Fuel Cell Research Collaboration with UTBM

02. April 2014

Since 2007, a research team at ETH Zurich, from which Celeroton AG spun out of, has been part of a scientific collaboration in the area of fuel cells with the University of Technology of Belfort-Montbéliard (UTBM). This collaboration has lead to the development of a 10 kW fuel cell model coupled with a Celeroton air compressor, that forms part of a Hardware-in-the-Loop test bench. Here we present an interview with two of UTBM’s scientists, Prof. Miraoui and Prof. Gao.

PEM fuel cell systems are often used for mobile and decentralized energy production and are built into cars, airplanes, trolley buses, combined heat and power (CHP) plants or emergency power units. All these applications have in common the difficult requirements of high efficiency and compactness. When every gram and every cm3 counts, then high-speed turbo compressors are perfectly suited to tackle these challenges. This has also been identified in a fuel cell research publication written by UTBM experts. This was the trigger for the scientific collaboration between UTBM and the team of the Power Electronics Systems Laboratory of ETH Zurich, and Celeroton AG respectively. The goal of the collaboration is to explore the interaction of ultra-high-speed turbo compressors with fuel cell systems with the aim to show the feasibility and advantages of such systems and to optimize control techniques (see here, here and here). The following interview with the two supervising professors of UTBM gives our readers an insight in this interesting research collaboration.


How long has UTBM been involved in fuel cell research?

Our research in the fuel cell field started in 2004.

What are the current challenges in fuel cell research?
While fuel cell technology will enable huge improvements, for example in energy efficiency there are still many challenges to be tackled. Currently, we are trying to reach cost reductions for the whole system. Also the performance and power density still have to be improved (with an objective on 1 W/cm2). Furthermore, our scientists are working on robustness, reliability, safety and life time issues. Our goal is to achieve a life time of 5,000 hours in general automotive applications and 10,000 up to 20,000 hours in public transport.

Besides these general difficulties we are facing some more specific challenges: In automotive applications for example the starting of the fuel cell system at -20° C needs to become possible.

Even though fuel cells could be a huge progress in green economy there still are some challenges from the ecological point of view as well: Production, distribution and hydrogen storage need to be monitored to reduce the impact on the environment, e.g. through the choice of a better production chain of hydrogen. We also still have to demonstrate the recyclability of our components.

How do you expect the market for fuel cells to evolve in the upcoming years?
We will see many automotive demonstrators, especially for public transport. But as the large list of improvements to be implemented (see above) shows mass production definitely is not for tomorrow, but maybe around 2020, as expected by the European Union council. In parallel the market of stationary applications should progress.

But we have to keep in mind that everything hinges on the hydrogen economy. We must review the overall efficiency from well to wheel or well to socket. We think that fuel cell systems and the hydrogen energy sector make sense only if the hydrogen is produced with renewable energy and with the highest efficiency to ensure the least possible environmental impacts.

What challenges are you facing on the compressor side and concerning pressure control?

The major tasks concerning the air management subsystem are:

  • Air supply: The air management system has to supply sufficient reactant flow to keep the desired oxygen excess ratio over the full power range. Insufficient air flow may damage the stack in severe cases. The air normally is circulated by an air compressor or blower, so the air compressor has to have good controllability.
  • Air cleaning: Any particle or chemical substance such as carbon monoxide can be harmful for the catalyst and the membrane. Therefore, the air has to be filtered before going into the stack. This means the compressor should ideally work oil-free.
  • Air pressurization: The air supplied to the fuel cell is generally pressurized from slightly above atmospheric pressure to 2.5 bar, depending on the stack requirement. Pressurizing air to the fuel cell implies a high efficiency and gives better water balance characteristics. Higher air pressure implies a more compact stack and higher power density whereas, compared with low pressure fuel cells, more parasitic power losses are produced by the air compressor.
  • Air humidification: The polymer membrane has to be maintained in a fully hydrated state to have optimal working conditions. The air management subsystem usually includes a humidification subsystem to fulfill this target.

The air compressor, the key component of the air management system, is the largest parasitic power consumption device in the fuel cell system. In severe cases it consumes up to 20 % of the generated power. The compressor efficiency including motor, power converter and controller depends on the compressor type, pressure level and speed. However, in all cases the compressor should be as efficient as possible.

