See also The EROEI of electricity generation.
In a future with the huge increase in variable renewables predicted in such places as zerocarbonbritain, the Offshore Valuation, and Sustainable Energy Without the Hot Air, it will be necessary to find ways of matching up supply and demand. Currently this is done by trimming the output of fossil fuelled power stations, with some unintended consequences for carbon emissions, at least in some places.
In part, balancing can be managed by ensuring as large a network as possible to increase the diversity of renewables and weather conditions as has been shown by Sinden, amongst others. In addition to this, demand side management can play a part (edit: potentially a very large part as pointed out by Dave in the comments, Tom Konrad, and Mark Barret in the work linked to below) in reducing the peaks of demand when power is not available. However, the most important way of matching supply and demand, both now and in the future, is through storing energy.
There are two types of storage – fuel stores and rechargeable stores. Fuel stores in the form of stockpiles of fossil fuels and dispatchable power stations have been the traditional way of matching supply to demand. This is supplemented to some extent by pumped storage and by interconnectors to France and Northern Ireland. These interconnectors are being added to as I write this with a North Sea Grid planned.
A way of looking at the effect of this requirement for storage on the efficiency of renewable energy is by calculating the EROEI of the storage and transmission technologies. This EROEI of balancing is also known as the lifecycle efficiency of the system. The EROEI in the case of rechargeable storage is a combination of the round-trip efficiency of the battery and the embodied energy of manufacturing, transporting and maintaining it.
The EROEI of balancing through transmission is a combination of the embodied energy of the Grid itself and the transmission efficiency. It is fair to count the whole of the embodied energy of electricity grids as the footprint of balancing. This is after all what grids are for. However it is less fair to lay it all at the feet of balancing variable renewables, as it has always been used to balance geographically dispersed and variable (or intermittent) fossil-fuelled power stations with equally dispersed and variable demand.
I’ve trawled around and found a few analyses of balancing technologies which I’ve shown on the chart below. The two grid systems have significantly higher EROEIs than battery technologies. It should be noted that only a proportion of the electricity produced will need to be stored, even with high renewables penetration, whereas all electricity will need to be transmitted over the Grid.
Edit: references are here.
In the graph, the top of each column represents the round trip efficiency. The red section at the top of each represents the loss of life-cycle efficiency from the embodied energy of manufacture and maintenance. The blue column represents the EROEI. Dark grey columns are technologies for which I couldn’t find any embodied energy figures and so just represent the round trip efficiency.
All of the figures required a few calculations to make them into EROEIs – in the case of the HVDC figures I’m not 100% convinced they’re right but you can check them for yourself at the bottom of the page). It would appear that grids, whether National Grid or a 4,000km long HVDC connection, are the most energy efficient way of balancing. Of course this depends on there being a match between supply and demand.
By the way, one obvious omission in all this is the other type of storage – fuel storage. Even though gas and coal have among the lowest EROEIs going for energy generation technologies, they are easily stored and that storage has a low energy cost. It would also be interesting to try and put a figure on what the lowest capacity factor for say CCGT would be before the EROEI fell to lower than one. That also raises the question of carbon return on carbon investment (CROCI) which I’ve not written about so far. The argument for balancing with fossil fuels may well not stack up so well in the carbon accounts as in the energy accounts, although that depends on how much we need to use it. At least one academic, Mark Barret, has created a detailed model of the UK’s electricity supply with only 5% fossil-fuelled balancing, which also incorporates additional HVDC inter-connectors and demand side management.
What would be interesting, now that we have figures for transmission, storage, and generation, is to try and model the impact of different mismatches between supply and demand, given different capacities of the storage and transmission methods available. A project for the future I think.
Harrison et al (2010) “Life Cycle Assessment of the Transmission Network in Great Britain”, Energy Policy
May, (2005) “Eco-balance of a Solar Electricity Transmission from North Africa to Europe”, Diploma Thesis for Technical University of Braunschweig
Denholm & Kulcinski (2004) “Net energy balance and greenhouse gas emissions from renewable energy storage systems” Energy Conversion and Management
At 1.7% losses per 1,000 km (p. 35) that is 93.4% efficient over 4,000 km. That means 1.071kWh generated per kWh delivered.
The embodied energy of a concentrating solar plant plus an HVDC line is calculated as 0.21MJ/kWh (p. 119) and of that the embodied energy is 3.5% (from the graph on p. 114). That makes 0.002 kWh embodied per kWh delivered.
Add the two together and you have 1.073 kWh embodied per kWh delivered and 1 / 1.073 = 0.93.