Active Power Control of a Natural Gas/Fuel Oil Turbine Power Plant with Adaptive Neuro-Fuzzy Inference System-Based on Modern Controllers


Tabakh R., TİRYAKİ H., BAYHAN N.

4th International Conference on Intelligent and Fuzzy Systems (INFUS), Bornova, Türkiye, 19 - 21 Temmuz 2022, cilt.504, ss.735-743 identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası: 504
  • Doi Numarası: 10.1007/978-3-031-09173-5_84
  • Basıldığı Şehir: Bornova
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.735-743
  • Anahtar Kelimeler: Power systems, Active power control, Gas/Fuel oil turbine power plant, PI, FGPI, ACO-PI, ANFIS-PI
  • İstanbul Teknik Üniversitesi Adresli: Evet

Özet

The settings of conventional controllers, which are commonly used in power plants, are determined based on system characteristics during the establishment phase, and, thus are unable to adjust to changing system dynamics over the life of the plant. To avoid this unsatisfactory state, the parameters of the controllers used in electric power plants should be self-adapting. As a result, the goal of this paper is to determine the optimal parameter settings using Ant Colony Optimization-based PI controller (ACO-PI), Fuzzy Gain Scheduled PI (FGPI), Adaptive Neuro-Fuzzy Inference System based PI (ANFIS-PI) and, conventional PI controller for proportional-integral (PI) controllers that will adapt to the changing dynamics of a natural gas/fuel oil turbine power plant's system model and, comparing the system output signal's transient responses. The results show that the ANFIS-PI controller outperforms the other methods.