The Myth of Coal Reliability
I have written frequently about how RE + storage is now cheaper than coal in India and elsewhere and one of the criticisms I often receive is that coal is more "reliable". Having worked for a (primarily hydroelectric) utility (Manitoba Hydro) where we made huge arbitrage margins off the inflexibility of coal, this argument always struck as a bit odd. Most utilities worldwide rely on Hydro or gas to maintain stability, and coal-dependent countries like India and Australia are currently paying the price. Here I am going to break down some of the "reliability" factors and discuss how coal compares to RE + storage, primarily from the perspective of India as the system I know best.
In my home utility, we used to assume coal was available ~85% of the time, but the latest data available I could find for India suggested availability was between 71%-77% (average of 74%). The availability of RE depends on how much storage you add, but it is disingenuous to compare the cost of 100% available RE to 75% available coal. A lot of people may be surprised the availability of coal is so low, but it is worth remembering that both supply and demand are diverse-not all states have peak demand at the same time, and not all plants are available at the same time. Yes wind and solar availability is perhaps more correlated than coal, though this is true to a certain extent for coal as well. For example, a friend of mine said that up to a third of Australian coal supply is offline in the current power shortage due to a combination of unplanned outages and fuel supply. Storage uptimes tend to be 99%+, which is one of the reasons RE + storage is much more reliable than coal (if a sufficient amount of RE is built, to have surplus to charge it).
While the Indian government target is for coal plants to achieve 1% ramp/minute, most coal plants do not in fact achieve this level, compared to hydro plants which are typically above 10%/minute (or even higher depending on the design) and battery storage which can ramp on the order of milliseconds.
I find it ironic that the same people who talk about the reliability of coal often have "stabilizers" in their home (or built-in stabilizers) for their refrigerator/AC. To the great credit of transmission operators like POSOCO the grid stability has improved in recent years (aided by grid interconnections, improved technology) and should further improve with the opening of an ancillary services market. However, coal-based (especially supercritical) systems are just structurally less flexible to meet variations in demand, a problem which long pre-dates the large-scale introduction of wind and solar. Coal is so notoriously inflexible that in many parts of Europe and North America which have fully transitioned to an open market, coal plants will actually be forced to take *negative* prices at low-demand periods, simply because they cannot easily ramp down or turn off and on again, as opposed to storage which can switch in milliseconds, or even wind and solar which can at least be turned off fairly instantaneously if required. It is telling that after regional blackouts in the 2016/2017 Australian summer, 100 MW of storage came to the rescue and has vastly improved system stability in South Australia, not coal.
Both RE and coal plants struggle to meet capacity, which is why when I discuss solar and wind, I typically recommend 2-4 hours of storage. In coal-dependent regions, one often sees regional load-shedding during peak periods. Most utility demand curves reach the top 10-20% of load for just 500-1000 hours. For example, in Ontario, the peak demand in 2021 was 23.0 GW, but the P90 demand was 18.3 GW. So the last 4.7 GW (~20%) is only being used <10% of the time. Building a coal plant to run with a CUF of 10% or less is prohibitively expensive (In India, think fixed costs of Rs. 15-20/unit), which is why Ontario has ended up with demand response and energy storage as the winners of its technology-neutral capacity auctions, a practice I hope to see replicated elsewhere in the world.
For daytime peaking regions, coal is even worse as output drops due to water scarcity and higher output temperatures (the thermodynamic efficiency of coal is a function of the difference between the boiler temperature and the ambient temperature).
Another line I often here is that coal is somehow good for "development". It is telling that the coal-producing states (Jharkhand, Chhattisgarh and Orissa) tend to have among the lowest reliability, whereas high RE states (Gujarat, Karnataka and Tamil Nadu) are among the highest reliability. While there are obviously lots of confounding factors, it's not like access to coal is correlated with high rates of energy access. The biggest challenge for energy access tends to be the cost of transmission/distribution, particularly in remote/hard-to-reach places. Coal has to be built in large "chunks" (at least 500 MW), which makes it ill-suited to remote areas with low demand. This is a problem for hydropower as well (my home utility had four communities using diesel due to the cost of building transmission to their remote locations). In contrast, solar (and sometimes wind, geothermal, biomass, etc.) can be built modularly to meet the demand of local communities. This is why, for example, remote areas in Ladakh, Nagaland, or islands can get much cheaper and more reliable power from RE + storage (currently, the typical solution is much more expensive diesel generators). If we really care about development in marginalized communities, RE + storage, not coal, is the "reliable" answer.
Responsive to Changing Demand Forecasts
I have heard the current energy shortage in India blamed on lack of planning due to covid-19. That is, the lowered demand in 2020 and 2021 led to utilities underinvesting in capacity for when demand rebounds. While I personally think a rebound in demand post-pandemic should have been expected, it should be noted that a coal plant takes 5+ years to be built, whereas wind and solar can be built in 12-18 months. As per Tesla's famous 2017 bet with South Australia, where 100 MW of storage was built in just 100 days, energy storage can be developed even faster. Like solar and wind, battery storage (though not pumped hydro) is modular, and can be built in small or large chunks, even in the face of unpredictable demand. This is useful not just in power-short countries like India but also in many Western countries which have seen declining demand in the last decade. Investments in coal (or hydropower) would have led to uneconomic oversupply.
To me a "reliable" system is not just reliable in a physical sense, but also in an economic sense in that it can provide predictable, consistent costs. As the recent price volatility in energy markets has shown, coal plants are anything but, with prices rising 300%+ in the past year alone. It is ironic that there is much hand-wringing about solar and storage prices rising 20-30% in the past year. While it is certainly unfortunate that more RE and storage wasn't built when prices were lower, RE + storage plants built today (still at a significant discount to building new coal) will be locked in for the 25+ year life of the project.
Comfort with Change
At the end of the day, when people say coal is "more reliable", I think they have not actually thought through the various parameters of reliability, as I have done here. What they mean is they are comfortable/used to coal, and changing the operating practices from using coal to using RE for energy and storage for capacity is a big change from both a bird's eye structural planning perspective as well as a day-to-day operational perspective. When I worked for a utility, even adding a (relatively) small 200 MW hydroelectric plant resulted in hundreds of hours of work to train operators, update SOPs, reconfigure protection and control, transmission settings, maintenance practices, etc. etc. I certainly sympathize with the hard work that is required to change operating practices for the energy transition. And I certainly sympathize with other aspects of the just transition such as retraining coal miners or coal plant operators for new roles. But let's stop with the myth that coal is more reliable than RE + storage. By any objective metric, it is not.
This article is also published the author's blog. Energy Voices is a democratic space presenting the thoughts and opinions of leading Energy & Sustainability writers, their opinions do not necessarily represent those of illuminem.