An Experimental Study of Modeling and Self-Tuning Regulator Design for an Electro-Hydro Servo-System

To design model-based controllers for industrial systems, an accurate mathematical model is necessary. Over time, most of these models are changed and the former controllers should be modified to achieve good performance. Hydro systems, according to their wide application over last forty years, are an interesting field of research. In this paper, based on the system identification and adaptive control theory, a special scheme for modeling and controller design of an old Electro-Hydro servo-system is proposed. Industrial process control suffering from noise and saturation as two main problematic factors. Here by conducting an experiment, the identification and control are employed in linear period of the nonlinear system. To specify the order of system, Matlab Ident toolbox, Hankel matrix and Hankel singular values are used. Recursive Least Square (RLS) method is implemented to determine estimated parameters. Designing a practical self-tuning regulator is a significant task too. Due to the inappropriate results of controlled system, a corrected regulator is generated. Finally, acceptable simulations are recorded and demonstrated.