Parameter Identification Of Heat Transfer Model For Continuous Casting Based On Particle Swarm Optimization

The heat transfer coefficient of secondary cooling is a very important parameter in the process of continuous casting.In the control of secondary cooling in continuous casting,the establishment of heat transfer model is the key to realize the dynamic control of secondary cooling in continuous casting.The accuracy of heat transfer coefficient is the main factor affecting the effective application of heat transfer model.In order to obtain accurate secondary cooling heat transfer coefficient,the commonly used parameter identification method is to use the slab surface temperature measurement to back-calculate,but due to the limitation of field conditions,the temperature of some cooling sections can not be measured on-line,so the empirical value is usually used,which leads to inaccurate heat transfer coefficient,especially when the working conditions change,the predicted value of model will deviate from the actual value.In view of the incomplete information of slab surface temperature measurement,this thesis studies how to identify the heat transfer coefficient of each cooling section.The main contents of this paper are as follows:(1)Establishment of heat transfer model for solidification of continuous casting.The heat transfer characteristics of continuous casting billet during solidification process are analyzed.A two-dimensional mathematical model of solidification heat transfer in continuous casting is established.The method of solving the differential equation of solidification heat transfer is given,which is also called the finite volume discretization model.The effects of casting speed,superheat and secondary cooling heat transfer coefficient on the output of the solidification heat transfer model are analyzed.(2)Establishment of an inversion model for calculating the heat transfer coefficient of secondary cooling.The optimization problem is an important part in the process of solving the inverse heat transfer model of secondary cooling in continuous casting.The parameters of the model and the fitness function are determined.Based on the established solidification heat transfer model and the actual temperature of slab surface,the inverse heat transfer coefficient model of slab surface in secondary cooling zone is established.(3)Solving the ideal model.In order to solve the inversion model of secondary cooling heat transfer coefficient,finite iterations of the inverse model are carried out by particle swarm optimization with different parameters.The accuracy of the identification results is almost the same as the original results.(4)Further solving the actual model.For the actual inversion model of secondary cooling heat transfer coefficient,the accuracy of the first identification results is low due to the lack of constraints in fitness function.In order to obtain accurate secondary cooling heat transfer coefficient and improve identification accuracy,constraints are added to fitness function and identification algorithm is improved.Finally,the optimization results are simulated with the simulation software of MATLAB.