作者: Sima Seidi Khorramabadi , Alireza Bakhshai
DOI: 10.1109/JESTPE.2014.2331188
关键词: Power control 、 Adaptive system 、 Adaptive control 、 Engineering 、 Control system 、 Fuzzy control system 、 Control theory 、 Control engineering 、 Control theory 、 Microgrid 、 Intelligent control
摘要: This paper presents an adaptive and intelligent power control approach for microgrid systems in the grid-connected operation mode. The proposed critic-based system contains a neuro-fuzzy controller fuzzy critic agent. agent employs reinforcement learning algorithm based on neuro-dynamic programming. feedback is made available to agent’s input as controller’s action previous state. evaluation or signal produced by together with back-propagation of error then used online tuning output layer weights controller. shows superior results compared traditional PI control. transient response time significantly reduced, oscillations are eliminated, fast convergence achieved. simple design improved dynamic behavior make it promising nominee systems.