Imperial College researchers and others are envisaging a future of trains powered by photovoltaic energy
Running electric railways directly from sunlight may be a cost-effective way of expanding both photovoltaic energy generation and railway electrification in the UK and around the world, according to a report from Imperial College. Originally triggered by a movement that sprang from an anti-fracking protest, the idea led to a feasibility study and may end up in pilot projects.
The report, Riding Sunbeams, proposes installing photovoltaic panels directly alongside railway lines and transmitting the electricity generated directly into the railway system as traction current, without first distributing it to the grid. This would take advantage of a coincidental match between the peak generating time for solar and a peak demand for traction current, but would bypass current problems in many areas of limited grid capacity. “Many railway lines run through areas with great potential for solar power but where existing electricity networks are hard to access,” explains Prof Tim Green, director of the Energy Futures Lab at Imperial College London. It would also potentially free up many sites for photovoltaic development.
The Imperial project began when the residents of Balcombe, a village in West Sussex, opposed proposals for test drilling to establish whether the area might be suitable for shale gas fracking. The protest inspired the establishment of a co-operative called Repower Balcombe, which aimed to generate the village’s entire electricity demand via renewables. One stumbling block to this goal was that the local grid did not have the capacity to accept any more solar energy without costly reinforcement. Prof Green, who lives in Balcombe, decided to look into the possibility of using solar power directly on train lines. “Quite quickly you realise that the answer might be more than the solution to one village’s problem and something to unlock untapped solar resource on a much wider scale,” he said in the foreword to the report.
Working with 10:10 Climate Action, a charity that runs projects aimed at helping communities to achieve cuts in their carbon emissions, Green and the Energy Futures Lab started to look at the subject in engineering terms.
The report was written by Leo Murray, director of strategy at 10:10 Climate Action, and Nathaniel Botterell, a post-doctorate research associate in the control and power research group in Imperial’s electrical and electronic engineering department. It begins by setting out some of the problems. The capacity constraints on the grid mean that it is not affordable to connect new renewable generating capacity across whole regions of the UK. Moreover, the withdrawal of subsidies for solar photovoltaic means the only commercially viable developments are those with an on-site final customer. In the meantime, on the railways, demand for traction power is increasing.
Another happy coincidence comes into play here. Solar PV arrays typically output DC power at between 600 and 800V. Electric rail, meanwhile, typically operates at 750V. This, the report says, means that the cost of power electronics needed to connect solar generation to DC traction networks should be competitive with grid connection costs.
For this to be most effective, it needs to be targeted at rail systems that use DC, and, unfortunately, about two-thirds of the UK rail system does not. Most of the network, is powered by AC running through overhead catenary cables, which is safer and better for high-speed and long distances. But in urban areas, overhead cables are much more difficult to install, because train lines run through tunnels, and under bridges and roads; in these areas, traction supply is instead conveyed through a DC third rail. “I think that focusing on the basics of integrating distributed energy generators into a railway’s system with the third-rail network brings a lot of benefits,” Prof Green said.
The most obvious benefit is that there’s no need to convert the DC output of the solar panels into AC, which inevitably leads to losses. But there are other advantages. In areas with DC power supply, there tend to be substations or track paralleling huts around every 3km along the route. By contrast, in AC-supplied areas substations can be up to 80km apart. The closer spacing more convenient for injection of DC supply; moreover, private wire supply from solar farms is effective over distances of up to 2km, so most lineside land would be within reach of a traction network connection.
The report notes that connection to AC systems is not impossible, particularly where connections can be integrated during new electrification works that are in progress or planned on large sections of the UK network. “This opportunity should be factored into future UK electrification rail education planning,” the report recommends. Each AC traction substation covers a large length of track, which could – with the required planning and infrastructure – be supplied by large solar farms or other renewable resources, such as wind farms or hydroelectricity. Another possibility might be to connect solar PV to the auto transformer sites located around every 10km along electrified routes running off AC, it suggests; these sites are used to boost the AC voltage.
Even using the simpler option of concentrating on the third rail sections, there are still issues that need to be determined. For example, the third rail is not only used to supply power; in most rail networks, it is also used to transmit information as part of the signalling system. There are also issues around managing how and when the solar power is sent to the third rail. Such technical aspects will be investigated by a collaboration between Imperial College and Turbo Power Systems (TPS,) which specialises in the distribution and management of power in the railway sector. “We couldn’t have better partners to work out how you can integrate solar power with our trains,” said Leo Murray.
The report concentrates on the advantages that direct connection of solar PV to rail would have in the south of England, as statistically this is the region that receives the most sunshine. The report’s authors claim that 15 per cent of commuter routes in Kent, Sussex and Wessex (sic) could be powered by solar PV. The London Underground could also benefit; much of the network in fact runs above ground, and 10:10 Climate Action has found 50 possible locations for solar installations, including derelict land, train depot rooftops, station car parks and even the possibility of floating solar on reservoirs. These sites, they claim, could provide enough electricity to power 6 per cent of the tube network.
But it’s not just about southern England. Another possibility would be the commuter network in Liverpool, while smaller city metros or trams, such as Manchester Metro link or the tram systems in Sheffield or Nottingham, could also install trackside PV on brownfield sites, the report suggests.
Even more potential exists around the world. Equatorial regions receive more and stronger sunlight than northern Europe, and mega cities in Asia, Latin America and the Middle East are all good candidates for solar rail. In India, for example, the government has a very ambitious target for installing PV capacity, demanding 100GW of solar capacity by 2022. The country has 15,000 miles of electrified railway with a target of some 2,000km of new electrified track per year to be built.
Ninety per cent of the Delhi Metro is currently powered by solar via a grid connection, and 10:10 Climate Action claims there is great potential for direct power. Closer to home, Spain has 4,000 miles of electrified railway, 50 per cent more direct sunlight than London, and in Barcelona, for example, planning regulations already stipulate the large buildings must have their own solar power, so there is clearly an appetite for this form of renewably powered transport.
£22bn for CCUS and hydrogen in northern clusters
It would make more sense to move the east coast project from Teesside down to South Humberside where there is existing demand for hydrogen (Phillips...