Reducing Harmonic Instability and Resonance Problems in PMSG Based Wind Farms

Unlike conventional power systems where harmonic resonances are coming from passive inductive and capacitive elements, large-scale power electronic systems like wind farms present a more complex system, where the fast dynamics of the power electronic converters may present an inductive or capacitive behavior. Therefore, the interactions between the fast controllers of the power converters and the passive elements may lead to harmonic instability and new resonances at various frequencies. This paper presents an optimum design technique for the Wind Turbine (WT) inner controllers in a PMSG based wind farm in order to reduce the number of resonances and to mitigate harmonic instability. In the approach, a PMSG based wind farm is modeled as a Multi-Input Multi-Output (MIMO) dynamic system by modeling the high bandwidth control loops of the power converters. Resonance frequencies and oscillatory modes of the wind farm are identified based on the MIMO matrix. Afterwards, a multi-objective optimization procedure based on Genetic Algorithm (GA) is proposed to put the oscillatory modes of the wind farm in suitable locations in order to minimize the number of the resonances and to guarantee a stable operation of the wind farm. A 400-MW wind farm is studied in the PSCAD/EMTDC software environment to confirm the validity of the proposed optimum design technique.