It is well known that industrial processes generally have nonlinearity, time-varying features and uncertainty and that active disturbance rejection control is effective in controlling these industrial processes. The control of industrial processes, especially those with time delay, is becoming more difficult.
In several different production processes of industrial practice exist time delays such as state delay, input delay, transmission delay, output measurement delay, capacity delay, etc. Because of the time delay, the manipulated value cannot immediately reflect the disturbance that the system receives, the response characteristic has a large overshoot, adjustment time gets long, and even the system becomes unstable. Thus, the time-delay system is accepted as a difficult one to control.
Although a number of control schemes have been proposed to overcome the effect of time delay, it is very difficult to build accurate mathematical models for systems that are becoming increasingly complex, which results in the limitation of the application of current control theory and methods.
Applying active disturbance rejection control to time-delay systems can achieve good effects. The design methods for the active disturbance rejection control system of time-delay systems include time-delay neglect, dimensionality increase, output prediction, input prediction and input delay. In all these methods, a delay object is approximated or transformed into a delay-free object, and then an extended state observer is designed for observation.
Kim Ho, dean of the Faculty of Automation Engineering, has proposed a strategy for applying a multivariable linear active disturbance suppression control method to the brine heating process where there is a bias interference between the time delay and the control variables.
To compensate for the time delay, he used the input delay method in the extended state observer design. In addition, he constructed and introduced the decoupling matrix using the mathematical model information of the object to further enhance the decoupling effect.
The simulation results show that the performance of the proposed control system is very high.
© 2021 Kim Chaek University of Technology