Annealing Furnace Online Blackbody Cavity Parameters Establishment And Optimization

As an important step of production process, anneal is the heat treatment process to the cold-rolled or hot-rolled steel strip. During the process, the steel strip is heated to the temperature of process in a certain period of time, kept the temperature, and then declined in accordance with a rate. The strip temperature control is one of the most important parts of the continuous annealing process. Strip temperature control directly affects the structure property and the mechanical properties of steel products. Then it can determine the quality and plate of the strip.Basing on the method of PID-Smith, we achieve the temperature control of the continuous annealing furnace. Comparing the effectiveness and precision of control, we can singularize the importance of temperature detection in the temperature control of the strip. Because the strip of continuous annealing furnace is in motion and the material emissivity of strip will be changed within a certain range, the error of using traditional radiation thermometry to measure the temperature of strip is larger. It also restricts the effect and accuracy of strip temperature control. Basing on the online blackbody cavity radiation thermometry method, we can measure the real temperature of strip by measuring the temperature of the cavity.Basing on COMSOL Multiphysics simulation platform, we research the motion strip temperature of the continuous annealing furnace heating segmental. According to the three laws of heat transfer, we make the mathematic model. According to the model, we embed the simulation platform and analysis the heat transfer process of strip and furnace temperature. According to the COMSOL simulation diagram and the related data of industrial field, we draw the relevant parameters of online blackbody cavity. Since the online blackbody cavity is determined by the relevant installation parameters, we need to combine model algorithm and optimize the installation parameters. Then we can achieve the ideal installation parameters of blackbody cavity to meet the accuracy requirements of strip temperature control.Basing on the installation parameters of online blackbody cavity achieved, we make use of the BP neural networks and least squares support vector machine to learn and train the data. Then we derive relationship between input and output parameters and verify the accuracy of the model by testing samples. By the process and the error requirement, we determine the parameter programming model which obtains the target function and the constraint condition of related parameters. Basing on the programming model above and the parameter training models of BP and least squares support vector machine, we use the genetic algorithm to optimize the installation parameters. Finally, we achieve the optimized installation parameters of the blackbody cavity.In this paper, the installation parameters of the blackbody cavity achieved above is closed to the actual parameters of Benxi cold rolling production line. It verifies the feasibility of this method and provides guidance to measure the strip temperature of annealing furnace in actual production.