A looper-thickness coordinated control strategy based on ILQ theory and GA-BP neural network

With the focus on the 1580-mm hot rolling production line of a factory, this research aims at improving the stability of the hot rolling looper-thickness control system and the quality of hot rolling products. A coordinated control model on the basis of inverse linear quadratic (ILQ) theory and BP neural network based on genetic algorithm (GA) is proposed with the aim at the complex multivariable and strong coupling relationship among the looper angle, strip tension, and strip thickness. Aiming at the linear and time invariant characteristics of constant tension rolling stage, a looper-thickness coordinated control strategy based on ILQ theory is proposed. Aiming at the nonlinear and time-varying characteristics of the lifting and setting stages, a predictive control strategy based on GA-BP neural network is proposed. The GA-BP neural network gain compensator is added to the ILQ control system. The experimental results show that the thickness, looper angle, and tension fluctuation of the system with the compensator are 0.37 mm, 4.4 deg, and 6.39 MPa respectively smaller than those without the compensator. The control strategy in this research provides theoretical basis for improving the quality of hot rolled products and facilitates further promotion and application.