Dynamic temperature modeling of continuous annealing furnace using GGAP-RBF neural network

Dynamic modeling of the quality control of a real large-scale continuous annealing process is studied in this paper. This continuous annealing process consists of several sub-processes and there exists unknown complex nonlinear mapping between the sub-process set points and the final annealing quality. The quality model should be constructed and updated based on the new data sequentially collected from the real process in order to optimize the set point of each sub-process dynamically. To meet this demand, a latest developed sequential learning algorithm called generalized growing and pruning RBF (GGAP-RBF) neural network is used to establish the required dynamic quality control model. On-line application of this quality model on the continuous annealing furnace in a steel factory has been conducted and the actual performance is as good as required.