Just outside London, a test is underway to see if it’s possible to power commuter trains by solar power.
One of the biggest benefits would be efficiency, as a conventional solar supply pumps into the National Grid, which converts the solar array’s DC electricity into its AC requirements. When the National Grid supplies it to the railways to power their trains, it’s turned back into DC supply — and that conversion process loses around 4-5% of the electricity originally generated.
Cut the losses by pumping DC solar power directly into DC absorbing railways and you lower the cost of providing the same amount of energy – making solar power more affordable. The really big problem is supply and demand rarely align with each other.
Not just that solar power is variable thanks to the vagaries of the weather, as it’s often far more stable than you might expect from the British climate, but that the demand from the railways varies wildly depending on how often trains pass through the area being supplied by the solar panels.
The whole project actually started down on the London-Brighton railway, in the town of Balcombe, where an opportunity to install a large solar array hit an expensive problem. It could have generated more power than the local town would need, but the National Grid in the area was already ‘full’ and couldn’t take any more supply without an expensive upgrade.
How do they propose to alter the voltage the generated DC power, the only economical way is to convert it back to AC (although at a much higher frequency than the national grid uses) step it up or down using a transformer, and then rectify it back to DC – which is how a switch mode power supply works.
So to say there is not going to be any use of AC I suspect is slightly false, however I agree it’s likely to be much more efficient than converting it at 50Hz.
JC: My reading of the article suggests that they are proposing that at least some of the power would be used directly as DC, since someone has noticed that the DC voltage produced by some of these solar arrays is “approximately” just what the railway needs. There is a lot of handwaving about supercapacitors and batteries (of course) because the power demands from minute to minute will not match up properly. That is why we hav a national grid. But it’s a start-up firm that wants investors, and they have spotted this apparent gap in the market. They may attract some investors, whether or not the idea actually works.
“It’s often far more stable than you’d expect…” Only when aggregated over a sufficient geographical area. At any specific location, on a typical British partly cloudy day in the brighter half of the year, solar radiation zips up and down wildly as the clouds scud over.
Here is a weather station near me yesterday, and you see how it solar irradiation oscillates wildly once clouds came over in the late morning, oscillating from around 900 W/m2 in the sun to around 250 W/m2 when cloud passes over.
https://www.wunderground.com/dashboard/pws/IBUCKING4/graph/2019-08-22/2019-08-22/daily
Also, even in August on a clear sunny morning, the angle of the sun means there’s not much available for the rush hour.
Once you start putting batteries in to smooth it, it starts getting expensive. Take what you need when you can, but feed the rest into the grid and let them do the smoothing for you. You might get a 5% conversion loss, but that is cheap in comparison to a battery.
Consistency is a problem both of supply and demand at a small scale. In your house, you might look at your smart meter and think you have been using a fairly consistent, let us say, 400W over a few hours. But that is smoothed over half hour intervals. When you turned your electric kettle on for 2 minutes, that was 2500W you added for that 2 mins. Thermostatically controlled appliances like fridges, freezers, etc trip in and out. Meanwhile your solar panel was wanging up and down as clouds scud over. If you are trying to run your house off the solar panel, the match between supply and demand often be poor. If you are thinking of having a battery to smooth it, it would be a much cheaper for the electricity company to put one at the substation and thus smooth out the swings across the (say) 200 properties in the substation circuit – that way you need much less battery capacity per property to cope with demand swings and solar swings.