The model uses source data from real-world industry records of UK wind and solar outputs over a 10-year time period.
If supply fails to meet demand at any moment, a blackout crashes the UK grid
UK electricity data records how much power wind, coal, gas, etc generated every 30 minutes since 2015, recording supply and demand across the ever-changing weather and seasons. That’s 175,000 data points. To model a clean grid, the first step is to eliminate all electricity generated from coal/oil/gas/biomass, then scale up wind and solar data to try to take its place:
| 
            | 
                 
             | 
(Graph derived from half-hourly figures from National Grid/Elexon 
         and Sheffield Solar) 
      
The best options to store electricity include batteries, pumped storage hydro and hydrogen. Any stockpiles of surplus energy from windy periods can feed back into the grid when there is not enough wind or solar.
| STORAGE CAPACITY - GWh - £ Bn | |
| STORAGE POWER - GW | 
Electrify Road Transport?
Electrify heating (phase out Natural Gas)?
Adds massive load to grid, mostly during winter. The only way to reduce this is to lower demand through energy efficiency:
Insulation 1 in of homes £ Bn
Heatpumps 2 in of homes £ Bn
| STORAGE CAPACITY - GWh - £ Bn | |
| STORAGE POWER - GW | 
   
The next steps look at different kinds of storage, and testing over a much longer period
 This model uses three main
                              
                      storage technologies  
 
Batteries - Very expensive, cost is determined by the amount of energy stored, but can be up to 90% efficient 
Pumped Storage Hydro - also expensive, costs depend on size of reservoir (amount of energy) and the power of turbines. 85% efficient
Hydrogen - electrolysers to make hydrogen and CCGT (turbines to turn it back into electricity) are both relatively cheap, and the amount of energy stored could be almost limitless. However, at least 60% of the precious electricity is wasted during the processes of converting to hydrogen and back to electricity.
 
This simulation can cover a time period up to 1 year
The basic looks at a whole year's data, but there are actually 10 years of data available (from 2015 onwards) detailing over 140,000 data periods soon to be developed.
Interconnectors linking the UK to other national grids on the continent currently make up a significant share of the UK's electricity. However, if all those other grids also rely on wind and solar, they may not be available to supply power just when the UK needs it. For that reason, this model excludes contributions from interconnectors.
Shifting power-hungry devices (car charging, heat pumps, industrial processes) to off-peak hours will undoubtedly play an important role in smoothing out the peaks and troughs in demand for electricity (in a very similar way to batteries), but it doesn't address the bigger problem of long term energy storage.
Studies have proposed that tidal barrages could supply up to 7GW of electricity, reliably, for many decades. Similarly tidal flow schemes promise predictable reliable power akin to baseload. But the numbers are too small to contribute significantly to the UK's power supply.
The promise that small modular reactors will be cheaper than large reactors looks like magical thinking to me. I'm not an expert, but the potential savings from the 'modular' construction approach seem less significant than the economies of scale that a single large reactor can deliver. Take the experience of Nu-Scale Power. Its costs ballooned three-fold, even after the project was scaled down in size before being cancelled at huge costs to Idaho consumers. That SMR suffered exactly the same massive cost over-runs as large projects such as Hinkley Point, Flamanville, Vogtle, etc.