Tuning methods for PI controller - Comparison on a highly modular drive

Abstract

This paper discuss PI controller optimization methods procedure. In its nature PI controller is a simple yet effective method for achieving desired reference, but tuning of the same is entirely different concept. There are numerous different methods for tuning a PI controller. Some optimizing one concept of control, i.e. current overshoot, other targeting different areas, i.e. settling time. Even with most advanced methods, sometimes, response has some unfavorable features. Sometimes PI tuning can require manual inputs, to lessen those inconveniences. Other times though, some tuning methods can prove to have better responses on the actual system than during simulations. Where better performances would be expected, sometimes, some unfavorable features can surface. This paper aims to show simulation and experimental comparison between two standard PI controller tuning methods. Methods were tested in Laboratory for Power electronic and electrical drives at Faculty of Technical Sciences. © 2013 IEEE.