Baseload power — a thing of the past?
Everyone knows the electricity system needs to be decarbonized by adding more renewable energy. However, how the composition of the power supply, or what we call the “power mix”, will look in the future seems to be still a matter of debate for many parties.
Some argue that baseload power is needed; we can’t run the grid system with sufficient stability without a baseload power supply.
This picture shows a representative image of the debate. The figure on the left represents how the electricity system works in many places today. We used to run the grid with baseload powerplant runs 24/7, while load followers and peakers like gas urbines fulfil the remaining demand fluctuation. Some believe that keeping baseload is a must, especially since renewable like solar and wind are unreliable. How can we keep the grid system stable if fully powered by intermittent renewable?
Meanwhile, on the right is the aspiration from the pro-solar and wind group. They believe that we can run the grid with fully solar and wind. Any fluctuation is balanced by battery storage. Which paradigm is true?
Baseload is a limitation of the powerplant, not a requirement for the grid system
My first argument on this debate is that running a powerplant in a baseload mode is not a mandatory requirement of the grid system. It must be noted that, as the name suggests, there is a base-load. Meaning that in any grid system, there is always a minimum load that will be demanded all the time. Look at your home; appliances like refrigerators run 24/7, right? So do some of the industrial loads, which must run continuously.
However, it does not necessarily mean that the minimum load needs to be served with a powerplant running in a stable, continuous way. It can, but not the only way. And that’s what power engineers are pursuing today: how to run the grid with a decreasing supply of baseload power plants.
In fact, power plants like coal and nuclear are run continuously because of their limitations. These baseload powerplants are CAPEX heavy, meaning that they are expensive to construct but dirt cheap to run (the composition of OPEX or fuel is small compared to the total cost of generation). And because of this, it is much more economical to run the powerplant all the time.
Another reason, it is caused by the technical nature of its boiler system. The power output can be ramped up and down but very slowly. It takes up to an hour for coal powerplant to go from zero to full capacity, while gas turbines only take 5 minutes.
Baseload operation is not compatible with variable RE
Apparently, there is no middle way between a traditional system with baseload and the new system dominated by variable RE.
I would argue that in electricity market that promotes variable RE, the baseload power will be more and more making losses.
Why?
As I explained before, coal (or coal with carbon capture if you want to make it low-emission) and nuclear power plants are CAPEX-heavy infrastructures. Running these plants in a variable way to match the variable RE’s output means that they will operate in fewer hours. Less operating hours means less revenue to pay back the initial investment.
Over time, these plants will suffer with higher penetration of variable RE. At some point, running the power plant will no longer be economical, and the owner will opt to shut down the plant.
Meanwhile, for load-followers and peakers, they are a perfect couple for the intermittent RE since they have the capability to match the fluctuations. They will be valued much higher in times when REs are not available.
In conclusion, it is either to keep variable RE to a certain limit or to sacrifice the baseload power for the sake of more solar and wind. Up until today, I haven’t seen any middle-win solution.
Zero fuel cost, variable RE is dispatched anytime it is available
In a competitive power market (not monopolized by a state-owned company), variable RE is in favour. Why?
Because it costs nothing to run the variable RE, as most of you know, we get free energy from the sun and blowing wind. The cost structure of solar PV and wind farms is mainly composed of CAPEX and fixed cost (like regular maintenance, you do it no matter whether you are running the plant or not). Fuel cost is essentially zero.
You may start to realize that variable RE is an extreme version of a baseload powerplant (from a financial point of view). For grid operators, it is always cheaper to run solar PV and wind anytime they are available.
If coal vs solar PV has a similar CAPEX of xxx $/MW, then the grid operator will always choose solar PV since dispatching it costs zero. In fact, this is what happens in several regions where some of the time only solar and wind are sufficient to cover the demand, and the electricity cost is zero, even negative.
Again, this only works in a competitive wholesale electricity market, while a state company with a monopoly has the option to run a baseload and curtail the excess solar and wind output.
The future?
Another threat for baseload power plants is that the cost of solar PV, wind and battery storage is declining (while baseload is not). Up to what point is anyone guess.
This only means one thing: The future of grid systems will no longer be dominated by baseload power plants.
If we look at microgrid, such as the ones in remote villages or islands, there are abundant examples of grid that runs with small share of baseload power. To give real examples: Maui Island, Hawai, 200 MW peak demand, 76% by solar and wind. For large grid system, Ireland (6 GW peak demand) has been able to showcase how a country could operate with variable RE dominance. It is supplied by 29% solar and wind annually, while occasionally it could reach 65% on some hours (NREL, 2018).
Baseload power plants will still contribute to the power mix. As you know, there are renewable baseloads, like geothermal, biomass, and coal, with carbon capture. But they will be supplying a minority portion of the electricity system.
However, operating grid dominated by variable RE is not without hurdles. Reliying on intermittent power sources means that a smarter grid system is needed. Technologies such as storage, smart grid, demand response, smart metering, etc will be highly demanded in the future. I will talk about this deeper in the next article, stay tuned!
This article is not possible without the support and review from Gugun Bonar, an expert in Power System from PLN Indonesia who is currently completing his master degree in University of Queensland, Australia.
