masoomeh taherkhani - Department of Electrical Engineering Iran University of Science and Technology Tehran, Iran
abdoreza rahmati - Department of Electrical Engineering Iran University of Science and Technology Tehran, Iran

Although learning-based control method base on BSpline Network (BSN) together with PD controller presents an excellent steady-state performance for Uninterruptible Power Supply (UPS) Inverters but it may not achieve high dynamic performance under non-periodic disturbances, which can be inappropriate for high performance UPSs. To further reduce the transient errors, in this paper a simple design and effective Static Output Feedback (SOF) robust controller instead of PD controller is adopted parallel with the BSN. Simulation results under various conditions show that the proposed controller can achieve not only better dynamic performance but also can return better steady-state performance compare to those of a BSN with PD scheme developed earlier whereas it's structure is also simpler.

B-spline neural controller, inverter, static output feedback controller, uninterruptible power supply