Fuel cell for prime power in data-centres

Topic description

Specific Challenge:

Pushed by the major trend of digitalization and the internet of things, the market demands more and more highly reliable, continuous power supply (prime power) for data-centres. It is expected that data-centres will use up to 4% of world’s energy consumption in the next few years. Most of the visible hyper scale data-centres (100 MW) are nowadays built in cold remote areas in proximity of renewable energy sources, primary wind and hydro power and where free cooling is available.

The more numerous but more discrete smaller “edge data centres”, situated close to the end-user, have also experienced a strong growth in cities and urban areas. Much smaller in scale, they offer an increased service level to the end-users expecting very fast data response time where every millisecond counts in customer experience. The power density of the data-servers has more than doubled over the last few years, allowing to increase the data treatment capacity within the same space, while the availability of electric power is becoming increasingly a constraint in those areas. Often it is simply not possible to increase the power supply in cities like London and Frankfurt, as the power distribution infrastructure is already extensively loaded.

The challenge is to overcome the constraints in the existing electrical infrastructure without increasing local emissions using solutions that are efficient from an energy and an economic point of view. This creates an opportunity for distributed generation. In addition, city centres often practice stringent policies to reduce and avoid emissions like NOx, SOx, particulate matter and noise, to protect the health of their inhabitants. This limits the use of power generation using combustion engines. Fuel cell generators can adequately address all these challenges as they can provide high power density within buildings or on roof tops and highly reliable power supply at acceptable cost levels.

An appropriate overall data-centre architecture is required to address all aspects simultaneously, particularly the redundancy to provide high reliability and availability.

Furthermore, edge data-centres are often located in large commercial buildings like banks and insurance buildings, hotels and hospitals, offering the opportunity for the use of the by-product heat. This can be realised either within the building itself or by providing the heat to a nearby district heating system. An appropriate thermal design provides the opportunity to reach higher total efficiencies relative to other conventional technologies at the same scale.

The project aims at demonstrating a highly reliable power supply to data-centres within urban areas where grid power supply is limited due to infrastructure constraints and where combustion-based combined heat and power (CHP) is not admissible due to emissions regulation.

The project will develop a building integrated solution using fuel cells that are adapted and suited to the requirements for strong load modulation, electrical integration/connection to the data centre and heat recovery for use in the building or a district heating network.

The solutions developed in the project should:

• Provide an appropriate data-centre electricity supply architecture based on fuel cells modules to provide a 99.999% availability to the data-centre;
• Supply at least 50 kWe of power to the data centre. The architecture developed should be modular, easily scalable to other system capacities and able to provide the five 9 (99.999%) availability;
• Demonstrate the solution developed in an operational environment (real data centre) for at least 8,000 hours;
• Provide overall solutions that cope with the strongly varying load of data-centres;
• Address service and maintenance requirements of both fuel cell modules and data servers in a coordinated manner to optimise the overall operating costs of the data-centre. This includes minimisation of system footprint and accessibility for maintenance;
• Consider the suitability of using the heat generated from the fuel cell and data-centre server either to supply the building needs or to a nearby district heating system.

Projects should identify and/or develop business models aiming to foster the replication of the solutions developed in the project in other data centres.

The consortium is expected to include at least a system supplier based in the EU or H2020 Associated Country with proven track record of units running in the field and a data-centre infrastructure provider. Activities should build on the experiences and achievements of earlier projects and bring further cost reductions as a means to improve and strengthen the competitiveness of the EU fuel cell supplier industry.

TRL at start of the project: 5 (for the electrical architecture) to 6 (for the fuel cell) and TRL at the end of the project: 7.

Any safety-related event that may occur during execution of the project shall be reported to the European Commission's Joint Research Centre (JRC) dedicated mailbox JRC-PTT-H2SAFETY@ec.europa.eu , which manages the European hydrogen safety reference database, HIAD and the Hydrogen Event and Lessons LEarNed database, HELLEN. A draft safety plan at project level should be provided in the proposal and further updated during project implementation (deliverable to be reviewed by the European Hydrogen Safety Panel (EHSP)).

Activities developing test protocols and procedures for the performance and durability assessment of fuel cell components should foresee a collaboration mechanism with JRC (see section 3.2.B "Collaboration with JRC"), in order to support EU-wide harmonisation. Test activities should adopt the already published FCH 2 JU harmonized testing protocols to benchmark performance and quantify progress at programme level.

The maximum FCH 2 JU contribution that may be requested is EUR 2.5 million. This is an eligibility criterion – proposals requesting FCH 2 JU contribution above this amount will not be evaluated.

Expected duration: 3 years

• Provision of solutions to supply the energy requirements of data centres located in urban areas that are reliable, modular, highly efficient and compatible with the requirements of the most stringent local air quality regulations;
• Cost-effective solution to overcome constraints in the existing electrical infrastructure of city centres and urban areas without increasing local emissions;
• Opening of new markets for EU suppliers including export opportunities for overseas markets. The project should provide business models for fuel cell and data centre providers planning to roll-out and implement the solutions developed in the project at scale;
• After an initial deployment phase in electricity constraint areas, it is expected that the use cases will expand more generally throughout urban areas and that the resulting products are compatible with increasing amounts of hydrogen within the existing gas infrastructure.

The solutions developed should achieve the following KPIs:

• Fuel cell system cost <€5,500/kWe;
• Fuel cell system volume <225 l/kWe;
• Single system availability of at least 98%;
• Overall availability of power supply to data-centre at least 99.999%;
• Electrical efficiency > 50% (methane LHV) and >45% (H2 LHV).

The conditions related to this topic are provided in the chapter 3.3 of the FCH2 JU 2020 Annual Work Plan and in the General Annexes to the Horizon 2020 Work Programme 2018– 2020 which apply mutatis mutandis.