Self-tuning PI Controllers via Fuzzy Cognitive Maps

Yesil E., Dodurka M. F., Sakalli A., Ozturk C., Güzay Ç.

9th IFIP WG 12.5 International Conference on Artificial Intelligence Applications and Innovations (AIAI), Paphos, CYPRUS, 30 Eylül - 02 Ekim 2013, cilt.412, ss.567-576 identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası: 412
  • Basıldığı Şehir: Paphos
  • Basıldığı Ülke: CYPRUS
  • Sayfa Sayıları: ss.567-576
  • İstanbul Teknik Üniversitesi Adresli: Evet

Özet

In this study, a novel self-tuning method based on fuzzy cognitive maps (FCMs) for PI controllers is proposed. The proposed FCM mechanism works in an online manner and is activated when the set-point (reference) value of the closed loop control system changes. Then, FCM tuning mechanism changes the parameters of PI controller according to systems' current and desired new reference value to improve the transient and steady state performance of the systems. The effectiveness of the proposed FCM based self-tuning method is shown via simulations on a nonlinear system. The results show that the proposed self-tuning methods performances are satisfactory.