Based on the fundamental principles of universal model adaptive control, the paper improved the method of universal model control law and compared by simulation. In order to ascertain the control parameters of universal model adaptive law that is learning step and penalty factor, made large numbers of step response experimentations on typical industrial self-balance model, summed up tuning rules of parameters in universal model controller when object's parameters: static gain K and process time constant T and postponed time r changed. In order to improve control performance, designed a UMAC-PID cascade control system and simulated on main steam temperature system and main steam pressure system, proved that the control structure can overcome disturbance quickly and more adaptive to load change and more economical. |