So first off my opinion
I really like pumped hydro.
Possibly because it’s such a gloriously mechanical answer to an electrical problem.
We’ve got too much electricity.
Pump water uphill.
Later we need electricity.
Let the water come back down.
There is something wonderfully simple about that.
Obviously the engineering required to actually do it isn’t simple at all.
But the principle is.
So it’s basically a battery?
Pretty much.
Except instead of storing energy chemically, we’re storing it as gravitational potential energy.
You normally have two reservoirs at different elevations.
When electricity is plentiful, pumps move water from the lower reservoir to the upper reservoir.
When electricity is needed, the water flows back downhill through turbines and generates electricity.
So:
Electricity → Pump → Water uphill → Wait → Water downhill → Turbine → Electricity
Again, there are losses.
We’re not getting all the electricity back.
But we’ve moved energy from one point in time to another.
This technology is not new
That’s another thing I like about pumped hydro.
Energy storage sometimes gets discussed as if storage arrived alongside lithium-ion batteries.
It didn’t.
Pumped storage has been around for decades.
Dinorwig in Wales is probably the UK example I’ve heard about most often.
Huge quantities of water can be released through turbines when the electricity system needs additional power.
So while the technology is old, the problem it’s solving arguably becomes more interesting as we add more variable renewable generation.
Why not build loads of them?
Geography.
There’s always something!
You need suitable terrain.
You need somewhere for the reservoirs.
You need a useful height difference.
You need enormous civil engineering works.
Planning and environmental effects matter.
And unlike a battery container that can potentially be installed on a suitable industrial site, you can’t manufacture a mountain and deliver it on the back of a lorry.
Well, not economically anyway.
So pumped hydro is potentially brilliant where the geography allows it.
That’s quite a limitation.
But once you’ve built it…
This is where pumped hydro seems very attractive.
These can be long-lived assets.
We’re talking about major civil infrastructure rather than battery cells that progressively degrade through cycling.
Turbines, pumps and other equipment obviously require maintenance and replacement.
But the reservoir doesn’t suddenly forget how to be a reservoir after ten years.
That potentially gives pumped hydro a very different asset lifecycle from electrochemical batteries.
What would I want from the data?
Quite a lot.
Reservoir levels.
Available stored energy.
Water flow.
Pump performance.
Turbine performance.
Vibration.
Bearing temperatures.
Electricity prices.
Grid demand.
Weather.
Expected renewable generation.
Maintenance condition.
Suddenly that simple idea of “water goes uphill, water comes downhill” becomes another optimisation problem.
When should we pump?
When should we generate?
How much capacity should we hold back?
What’s the condition of the machinery?
How much energy is actually available right now?
That’s where the data engineer turns up.
The physical world matters
This is another good example of why I think understanding the asset matters.
A database might contain:
reservoir_level = 92%
Useful.
But 92% of what?
How does that translate into MWh?
Does the usable energy change depending on the hydraulic head?
Are there environmental limits on how far the reservoir can be drawn down?
Is all of that 92% actually available to generation?
The number only becomes properly useful when we understand the physical system producing it.
So where is my head at?
Pumped hydro feels like one of those technologies that doesn’t need to be fashionable.
It just needs to work.
It’s large.
It’s physical.
It’s geographically constrained.
And it can store a serious amount of energy.
I don’t think it competes directly with every battery installation because again, different storage technologies solve different problems.
But if you’ve got the right geography, moving water uphill when electricity is plentiful and letting gravity return the favour later seems remarkably sensible.
Sometimes the giant battery really is just a mountain.
That’s where my head is at the moment.
Gareth Winterman