a maximum online averagecontrol precision of0.625%.（3）The transfer function for the annealing furnace’s combustionchamber is established by analyzing the relationship among suchvariables as strip steel heating temperature, gas flow, and exhaust gastemperature. The design of fuzzy adaptive PID controller of temperaturecontrol circuit is presented. and the MATLAB simulation, via which thePID controller’s parameter is obtained, is established. Moreover, theregulation performance of fuzzy adaptive PID controller and normalcontroller is compared. The newly designed fuzzy adaptive PIDcontroller’s systematic fuzzy reasoning and decision making, as well asonline setting of parameters-Kp, Ki, Kd-are achieved via comparingsimulation curves and parameters optimization Finally, the combustiondynamic control is carried out in a real annealing furnace.