,
28 tweets,
5 min read
Read on Twitter
1/ It all comes down to the economics of adoption. Coal is becoming a stranded asset; banks are not backing coal plants (which coal people see as a conspiracy rather economics).
2/ Cost, siting, price risk (the risk that your investment today will become uneconomical in competition with future, more advanced installations), and scale (which sets the magnitude of the risk), time to ROI, and risk uncertainty.
3/ They're going to need a lot more storage, but how much cannot be determined just by adding a few numbers.
The type of generation and its geographic distribution, transmission, the rise of rooftop solar, the evolution of demand, the ability to shift the time-of-day of loads.
The type of generation and its geographic distribution, transmission, the rise of rooftop solar, the evolution of demand, the ability to shift the time-of-day of loads.
4/ Storage is not the only way to match the demand curve with the supply curve. It is the most flexible, but it is not as inexpensive as, say, importing wind from another part of the country when it falls short here, or charging electric cars at a different time of day.
5/ The choice of storage technology will be dynamic. Pumped hydro faces siting and scale costs; a pumped hydro project in California near Sacramento was dropped as batteries became more competitive and the projections of need for it became less clear.
6/ But the large energy capacity of pumped hydro makes it a valuable component. Alternatively, with vanadium flow batteries, you just make the tanks bigger and fill them with more vanadium solution. Hydrogen storage has potential for long-term as well.
7/ These alternatives have fewer restrictions on site selection. Current hydrogen technology has poor efficiency, requiring more generation. But these tradeoffs will be made in a series of decisions as the markets and technology mature.
8/ There's a simple rule of thumb. Big projects require long-range planning and long-range economic projections. You will see big projects (hydro, nuclear, concentrated solar, large wind farms done as a single project) when there are clear economic needs over a long period.
9/ But the rest will be smaller or more incremental projects, from rooftop solar and home storage systems, to adding turbines to a wind farm or small-to-medium photovoltaic farms near points of use or out in the desert.
10/ And these projects benefit from knowing how the net demand curve (demand - supply) has evolved.
And storage will be added where it's needed, designed for the roles it's needed in. This is economically efficient.
And storage will be added where it's needed, designed for the roles it's needed in. This is economically efficient.
11/ So the lack of storage is not a great concern to me. After the big 2016 blackout in South Australia, they have come to recognize & address their deficiencies. Two major battery facilities have been added—the Hornsdale Power Reserve by Tesla was briefly the world's largest.
12/ One deficiency was the short-term storage referred to as "system inertia" because it has historically taken the form of inertia in rotating generators. It is the ability to supply additional power for a brief time in the face of sudden changes.
13/ Wind can be configured to provide it, using the inertia of the turbine blades. (It wasn't at the time of the outage). Solar can provide it if operated at reduced capacity.
But batteries do it better, for longer.
But batteries do it better, for longer.
14/ But the 100 MW/129 MWh Hornsdale Power Reserve and the new Yorke 30 MW/8 MWh are not there to move power between times of day, aka "arbitrage". Their role is actually much more valuable. Hornsdale's $A90 cost has nearly been paid back in $A40/yr savings. 
15/ One role, formerly filled by gas turbines, is to provide reserve power. In the event of the loss of a power plant or transmission line, they can step in and supply power while the problem is resolved, possibly by bringing another power source online.
16/ Hornsdale has done this many times, and done it so much more quickly and accurately than the gas turbines that it now part of the first line of response to incidents.
17/ Likewise, in California after the Aliso Canyon methane leak, batteries were installed in a few months to replace the gas turbines that depended on the Aliso Canyon storage facility.
18/ But another valuable role is frequency control and ancillary services. This is an essential service to keep the grid synchronized and avoid knocking power plants offline. And this is something batteries, wind, and solar (in decreasing order) can provide.
19/ This is basically short-term energy storage, and is the "system inertia" referred to in #12. If the frequency drops, it pushes the voltage higher, taking a bit of the load and letting the rotating generators speed up.
20/ If the current leads or lags, they absorb and return power at the right moments to bring them back in phase.
Traditionally, these have been handled by rotating devices or switched banks of capacitors, but the inverters that convert DC to AC do it better.
Traditionally, these have been handled by rotating devices or switched banks of capacitors, but the inverters that convert DC to AC do it better.
21/ These high-value services, combined with quick ease of deployment, have driven the growth of batteries to date. Technology adoption always hits the high-value targets first.
22/ But as renewables approach 100%, the need to store power for longer periods will become more valuable. The alternative to storage is to have sufficient excess wind, geothermal, and hydro at all times and throw away the excess.
23/ But that is not economically efficient, so storage WILL be added. It is also not robust; lightning strikes frequently take out transmission lines. South Australia is rapidly decentralizing their grid.
24/ In addition to rooftop solar, 10s of thousands of behind-the-meter batteries are being installed in homes, flattening the demand curve and reducing the impact of any load-shedding or blackouts.
25/END These decentralized micro installations are the vanguard of the time-shifting arbitrage that will be needed. And by making it e asier to match generation and load, they encourage further replacement of coal plants with wind and solar.
I covered it briefly. #6, but also #7
But I suspect you're thinking of it as transportation fuel rather than storage? I would prefer to consider that as a new dispatchable load, since it doesn't return to the grid. The same for charging electric cars.
But I suspect you're thinking of it as transportation fuel rather than storage? I would prefer to consider that as a new dispatchable load, since it doesn't return to the grid. The same for charging electric cars.
In fact, generalize that to synfuel production, not specifically hydrogen.
And this is one of the reasons I am not sold on a hydrogen transportation economy. The other is the scale of the infrastructure investment needed creates a chicken-and-egg situation.
But hydrogen grid storage might lower that barrier as a side-effect.
But hydrogen grid storage might lower that barrier as a side-effect.
