Optimizing Disturbance Rejection by Using Model-Based Compensator with User-Defined High-Frequency Gains

The paper presents extension of model-based disturbance rejection method based on Magnitude optimum method. The original method requires the pre-chosen values of controller and compensator filter values, which might result in a relatively noisy process input signals that are not tolerated in practice. In order to keep the process input noise under control, the proposed tuning method requires user-defined controller and compensator noise gain for high frequencies, instead. The experiments on process models show that the proposed approach is successful even though the actual and the identified process models do not match.