An algorithm for automatic tuning of PID processes is proposed. The authors utilize the area under the controller response and the process response to calculate initial conditions and to constrain the least square regression to an adequate estimation of the process transfer function. The tuning algorithm is intended to guide the user in performing a loop by loop tuning of the utility control loops. Thus it supplies the user with full graphics support, reliability indices and tuning menu. The reliability indices are based on the correlation between the estimated response in the closed loop and the data. The tuning menu includes clear definitions of the tuning goals, i.e. tracking, regulation, etc. All of these guided the user to achieve the utility tuning goals with relatively simple steps.
展开▼
