Extended fictitious reference iterative tuning and its application to chemical processes

A new, practical method for direct tuning of PID controllers using operation data under feedback control is proposed. Neither open-loop nor closed-loop identification is required in the framework of direct tuning, but conventional methods have the following problems: iterative experiments are needed, changes of the manipulated variable are not evaluated, and it is difficult to properly determine a reference model without information on a process. In the present work, to solve these problems and make direct tuning practicable for process control, extended fictitious reference iterative tuning (E-FRIT) is proposed. The major advantage of E-FRIT over other direct PID tuning methods is its practicability, which is realized by focusing on real application of E-FRIT to chemical processes. In the proposed algorithm, only the desired closed-loop settling time needs to be determined by users. The great usefulness of E-FRIT is demonstrated through its industrial applications. The MATLAB program of E-FRIT with the latest algorithm is open to the public through its official website; it has been widely used in the process industry in Japan.