Monday 14 April 2014

The 100% Solar Powered National Grid?

Image: Viridian Solar

Is your Car the Star?

I was intrigued by a recent article in the Telegraph (of all places)  in which the writer predicted that solar will dominate global energy production.

The article extrapolated the falling costs of solar photovoltaic generation in recent years and combined this with a whistle-stop tour of technical innovations in the sector - perovskite solar cells, flow battery technology  and robotic cleaning of panels in solar farms.

The article predicted the end of the fossil-fuel era, and concluded that wind turbines would at best be a regional niche technology because their costs were static by comparison with the enormous technological strides that will continue to be made in solar.

As is common in such articles, the intermittent availability of solar power was mentioned only in passing along with the solution – energy storage technologies.

The article made me wonder about energy storage.  How much energy storage would be needed if solar were to win the race for global dominance of energy production?  Is it a realistic prospect? 

Solar power production is intermittent in three ways.  First there is the unpredictability of cloud cover and light levels during the daytime.  Second there is the more predictable effect of the sun travelling across the sky during the day and disappearing at night.  Finally there are the seasonal variations in day length and clear skies from summer through to winter.

One Helluva Battery

Let’s do some simple analysis to try to quantify the challenge.  To keep things simple, lets take a look at the following question:

‘How much energy storage would be required to hold the electricity generated during daylight hours for use overnight?”

The graph shows the half-hourly electricity demand from the UK national grid for 23rd April 2013.  I chose a day that was pretty sunny, but in spring rather than summer or winter.  The total electricity consumption that day was around 826,000 MWh.

On the same day in April 2013, the 44.5 kWp south-facing PV array on the roof of Viridian Solar’s factory and headquarters in Cambridge produced 233 kWh, or 0.233 MWh.

If we were to scale up the PV array to produce the same number of units of electricity as the UK demand that day, it would be:

44.5 x 826,000 / 0.233 = 158,000 MWp or 158 GWp

To put this figure in context, the UK government central forecast predicts installation totalling 10 GWp within a decade in its solar strategy, however the UK installed 1.1GWp in the first quarter of 2013.

If progress continues in the reduction of costs of solar modules, then it’s not so impossible to imagine that the deployment of this level of capacity over the coming decades might happen.

I've scaled up the output curve from the Viridian Solar PV array for that day as if it were 158 GWp and added it to the graph to show the timing of solar power generation compared to the timing of electricity demand.

The energy storage required for the solar output to keep the country going through the evening, night and until the solar output exceeded the demand the next morning would be 428,000 MWh or 428 GWh.

By using the current level of electricity demand we’ve ignored the possibility that energy efficiency gains might reduce total electricity demand significantly and thus drive down the amount of energy storage we would need. 

We've also ignored the new technologies emerging that intelligently move the timing of electricity use for applications that are not time critical (such as running air conditioning and refrigeration).  Such technology could change the demand curve to better fit the solar output, also reducing the amount of storage required.

However, some 40% of our energy use today is for heating and currently most of this does not use electricity but instead burns gas and oil.  This massive demand for energy will also need to be de-carbonised and it seems likely that it will require electricity to operate heat pumps.  Clearly heating demand can be reduced with better thermal insulation and solar heating panels on our buildings, but it seems that we’ll be doing well if we can hold our electricity demand flat relative to current levels.

So what does 428,000 MWh mean in the real world?

Consider Dinorwig pumped storage power station.  This massive civil engineering endeavour in Snowdonia National Park comprises 16km of tunnels, 1 million tonnes of concrete and 4,500 tonnes of steel. The turbine hall is Europe’s largest man-made cavern.  Cheap night-time electricity powers reversible turbines to pump water from a lower reservoir at the bottom of the mountain to a reservoir higher up the mountain.  At times of peak demand the water is allowed to run back down through the turbines and generates electricity.

It cost £425m, took ten years to build and was completed in 1984, that works out to around £2bn in today’s money.

It has a power output of 1,728 MW for six hours, a total of 8,640 MWh.

We’d need to build another 48 of these babies to store solar PV electricity through a 24-hour period to meet demand.  Assuming sites could be found, and that their costs were similar to Dinorwig, then the bill looks like it would be a cool £100bn.  That’s a very big number indeed, equivalent to around 2.5 times the estimated cost of the HS2 rail line.


EVs for PVs

But maybe there’s a cheaper way.  Maybe the answer to the intermittent nature of renewable energy is sitting in driveways and garages all over the country.

Would it be possible to use the batteries of electric vehicles to store energy from times of excess generation and release it at times of need?  What do the numbers look like?

One of the main thrusts of electric vehicle development is battery storage.  The challenge is that although most vehicles are used for relatively short journeys most of the time, occasionally we like to go away to visit family or friends further afield.  We want our vehicles to have enough range to cover these infrequent long journeys.

Here’s the capacity and estimated range of a few of the electric vehicles currently on sale.


Most manufacturers have settled on a battery size around 25kWh as the right compromise on range and cost.  It seems likely that for mass adoption the manufacturers will need to develop vehicles with higher ranges.  However, let’s use 25kWh as a median value for the battery size in a car.

The annual average mileage for a car in the UK is 8,200 miles, or 22.5 miles/day.

Taking the average miles/kWh from the table means that to meet the average mileage a battery of only 6.3kWh is required.  On most days, the car has 19kWh of excess capacity that is only needed for those unusual, longer journeys.  If, through smart metering and information technology, this excess storage could be made available (perhaps yielding a small income to the car owner), how much storage could it provide?

There were 29.1million cars on the road in the UK in 2013.

If we assume that the entire UK car fleet changes over to electric vehicles then the excess battery storage this would represent is 515,000 MWh, higher than the amount that I calculated would be needed to spread PV generation out through a 24-hour period.

Hold on a Minute

Before we get too carried away, there are holes in this rosy picture of the future that need to be mentioned. 

We’ve only thought about storing energy from one (sunny) day to the next, but we need to solve the problem of having energy available in winter too, or on days where the sunlight is lower. 

One option is to install a massive excess of solar PV – many times more to meet demand in winter, 90% of which would then be idle on sunny days.  Solar panel prices would have to drop spectacularly to make this approach viable.

Another concept, popularised by the Desertec Foundation is to locate the solar panels in a place where it’s sunny more of the time and day length doesn't vary so much.  We then build low-loss power cables to connect from the desert to population centres.
It seems to me that wind, wave and tidal power generation will have a prominent role to play, due to their complementarity to solar – the wind still blows at night and average wind speeds are higher in winter.

Whatever the future holds, its clear that energy storage has a significant part to play, and how ironic if the car, which did so much to drive the development of the global petrochemical industry, will play a central role in bringing our reliance on fossil fuels to an end.