PO189

Testing LVRT capabilities of wind converters oriented to offshore in a low scale test bench considering dynamic oscillations.

Markel Zubiaga 1, Guillermo Nuņez1, Danel Madariaga1, Iņigo Garin1, Jokin Aguirrezabal1, Ignacio Zubimendi1, Eduardo Burguete2, Mikel Zabaleta2, Igor Larrazabal1
1Ingeteam Power Technology, Zamudio, Spain, 2Ingeteam Power Technology, Sarriguren, Spain

Abstract

 

Power grid emulators have been spreading in recent years in order to test distributed renewable generation systems. Their main advantage is the possibility to emulate extreme grid disturbances as voltage faults and recreate any operating point of the grid. These characteristics make grid emulators an excellent way to validate the capability of any equipment upon specific grid operator requirements.

Offshore wind turbines are connected to inter-turbine grids several kilometers away from shore. The definition of these grids can vary significantly from one location to another and their definition is not as clear as it is onshore.  Thus, besides usual testing procedures to fulfill grid codes, the emulation of dynamic oscillating transients of the offshore grid can provide an important insight to ensure the correct integration in the grid of these wind turbines.

The developed test bench has not enough rated power to test the real equipment (up to 12MW) because of practical reasons. To perform fault tests with a grid emulator, at least 50% more rated power than the equipment under test (EuT) is convenient. To optimize the cost of the system and operability, the EuT is scaled.

 

Method

 

The developed test bench allows the generation of specific voltage dips profiles (and conditions) to perform further tests. Testing the scaled wind converter on both, on worst case voltage dip profile (classic approach) and on simulation estimated data (or real recordings) of:  voltage evolution, oscillating transients and frequency profiles.

The control hardware and the control code in the real equipment and the scaled one are the same, with the same control logics, auxiliary equipment and input/output signals. The components of the main power circuit are scaled maintaining the same per unit value.

The grid emulator is based on a back-to-back conversion line rated at 750kVA - 690V. The scaled EuT is composed with two 3L-150kVA conversion lines.

 

Results

 

The control code implemented is the same for three different cases: simulation scenario, test bench and real equipment. The voltage fault data obtained using simulation models can be used to configure the grid emulator. This way, some specific control structures oriented to improve the equipment behavior under grid disturbances can be tested in a low scale replica. So it is possible to verify the simulation results oriented to grid disturbances.

The test bench can be configured to reproduce any voltage profile, oscillation (<1 kHz) and fundamental frequency profile. Thus, even real event recordings can be tested.

 

Conclusions

 

By using simulation scenario data (or real recordings), it is possible to configure the grid emulator to reproduce the same event. These allows going one step forward from the simulation analyses and test the equipment (low scale replica) under specific voltage dip profiles and conditions. The expected fault conditions at the real site. Verifying the correct behavior of the control structures (Phase locked Loop, midpoint regulation, Filter under oscillations, etc.).

 

Objectives

 

Testing the scaled wind converter replica on both, on worst case scenario voltage dips (classic approach) and on real voltage fault recordings or simulation estimated voltage/frequency profiles, can help minimizing the uncertainties associated to going offshore. Furthermore, if after the commissioning of the wind turbine, an unexpected event is reported, a grid emulator presents a way to recreate the event in a controlled environment in order to improve the functionality of the wind turbine.