Development of Frequency Limited Model Reduction Algorithm with Error Bound and Application to Continuous-Time Variable-Speed Wind Turbines for Power System

摘要

The frequency-limited model reduction technique presented by Gawronski & Juang produces an unstable reduced-order system. To overcome this main drawback, many techniques were proposed as a solution to preserve the stability of the reduced-order system. However, these available approaches produce a large variation to the original system and yield a large approximation error. The proposed approach not only ensures the stability of the reduced-order systems but also provides low approximation error when compared with existing approaches. Moreover, it also yield easily computable a priori error bound formula. Simulation results show proposed scheme yields low approximation error when compared with existing approaches while performing model reduction of variable-speed wind turbine systems which shows efficacy of the proposed work.