Abstract

Background: For controlling a back-to-back three-level (3L) Neutral-Point-Clamped (NPC) converter of a Permanent-Magnet Synchronous Generator (PMSG) based Wind Energy Conversion System (WECS), a Refined Model Predictive Control (RMPC) strategy is presented in this paper.

Methods: The RMPC strategies of PMSG-side and network-side converters aim at prevention of the high jumps in line and pole voltages of both the converters. A suitable subset of Switching States (SSs), which can prevent these high jumps, is predetermined for each SS. In each control cycle, the RMPC algorithm determines an optimal switching state from an appropriate candidate set of SSs.

Results: The functioning of the RMPC algorithms is checked with simulation tests. The effect of parameter uncertainty on the functioning of the RMPC algorithm is also studied.

Conclusion: The test results indicate that the RMPC scheme can block high jumps in pole and line voltages while extracting maximal power from wind, controlling network side reactive power, and balancing the capacitor’s voltage.

Keywords: Back-to-back, high jumps in voltages, model predictive control, neutral-point-clamped, parameter mismatch, PMSG, three-level, wind energy.