What advantages in fuel cell systems do you see with high-speed turbo compressors in comparison to other compressor technologies?
The centrifugal technology has major advantages concerning compactness, low noise and high efficiency which make it more suitable for automotive applications than other compressor technologies. For automotive compactness requirements ultra-high-speed compressors seem to be the best solution. The advantage of this high rotational speed is a decrease of the impeller radius and therefore an increase in power density in turbo machinery. Also the motor power density is roughly proportional to the speed. In summary the centrifugal compressor exhibits benefits, especially for automotive fuel cell applications.

However, control problems such as pressure and surge control introduced by the centrifugal compressor still are to be solved. Moreover, the fuel cell characteristics and load situations also have to be considered in controller development. Therefore, new control technologies are needed to cope with the control problems to satisfy the demands from both the compressor and fuel cell. In fact many control methods have been proposed recently to deal with the air management problem of fuel cell systems.

How have you experienced the research collaboration with Celeroton so far?
Our collaboration with Celeroton began in 2009. We quickly identified the potential of high-speed compressors after our first discussion. The professionalism of the research and development team has enabled us to obtain a first prototype in order to work on the management of the air loop quickly.

Today the first part of the work has been finalized. The proposed controllers have been implemented and then compared by a hardware in-the-loop simulation. It shows that the dynamic decoupling controller (DDC) can reduce interaction between mass flow and pressure control effectively. In the end a 10 kW fuel cell model was developed for further validation of the controller. According to the load variations the compressor is able to supply variable mass flow while maintaining the pressure constant.

You also do research for industries that plan to use fuel cells in the future. What is the roadmap from research to serial production? What are the biggest challenges?
Fuel cell industries hope to follow the roadmap given by the US Department of Energy (DOE) (2020: lifetime 5,000 h, stack cost 43 €/kW for 500,000 vehicles). We think this might be possible by innovation in design and materials thanks to innovation in production technology and economies of scale.

Besides that we think that the real challenge is hydrogen. If we cannot refuel cars or static fuel cell generators, it is not even necessary to develop fuel cell systems. Today, more than 95 % of the hydrogen is produced with fossil energy (gas, petrol and coal). We do not solve the CO2 problem if we use these sources. The only way is to use renewable energy sources or, if necessary, next generation nuclear plants to produce hydrogen in order to supply fuel cell systems. It is a global task.

Thank you very much for this interview.



Abdellativ MiraouiProf. Abdellatif Miraoui was born in Morocco in 1962. He received the M.Sc. degree from Haute Alsace University, Colmar-Mulhouse, France, in 1988, and the Ph.D. degree and the Habilitation to Supervise Research from the University of Franche-Comte, France, in 1992 and 1999 respectively.

He is the President of Cadi Ayyad University, Marrakech, Morocco. He has been a Full Professor of electrical engineering (electrical machines and energy) at the University of Technology of Belfort-Montbéliard (UTBM), Belfort, France, since 2000.

His special interests include fuel cell energy, energy management in transportation, and design and optimization of electrical propulsions/tractions.


Fei GaoProf. Fei Gao (S’09-M’11) received the Master’s and Doctor’s degree in electrical engineering from the UTBM in 2007 and 2010 respectively.
Since 2011, he has been an Associate Professor at the UTBM. His main research interests include fuel cells and their applications in transportation, multiphysics modeling and real time applications. He is also the chairman of fuel cell modeling axis of the Federation for Fuel Cell Research CNRS FCLAB in France, and the head of the energy production division of the energy and environment department. He serves as an editor for the IEEE Transportation Electrification Newsletter and the secretary of the IEEE IES Technical Committee on Automotive Technologies (IEEE TCAT).

< back


High-speed magnetic bearing chopper for infrared nanospectroscopy

Infrared (IR) nanospectroscopy is a non-destructive method for the analysis of molecular structures below the ...

> more

Design of ultra-high speed optical beam chopper rotors

High-speed beam choppers usually require a fast circumferential and rotational speed and vacuum operation, for which ...

> more