Research On Decoupling Control For Strip Flatness And Gauge Based On Improved PSO

With the development of China's economy,the domestic steel industry is advancing by leaps and bounds.Many of the world's leading steel industries have come from China,but limited by technical reasons,today China still uses crude steel as the main product of steel.Strip shape and thickness are two very important parameters in the hot rolling process.When the two parameters are studied in the early stage,they are generally distinguished and studied independently,which leads to the poor quality of steel products.In fact,in order to accurately control the shape and thickness of the plate and improve the quality of the product,the flatness thickness system(AFC-AGC)should be considered as a comprehensive multivariable control system.In view of the strong coupling,nonlinear and multivariab le characteristics of AFCAGC synthesis system,a large number of control methods have been developed by relevant experts at home and abroad.It has been proved that the conventional control measures can not achieve satisfactory control effect in the end.In order to deal with this situation,it is urgent to develop control theory and means that are applicable to the system.Based on the actual data of sheet thickness in hot rolling field,a comprehensive mathematical model of strip thickness for hot strip rolling is established in this paper.Then,a decoupling strategy based on the improved immune particle swarm optimization algorithm(IPSO)and PID neural network(PIDNN)is proposed to form a new PIDNN controller.The join weight of PIDNN is optimized by IPSO,which effectively solves the defect of premature convergence caused by gradient learning algorithm.The simulation results of MATLAB show that this algorithm has obvious advantages over traditional PID and single neuron PID control methods in both dynamic and static performance.It can provide some reference for decoupling problem in control field.Figure 33;Table 3;Reference 50...