Research on flatness intelligent control via GA–PIDNN

The traditional flatness control methods have the problems of limited control accuracy, slow responding and difficultly establishing a precise mathematical model, in addition, the traditional BP optimal algorithm exists the shortage of easy trapped in local minimum, flatness intelligent control based on GA–PIDNN was proposed. PIDNN controller does not rely on the mathematical model of controlled object and has excellent characteristics in the control system. Genetic algorithm (GA) has good parallel design structure and characteristics of global optimization. Flatness recognition model and flatness predictive model are established via GA–PIDNN by combining the merits of GA and PIDNN, on this basis, for the 900HC reversible cold rolling mill, the GA–PIDNN controller was designed to control the flatness defects effectively. Through the comparative study of PIDNN optimized by BP optimal algorithm, the simulation results show that this control method can quickly track flatness target value, improve flatness control accuracy, achieve good control result, and it is an effective flatness control method.