A Set-Membership Affine Projection Algorithm-Based Adaptive-Controlled SMES Units for Wind Farms Output Power Smoothing

This paper presents a novel adaptive control scheme of the superconducting magnetic energy storage (SMES) units with the purpose of smoothing the wind farms' output power. The adaptive control scheme is based on the set-membership affine projection algorithm (SMAPA), which provides a faster convergence and less computational complexity than the normalized least-mean-square algorithm. In this study, two grid-connected fixed-speed wind farms are considered. The control strategy of the SMES units is based on a pulse width modulation (PWM) voltage source converter (VSC) and a dc-dc converter. The VSC and dc-dc converter is used to control the reactive and active power exchange with the power system, respectively. The SMAPA-based adaptive proportional-integral (PI) controllers are utilized to control both converters. For realistic responses, real wind speed data extracted from Hokkaido Island, Japan, and two-mass drive train model of the wind turbine generator system are used in the analyses. Also, a real 10 MVA SMES unit that was installed at a power plant in Kameyama, Japan, is connected to the point of common coupling of the wind farms. The validity of the proposed control scheme is verified by the simulation results, which are performed using PSCAD/EMTDC environment. With the SMAPA-based adaptive-controlled SMES units, the wind farms' output power can be smoothed easily avoiding huge effort for fine tuning the controllers' parameters.