TY - JOUR
T1 - Local Reactive Power Control Methods for Overvoltage Prevention of Distributed Solar Inverters in Low-Voltage Grids
AU - Demirok, Erhan
AU - Gonzalez, Pablo Casado
AU - Frederiksen, Kenn H. B.
AU - Sera, Dezso
AU - Rodriguez, Pedro
AU - Teodorescu, Remus
PY - 2011/12/7
Y1 - 2011/12/7
N2 - voltage (LV) grids by means of solar inverters with reactive power control capability. This paper underlines weak points of standard reactive power strategies which are already imposed by certain grid codes, and then, the study introduces a new reactive power control method that is based on sensitivity analysis. The sensitivity analysis shows that the same amount of reactive power becomes more effective for grid voltage support if the solar inverter is located at the end of a feeder. Based on this fundamental knowledge, a location-dependent power factor set value can be assigned to each inverter, and the grid voltage support can be achieved with less total reactive power consumption. In order to prevent unnecessary reactive power absorption from the grid during admissible voltage range or to increase reactive power contribution from the inverters that are closest to the transformer during grid overvoltage condition, the proposed method combines two droop functions that are inherited from the standard cos ϕ(P) and Q(U) strategies. Its performance comparison in terms of grid losses and voltage variation with different reactive power strategies is provided by modeling and simulating a real suburban LV network.
AB - voltage (LV) grids by means of solar inverters with reactive power control capability. This paper underlines weak points of standard reactive power strategies which are already imposed by certain grid codes, and then, the study introduces a new reactive power control method that is based on sensitivity analysis. The sensitivity analysis shows that the same amount of reactive power becomes more effective for grid voltage support if the solar inverter is located at the end of a feeder. Based on this fundamental knowledge, a location-dependent power factor set value can be assigned to each inverter, and the grid voltage support can be achieved with less total reactive power consumption. In order to prevent unnecessary reactive power absorption from the grid during admissible voltage range or to increase reactive power contribution from the inverters that are closest to the transformer during grid overvoltage condition, the proposed method combines two droop functions that are inherited from the standard cos ϕ(P) and Q(U) strategies. Its performance comparison in terms of grid losses and voltage variation with different reactive power strategies is provided by modeling and simulating a real suburban LV network.
U2 - 10.1109/JPHOTOV.2011.2174821
DO - 10.1109/JPHOTOV.2011.2174821
M3 - Journal article
SN - 2156-3381
VL - 1
SP - 174
EP - 182
JO - I E E E Journal of Photovoltaics
JF - I E E E Journal of Photovoltaics
IS - 2
ER -