While noble in its intent to be able to directly compare the cost of energy sources, LCOE (Levelized Cost of Energy) is fatally flawed and is leading to incorrect and potentially dangerous decisions. From an early age, we are taught the fallacy of apple to oranges comparisons. This is exactly the problem with LCOE. At its core, it is calculating the cost to generate an electron, period. It is not comparing energy generation systems only the cost of generating an electron from a technology.
Renewable energy is intermittent and non-dispatchable. There is no electron generation during the night or when the wind doesn’t blow. The world can’t put life on hold during low or no generation from renewables. So, the very low cost of renewable generation is meaningless if I don’t value the power when it is available, and it is infinitely expensive if it isn’t available when I need it. The LCOE is an apple to oranges comparison. It ignores all the other aspects of a commercial and viable energy system, such as the ability to meet the customer’s needs when they need it. To really make a comparison worthwhile, we need to expand the scope of the analysis from an electron to a viable and practical energy system.
To accomplish this, let’s define a viable energy system as a generation process that is able to meet the customer requirements for energy 24 hours a day, 365 days a year. So for our discussion we are going to assume a 1 megawatt 24 x 7 X 365 load.
Meeting this load requirement with a combined cycle natural gas plant is quite simple. It has the ability to run 24×7 365 days a year (Ok, it would require some downtime for maintenance, but that would only reduce the capacity factor by a few percent). Now we will often see combined cycle natural gas plant capacity factors in the 40% range. That is the actual dispatched capacity required to accommodate other less flexible baseload plants such as coal and nuclear and of course solar and wind are given preference. This 40% capacity factor does not represent what the combined cycle plant is capable of, only what is currently dispatched. A natural gas combined cycle plant is absolutely capable of a capacity factor of 90%. This is in contrast to the capacity factor of solar at 20%, which is all that the technology is capable of achieving.
So, to make our apples-to-apples comparison, we need adjust the solar plant to supply power 24 X 7 X 365. Since solar is only available when the sun shines, we will need to store solar for the 14 or so off hours. To accomplish this, we need to have a larger solar plant than our load, so we can service the load at night. To store enough power during the solar day to store for the evening load, we must overbuild the solar by 65%. So, to service a 1 megawatt round the clock load, we will need a solar plant equal to 3 megawatts in comparison to our 1 MW combined cycle natural gas plant.
So just from a power perspective, we need a solar farm approximately 3 times the load and store the excess generation for nighttime. Unfortunately, we’re not finished. There is a considerable fall off in production during winter as compared to summer. Typically winter production is about 40% of the summer production. Since our solar plant needs to service the 24 X 7 X365 load year-round, we need to make it even bigger so that we have enough production to satisfy winter load from winter production. This means that our 3 MW solar farm needs to produce 3 MW during the winter. To accomplish this, our 3 MW needs to be divided by 40% too come up with the actual required size of 7.5 MW. Again, we need 7.5 times more solar capacity than natural gas capacity to achieve the same load service. Now someone is going to say, this is unnecessary, since long duration storage will solve this mismatch. Unfortunately, this is not the case. Long duration storage is between 10 and 100 hours. Not weeks and months.
Now that we have the requisite solar size, we need to add the battery requirements. Assuming long duration storage at a cost of $ 200 kWh, a 1 MW hour battery will be $ 200,000. We need approximately 12 hours of such storage, so add an additional $ 2.4 million to the solar price tag.
As a quick review, to make an apples-to-apples comparison we need to compare 7.5 MW of solar with 12 MWh of storage to meet the same load services capabilities as a 1 MW combined cycle natural gas plant.
Just for reference and comparison here are the LCOE’s calculated by Lazards in their 2023 report:
Solar: median $ 60 MWh
Combined Cycle Natural Gas: median $ 70 MWh
Now let’s adjust these numbers for our apples-to-apples comparison.
We know that the amount of solar required is 7.5 times more than our load and we need to add an additional $ 2.4 million for batteries. This brings the LCOE for apples-to-apples solar comparison to $ 800 / MWh
Combined cycle natural gas LCOE moves in the opposite direction, since its capacity factor moves from 43% to 90%. This results in a reduction in LCOE to $ 37 / MWh.
Doing the math, solar is 21 times more expensive than natural gas from a functional load servicing perspective.
Just for full transparency, these are back of the envelope calculations but illustrative enough to show the absurd and dangerous distortion caused by LCOE.