Let’s continue with the formulas in the 100% Renewable Electricity Calculator. In previous post, I looked into the first of the formulas of the backup strategies (determining the cost of the capacity of the needed backup power plants). In this post, I will look into the formula that calculates the cost of the fuel that is needed to fill in the gaps left by solar and wind. The backup strategy that was proposed in the calculator is Power-to-Gas and then back to Power.

Initially, it was not really clear what kind of Power-to-Gas was used. Different (sometimes contradictory) labels are used in the spreadsheet, mostly “Power-to-Gas”, but also “Power-to-Gas methane” and “cost hydrogen” when calculating the backup fuel cost… So, is it Power-to-Hydrogen or Power-to-Methane that is integrated in the spreadsheet? Following the link in the “source and figures” sheet, it is definitely Power-to-Methane via methanation (creation of methane from hydrogen and carbon dioxide).

I could see methane work in a grid. It is a high density fuel that can be stored for a longer period and could work in existing networks/installations. It could be a solution for seasonal storage, more than for example hydrogen could be. It will however have its drawbacks like for example high conversion losses. So, I wonder how the calculator deals with those drawbacks and how seasonal storage is implemented.

This is a very simple schema of how Power-to-Methane-to-Power works (inspired by figure 4 of this paper (percentages are also from this paper):

The input is excess electricity produced by the (overdimensioned) solar and wind capacity that is used to make hydrogen (step 1). This hydrogen is then combined with CO₂ to make methane via the process of methanation (step 2) and stored (step 3) until it is being fed to gas-fired power plants (step 4) in order to fill in demand when solar and wind electricity production is insufficient.

The process has a 33% roundtrip efficiency, meaning that from the surplus power that is fed in, only 33% will finally come out in the form of electricity that is used as backup power when solar and wind electricity production is lacking. That 33% seems rather high though. The authors of that paper took a 58% efficiency of the final step (they called it an “improved gas turbine”). While such high efficiency in gas-turbines is not unheard of (modern installations can run slightly above 60% efficiency under optimal conditions like when brand new and running at full load), how realistic is this 58% for load following backup power plants? The whole point of such backup power plants is to follow the load on the grid and produce electricity in the amounts that are necessary. These are not base load power plants, the backup power plants as intended in the spreadsheet will cycle up and down which will reduce their efficiency.

Back to the spreadsheet. The Power-to-Methane component consists of two parts: the cost of the backup plants that burn methane when needed (previous post) and the cost of the methane production from excess electricity (this post). The cost of methane production is calculated in the sheet “calculator in cell O20. It is wrongly labeled as “cost hydrogen” (following the source of the number that Nitsche uses, this should be “cost methane”).

This is how the formula looks like:

  surplus electricity × (methane production cost × efficiency CCGT) ÷ 1,000,000

(CCGT = Combined Cycle Gas Turbine, a more efficient gas turbine because the waste heat produced from the gas turbine is directed back into a steam turbine, therefor producing extra electricity. This feature however works not so good under variables loads, therefor less suitable for load following power plants)

Not sure why these brackets are placed there in this formula. Firstly, they are not needed at all in the formula (multiplication and division have the same precedence) and secondly, they don’t clarify anything (maybe on the contrary, the methane production cost is more related to the surplus than to the CCGT efficiency).

Let’s look in the elements of this formula:

1. The constant value of 1,000,000 is a conversion from dollars to billion dollars (divide by 1,000,000,000) combined with a conversion of kW to MW (multiply by 1,000), therefor resulting in 1,000,000.

2. The source of the CCGT efficiency value is a EIA webpage detailing average heat rates for different turbine types. The selected heat rate value is indeed for CCGT, but at the bottom of the page there is the remark that these heat rates are at full load conditions. This seems logical. CCGT is mostly used as base load, meaning that it will run at or close to optimal conditions. This however raises the question whether CCGT is the right choice for backup in the spreadsheet? These backup power plants will cycle up and down or spinning idle (or are switched off), depending on the production of solar and wind at that moment. They will therefor run at a much lower efficiency and will burn relative more methane. Load following gas turbines typically run at an efficiency of 20 – 30%, not at full load. Meaning that the formula uses the efficiency of power plants that are not used as it is intended in the spreadsheet.

3. The link to the methane production cost details goes to the report “future fuel for road freight” and its subject is the economical, technological and environmental performance of alternative fuels for trucks in France (among which methane generated by a Power-to-Methane process). The report calculates a cost of €0.184/kWh for producing the methane, including transport, distribution and compression (bottom of table 36):

   

   Converted to dollars that is $0.2/kWh and this is the value that is used in the spreadsheet.

   Nitsche defended his use of the cost from this report in the comment section of another post in which he got the response that the report is about the production of methane as truck fuel, not as grid backup. Nitsche replied that this indeed was the case, but that methane could be used for other reasons than truck fuel and that he was only interested in the cost data. I could surely understand that it doesn’t matter for which purpose the methane is produced, but can this setup be extended to how it would work in a grid? The calculation is only about the fuel, it doesn’t include storage costs for the methane until being used in backup plants (step 3).

   Interestingly, also the efficiency of that Power-to-Methane plant was given (at the top of table 36):

   

   For every input of 562 MW as (surplus) electricity, there is an output of 245 MW as methane. That is an efficiency of 44%. If I now combine all efficiencies from the source material until now and assume a 30% efficiency of load-following backup power plants, then this is how the losses of the process look like:

   

4. Finally, the surplus electricity is electricity that was produced by solar and wind, but could not be used because demand was lower than production. The formula uses ALL of it, then multiplies with the methane cost per kWh and the efficiency of the backup power plants, but this is not really what happens in reality.

None of those three last elements in this formula are in line with reality. Then there is also the question “Where is the methane storage implemented in this calculator?”. There is (optional) battery storage incorporated, but there doesn’t seem to be something similar for methane.

The answer to that question will lead us to the fundamental flaw of the calculator and the reason why it can never ever come up with a realistic cost of a 100% renewables grid in the first place….

via Trust, yet verify

https://ift.tt/vAlSQRL

October 8, 2023 at 06:44PM