On the Use of Neuro-Fuzzy to Control a Three-Phase Uninterruptible Power Supply

Abstract

This paper introduces the application of neuro-fuzzy to control a three-phase uninterruptible power supply (UPS) for critical loads. The introduced controller incorporates adaptive neuro-fuzzy inference system (ANFIS) in rotating reference frame with three input variables and one control output variable. With this methodology, the problems caused by nonlinear loads and parameter variations of UPS system are solved. Moreover, it achieves high dynamic performance and accurate voltage regulation of the load with minimum total harmonics distortion (THD) and good steady-state characteristics. Simulation and experimental results on a three-phase prototype show effectiveness and advantages of the proposed controller especially under rectifier-type loading condition