An ADRC Parameters Self-Tuning Controller Based on RBF Neural Network for Multi-Color Register System

To improve the control precision of nonlinear register system for flexographic printing, a feedforward active disturbance rejection (ADRC) parameter self-tuning decoupling strategy based on radial basis function (RBF) is proposed to address existence coupling interference and multiple working conditions. Firstly, according structure printing equipment registration principle, mathematical model global established linearized using small deviation method. Secondly, decoupled controller designed by integrating control, ADRC, RBF, in which used eliminate errors caused modeled disturbances, ADRC performs estimation compensation unmodeled RBF realizes parameters. Finally, different operating conditions are simulated compare verify performance controller. Simulation results show that has better compared traditional PID its error peak reduced about 32% achieving high accuracy multi-color system.