The EU climate and energy package — known as the 20-20-20 target — places Europe as a world leader in the field of sustainable and renewable energy. By 2020, the continent could add around 600 TWh in new renewable electricity generation.
However, predictions for offshore wind energy suggest that by 2020, offshore wind will be responsible for 28 percent of the total wind energy generation, itself a third of all renewable electricity generation, according to EU forecasts. This estimate equals a total of 44 GW of installed offshore capacity throughout Europe by the end of this decade, an average of 4.1 GW annually.
Since most of the offshore energy in Europe is to be installed in the Baltic or North Sea, the integration aspects of this offshore power to the different national electricity grids constitute a very important challenge to what is already a very challenging engineering task. Several European studies, for example Tradewind and OffshoreGrid, have shown that electricity networks in Europe will require major reinforcements. Nevertheless, along with important onshore grid reinforcements Europe will also need to develop an offshore grid infrastructure to efficiently integrate large amounts of offshore wind.
The first step towards a European offshore grid network was taken on 7 December 2009, at the EU Energy Council in Brussels where nine European countries — Belgium, Denmark, France, Germany, Ireland, Luxembourg, the Netherlands, Sweden and the UK — signed a political declaration for joint cooperation on the development of a transnational electricity infrastructure in the North Sea. A year later, the countries taking part agreed to make available, by 2012, a series of deliverables on grid configuration and integration, market and regulatory issues, planning, and authorisation procedures for the construction of the mooted transnational offshore grid.
Inside this vision of Europe’s future grid, the North Sea Transnational Grid project (NSTG) aims to identify and study the technical and economic aspects of connection of offshore wind power and trade between countries. The project is jointly executed by the Energy Research Centre of the Netherlands (ECN) and the Delft University of Technology; it was started in October 2009 and will continue for four years.
Transmission technology for the NSTG
There are two transmission technologies available for the connection of offshore wind farms to onshore networks: high-voltage AC (HVAC) and high-voltage DC (HVDC). The choice of which transmission technology to use will be based on efficiency and economic viability.
In comparison with HVDC systems, HVAC transmission systems have a wider dissemination, are more straightforward to install and present a lower offshore footprint. To date operational offshore wind farms in Europe have been connected through HVAC systems. The main reasons include the fact that only a few offshore wind farms currently have power ratings above 200 MW, and almost all are less than 30 kilometres from shore.
However, it is not always economically viable (or technically feasible) to use AC. For instance, the BARD Offshore 1 (or BorWin1) wind farm, scheduled to be operational in 2012, will be connected using HVDC. The 400-MW BorWin1 will be located 130 km from the German coast, justifying the choice of DC as to cross long distances by means of submarine cables (~60-100 km) the HVDC solution starts to be preferable. Traditional HVAC lines have higher losses (due to skin effect and capacitive leakage current) and demand additional equipment to provide reactive power compensation. On the other hand, DC cables do not suffer from leakage current and thus, in steady state, their electricity transmission is limited only by cable resistance.
The Role of Modularity
The North Sea Transnational Grid project, with its intention of interconnecting around 60 GW of offshore wind power between several countries in the North Sea up to 2030, is a very ambitious initiative. For projects of such dimension and complexity, choosing the most appropriate construction architecture is extremely important right from the start. The NSTG will have to organically grow with time from its initial, inherently simple phase to its desired final form, expected to exhibit a much more complex topology.
System architecture may be defined in several ways. One possible definition involves verification of how the operative elements of a system are arranged into blocks, and how these blocks interact. Two distinctive types of system architecture — integral (or closed) and modular (or open) — are generally recognised. In analysing the complexity involved in the development of a system such as the NSTG, it is immediately apparent that an integrated architecture is not the most convenient choice for construction and expansion of the system. Modifications to features and/or components are likely to occur regularly during the initial and development phases. Nevertheless, there should be little redesign of the whole system given technical difficulties and the high costs involved.
For complex systems such as the NSTG, one potential solution is to adopt, from the early stages of development, a modular architecture approach. In a modular-architecture system, each module may be designed practically independently from each other, which allows changes to be made to one module without affecting the others. Therefore, it becomes important to be able to clearly distinguish the objectives and primary functions of each system’s modules and the possible interactions between them. The task of establishing the modules’ functionalities inside the system can be accomplished through design hierarchy and standardisation.
Design hierarchy & standardisation
Design hierarchy and standardisation are two important concepts for complex systems such as the NSTG, since more than one stakeholder will be involved and, indeed, necessary for funding and development of the entire system.
In the modularisation process of a complex system, the first task is to establish which are the parts that can be considered modules or subsystems. For instance, in an offshore transmission grid, wind farms, HVDC converter stations, DC transmission cables and potential protective systems naturally constitute the basic building blocks or modules involved in the installation.
