Demand performance

University project aims to optimise efficiency of fuel cells in future UAVs


Improving the energy use and efficiency of fuel cells powering future UAVs and stealth ground vehicles using computational intelligent modelling is the aim of researchers at Cranfield University.

The research will take the form of producing algorithms that will drive the cells to their optimum operating performance.

Robotic systems such as those used in UAVs are required to perform complex trajectories with numerous transitory phases. This means that the load on the fuel cell is constantly changing, and it may not need to produce as much power at certain points, such as when the vehicle is moving slowly or is cruising.

At present, fuel cells are configured to operate at their optimum temperature as soon as possible after start-up, meaning that their fuel use and power output is constant — regardless of demand.

This project aims to mathematically model the cell control system and optimise its performance at various loads, meaning that the amount of power generated at any point is sufficient to meet the demand required. The resulting solutions are expected to improve endurance and robustness of the fuel cell system.

The team will also look at how to shorten the time between start-up and optimum efficiency being reached to improve the vehicle’s potential where rapid deployment is vital.

Although the initial target for the system will be the aerospace market, the researchers are also looking at its potential for use on hi-tech military ground vehicles, as the challenges that need to be met are very similar. Such transporters are being investigated for use as stealth vehicles as they operate almost noiselessly.

The team is also looking to attract an industrial partner which perhaps has an interest in either or both markets.

Potentially, there could also be applications in the leisure industry. At present, fuel cells have been marketed as an environmentally friendly way of powering lighting, heating and cooking facilities in caravans. But the cells may cost thousands of pounds and so have not proved popular.

By improving efficiency, it may be possible to reduce their cost, allowing smaller units to be used.

‘We are looking at taking a greener approach to fuel cells, which currently have technological limitations in this respect,’ said Dr John Economou of the university’s Guidance and Control Group.

‘We will be looking at how to match the requirements of the robotic systems with the performance of the fuel cells.

‘For instance, if the system needs to travel from A to B, we must discover the best and most economical way to do this. The research consists of a complex problem with what should be a relatively simple solution.’

To design a highly-efficient way to operate a fuel cell for power, the group is looking at the theoretical performance abilities of energy generation systems, based on the technology that is in existence today.

The plan is to develop and produce complex models of this, to see which system shows the most promise in terms of fuel efficiency.

‘We are exploring the possibilities of what is out there,’ explained Economou. ‘The idea is to engage with manufacturers of the fuel cells and refine their performance.

‘By adding something to devices that simply make improvements to existing systems, manufacturers will not have to change their system’s construction or design.’

This will make any advances cheap to implement, as users could also retrofit the Cranfield system.

To help with the research, Economou is looking for a doctoral student with a background in electrical and electronic engineering, automatic and control engineering or mathematics.

‘It may be an ideal opportunity for someone from industry to gain experience,’ he said.