By mean of the implementation of six large-scale demonstrations , the project intends to remove in three years the barriers to integrate both onshore and offshore wind power to a greater extent into the European system by 2020.
The large-scale demonstrations aim at proving the benefits of new power technologies (or the new kind of their application), the majority coupled with innovative system management approaches.
In particular, secure meshed HVDC networks were validated in France (“DEMO 3: DC Grid”) using simulations and limited scale mock-up, and by developing and testing a full scale prototype HVDC Circuit Breakers (DCCB) .
Within summer 2013, the DEMO 3 activities provided and demonstrated key building blocks for designing future High Voltage Direct Current (HVDC) networks which can be securely operated. DEMO 3 has also demonstrated innovative design for the integration of HVDC lines into existing AC systems. Investigations covered a significant scope of system and security management, ranging from the stable and reliable steady state operation to the detection, elimination of and recovery from large disturbances like DC network faults . The need and requirements for specific equipment and systems like master controllers, DCCBs and associated protection strategies were characterized in simulations and through a laboratory test mock-up; these requirements were confronted to current technological advances, especially through the large-scale demonstration test of a DCCB prototype, but also through operations on the first meshed HVDC Grid mock-up with physical cables and protection devices. Finally, the benefits and impacts of meshed DC Grids were studied in the context of the North Sea area, in comparison with the current approach of point to point DC connection of wind farms.
Recognizing that future offshore HVDC grids would most probably be built stepwise, the DEMO 3 activities distinguished three stages beyond radial DC connection of wind farms:
A first stage with small DC backbone grid-shaped (Fig. 1) should be implemented, which can be readily constructed and extended with currently available technologies, without specific equipment or systems like DCCBs or DC grid (DCG) master controllers.
Fig. 1 – Tree-like (plain lines) and meshed (plain and dashed lines) DC backbone grid examples (VSC are Voltage Source Converters; GS “Grid Side”; WS “Wind Side”)
Autonomous power flow controls (PFC) for both the DCG converters and the offshore wind turbines were exhibited, which demonstrated that flexible power flow control in normal and disturbed conditions, ancillary services to the AC mainland network (voltage support, frequency control, Power System Stabilizer), and Fault Ride-Through capability can be provided by such DCGs using local measurements only.
Fault clearance would then involve de-energizing the complete network from the onshore AC grid. Therefore the maximum power infeed from these networks must remain below acceptable values for such events, e.g. a few GW depending on the Frequency Containment Reserve- FCR (formerly primary reserve) of the synchronous zone it connects.
An intermediate stage - by 2020 - relies in simple meshed networks, for which specific equipment or systems like a DCG master controller would be required in addition to the controls mentioned for the backbone structures. To establish these requirements and assess the operation of such networks, a representative network topology with 5 Voltage Source Converter (VSC) terminals was used in simulation first. In a second stage, the various embedded controls in the converters (master-slave control, voltage control , and coordinated control) were experimentally validated through the grid behaviour, using a scaled-down five-terminal meshed DC Grid mock-up using Hardware In the Loop (HIL) simulation (Fig. 2) on actual and simulated equipment.
Fig. 2: DCG mock-up using Hardware In the Loop (HIL) simulation
In case of DC network fault, the rate of rise and amplitude of the fault current are dramatic. Therefore, a protection system based on DCCBs (DC Circuit Breakers) is required to selectively detect and clear DC network faults, as the loss of the complete DCG would not be acceptable. Three different classes of requirements were identified for the duty of DCCBs, depending on the ratings of the grid, but also the portions of the grid to protect. Two of them are met by the performances of the fast switch HVDC Circuit Breaker demonstrator (Fig. 3) which was designed to meet their stringent speed requirements at acceptable cost, as witnessed by an independent observer and the EC Technical Reviewer.
Fig. 3: DCCB demonstrator architecture
Fault clearing time constraints imposed also the pilot development of very rapid and selective optical fibres protection strategies. This scheme, based on differential overcurrent relays, was shown to be effective for cable distances no longer than about 200 km. Moreover, it was validated experimentally on the DCG mock-up, where opening orders are sent in less than 3ms.
An economic analysis focused on comparing possible DC topologies (radial DC connections , point-to-point DC links , HVDC multi-terminal grids or meshed HVDC networks ) in line with the development of offshore wind generation in the North Sea, based on the long-term planning and reliability assessment methods used by European TSOs, was carried out. It was quantitatively established that DCGs use HVDC underwater cable capacities more effectively than radial DC connection schemes to feed offshore wind power back to the continent, along with the additional benefit of interconnecting energy production areas at the European scale. The DCG can also implement very beneficial functions for the operation of the onshore AC grids connected to it, which were not assessed in the framework of the study: improved AC security margins through appropriate power injections via the onshore DCG terminals; ancillary services like voltage control, frequency support , synthetic inertia or damping of inter-area oscillations;black start capability restoration of the AC system from the offshore grid.
Global cost-benefit comparisons between radial DC connection and DCG schemes were carried out while varying parameters like the cost of CO2 emissions, of new DC technology (including the DCCB) and cable capacities. From this analysis there is no clear advantage or disadvantage between the studied schemes. Grid schemes are more costly in terms of investment but provide added benefits for operation and remain thus competitive overall. At the 2020 or 2030 horizon, other uncertainties like regulation criteria on structural adequacy of the European generation mix could also play a significant role in the balance.
The TWENTIES project gathered 26 partners from 11 countries:
- REE, Spain (TSO)
- RTE, France (TSO)
- Elia, Belgium (TSO)
- Energinet, Denmark (TSO)
- TenneT, Netherlands (TSO)
- 50 Herzt, Germany (TSO)
- CORESO, Belgium
- DTU Energy, Denmark
- COMILLAS-IIT, Spain
- FRAUNHOFER IWES, Germany
- SINTEF, Norway
- INESC-PORTO, Portugal
- UCD, Ireland
- RSE, Italy
- University of Strathclyde, UK
- University of Liege, Belgium
- KUL, Belgium
- ULB, Belgium
- Alstom Grid, UK
- ABB, Spain
- Siemens Wind Power, Germany
- Gamesa, Spain
- EWEA, Belgium
- Dong Energy, Denmark
- Iberdrola, Spain
- EDF, France.
This project article is linked with the following knowledge articles:
For more information, please visit http://www.twenties-project.eu/node/1