In a modular system, the design engineers are the ones responsible for establishing a set of design rules, which account for:
- System components: What are the system’s modules and their roles;
- Interfaces: How modules inside the system are connected to and interact with each other;
- Test procedures: Set the performance levels of a certain subsystem and allow for comparison.
Therefore, in a modular system, the design hierarchy is composed of levels. The global design rules are set at the highest level, next are the modules’ interfaces, system integration and communication. Finally, there are the design parameters that concern only modules.
Offshore DC networks will require rules — similar to how AC grids operate with regard to the transmission system operator grid code. During early development stages the design rules might appear simple or incomplete. However, as the characteristics of the NSTG and the modules inside it become better understood, the design rules will also tend to be developed further.
The NSTG global design rules, or the design parameters in the top level, must be established first due to the fact that they directly affect all the modules that are part of the system. Examples of global design rules inside the NSTG could include, but are not limited to, the DC voltage level — nominal, steady-state and transient range — in the offshore grid, the size and topology of each HVDC station, multi-terminal DC protection philosophy, multi-terminal DC control and the power transfer capability of the DC cables.
The crucial point is that changes made to the global design rules will have large implications on all system modules, and are thus expected to be expensive and difficult to perform. In comparison, modifications to features inside a module in the lower levels have limited extension and should be easier and cheaper to perform. For instance, changing the DC voltage level of the multi-terminal DC (MTDC) system would be one of these far reaching modifications that are bound to be costly and technically difficult. Thus, once the DC level inside the NSTG system is established, there will be very little room to change it. At the last IEEE European Power Electronics Conference, in Birmingham, an article was presented showing that Siemens has been awarded three projects for the connection of offshore wind farms in German waters. However, all three projects will have different DC voltage levels — Borwin2 (±300 kV), HelWin1 (±250 kV) and SylWin1 (±300 kV) — even though they are located relatively close to one another. This lack of standardisation will make the interconnection of these offshore wind farms to a future North Sea Transnational Grid much more challenging and expensive.
System engineers must carefully establish and take global design rules into consideration before systems like the North Sea Transnational Grid can be developed and built. Furthermore, the proper development of global design rules can lead to DC grid standards which could allow for costs reduction by having a single common design, allowing systems to be built incrementally and by different equipment suppliers, thus supporting incremental investment plans.
In this way, large pan-European DC grids would be developed “organically” — first by the construction of a few small independent DC grids (four to six terminals) that, in a later stage, would be combined to form a larger offshore network with a much more complex topology, such as a meshed multi-terminal DC network.
Multi-terminal DC networks
Multi-terminal HVDC (MTDC) transmission systems are characterised by more than two HVDC converter stations somehow interconnected on the DC side of the transmission system — that is, multiple HVDC converters linking different AC power networks through a DC transmission network.
The MTDC configurations can be classified according to the type of HVDC technology implemented at the converter stations: line-commutated current-source converters (CSC) or forced-commutated voltage-source converters (VSC):
- CSC-MTDC: all the converter stations use the line commutated current-source converter HVDC technology;
- VSC-MTDC: all the converter stations use the forced-commutated voltage-source converter HVDC technology;
- Hybrid-MTDC: when both HVDC technologies (CSC and VSC) are used together.
Even though there are well above 100 HVDC-based transmission systems installed around the world, only three currently have more than two terminals: The Hydro-Québec/New England scheme in Canada; the SACOI-2 scheme between Italy and France; and a back-to-back scheme using VSC technology at the Shin-Shimano substation in Japan.
Due to their physical characteristics, VSC-HVDC transmission systems will most probably be the technology initially chosen to connect offshore wind farms, since in offshore projects the converter station’s footprint is a critical variable. State-of-the-art voltage-source converters for transmission purposes make use of modulation schemes or multilevel topologies, which allow them to have smaller space requirements than current-source converters. In addition, in VSC-HVDC stations there is no need for bulky special transformers able to block DC voltages, as is the case for CSC-HVDC stations.
Modularity and standardisation will most definitely play very important roles in the construction and development of multi-terminal DC networks for integration of large-scale offshore wind energy. To allow offshore networks to grow organically with time, the global design rules of such systems will have to be carefully discussed with all involved stakeholders in order to avoid expensive and difficult changes during later development stages.
DC load flow analyses performed for a possible NSTG layout involving the countries with the highest expected offshore installed capacity have shown that a control strategy where more than one node is controlling the DC voltage inside the MTDC network — thus working as slack nodes — is superior when compared to a strategy in which only one node is given that task.
RodrigoTeixeira Pinto and P. Bauer are researchers in the Power Processing Group at the Technical University of Delft in The Netherlands.