Early in 2015 respected Australian energy market blog WattClarity put some simple estimates of required battery storage capacity for the NEM on the back of an envelope. To summarise:
- Actual demand and wind data was used
- Wind was low, mainland NEM-wide, for 62 hours
- Median demand was set at 21,000 megawatts
- The necessary electric vehicle battery capacity to fill the gap was gigantic at 1,271,000 megawatt hours (MWh)
- It was 992,000 MWh if the wind fleet were 10 times its current size
Since those three days of relative calm, to say that the hyperbole around the imminent triumph of battery storage has only escalated is an understatement.
— David Hess (@6point626) August 11, 2016
But the assurances are invariably unaccompanied by arithmetic, and WattClarity’s calculations, like its technology agnosticism, remain as rare, much-needed examples.
But how many?
The assumption by many and the party-line for some in particular is that batteries will take the place of coal and gas as solar and wind dominate the electricity supply, pushing out the antiquated power stations. WattClarity’s estimates provide the basis for answering the inevitably absent question of how many?
But first, would solar capacity make a substantial difference? APVI records for those three days (28th to 30th of March) are incomplete, but assuming a calculated average 51% of nameplate output (1,876 MW) at noon maximum for NSW, QLD, SA and VIC, and generation from 6:30 am till 5:30 pm, solar supplies 10,320 MWh in total. Let’s multiply this by 10, like was done for wind, which provides daily peaks exceeding the average residual demand (16,000 MW), and as such the remaining nightly gaps of demand are reduced to a rough total of 682,800 MWh.
But how much?
Lithium-based battery technology, by far the dominant form for the foreseeable future, has dropped in unit price very rapidly. The AFR article linked above suggests a cost of USD $600 per kWh will be possible soon – which would today be AUD $798,000 per MWh.
Thus the necessary batteries on the back of this envelope cost on the order of AUD $545 billion.
But for how long?
10 years. We can be sure of that much. ESCRI-SA provides details on the expected 10 year lifespan in its on-going storage study, and this is consistent with the already impressive chemistry involved in lithium batteries.
To paraphrase Sir David MacKay, “I’m not anti-batteries, I’m just pro-arithmetic.” The ESCRI-SA project is fascinating, as is the innovative AGL virtual power station, and projects like Kingfisher present the cutting edge of aligning solar power with demand. But we’ve got to appreciate the scale here. We don’t have ten times the wind and solar as used for these thought experiments, and getting anywhere near it will be very hard work. And no amount of enthusiasm can overcome the reality of hard physical chemical limits to which the materials in batteries are immutably constrained – a given density will yield a certain voltage for a particular period, with no real reason to hope for a “big leap forward“. We’ve got to ask hard questions when batteries are proposed as the panacea to our very imminent energy supply challenges. Maybe just as importantly, we need to look at who is proposing it. Let’s evaluate all the options, from that sensible fulcrum.
To that end, there are already notable examinations of the potential value of battery storage. Crucially, they are not shy about including nuclear capacity in the analysis:
There is no silver bullet to decarbonize the electricity sector: the least-cost generation mix includes a diverse mix of resources and wind, solar, and flexible nuclear technologies co-exist in the optimal low-carbon generation portfolio, regardless of the level of energy storage. Under an emissions limit of 100tCO2/GWh, nuclear’s contribution to total energy supply ranges from 18–40%, depending on the amount of energy storage installed, while solar and wind shares are in the 9–15% and 23–43% ranges, respectively. Likewise, flexible nuclear contributes 52–68% under a tighter 50 tCO2/GWh limit while solar contributes7–14% and wind 12–19%, depending on the storage capacity.
- Safaei and Keith in Energy & Environmental Science, open access.
The optimal energy capacity of bulk electricity storag (BES) turns out to be small in general, even when we impose ∼70% emission reductions compared to business-as-usual. The mechanical storage fleet was sized to supply the average electric load for one full day on its own. This value sharply drops as the energy cost increases while the power cost simultaneously drops; i.e. moving to electrochemical systems. These relatively low energy capacities signal the unimportance of large-scale storage of electricity over long time horizons (e.g. seasonal storage) from an economic point of view. This is driven by the lower competitiveness of BES systems coupled with wind in comparison to low carbon and dispatchable generation facilities, like CCGT and DZC (nuclear or fossil with carbon-capture) modeled here.
There are cautions from industry leaders like Ibedrola and Tesla, which should be heeded. Where pilot projects – like GRID4EU and Yanbaru – have not performed, we must understand why. Notwithstanding their lifecycle carbon intensity, storage technologies need and deserve rigorous and impartial assessment alongside our future energy sources, not and the disrespect of ceaseless, motivated oversell.