Handle power plant co-firing and fuel shifting

[jrissman]

Megan and I discussed the most tractable approach to handle co-firing (plants using different fuel types, such as a start-up fuel and a main fuel, or splitting their generation between two fuels). This generally involves using different equipment (i.e., a fluidized bed coal unit and accompanying natural gas equipment), which can have different efficiencies, emissions profiles, etc. We can't just divide up fuel types at the end (i.e., the "hard coal" plants across the fleet are burning 97% coal and 3% natural gas) because this wouldn't reflect the different heat rates and other properties of the fuels used.

The approach we thought of is along these lines:

• We already track power plants by type as a quantity of MW of generation (at a given vintage). We already aren't tracking specific plants or units.
• For co-firing plants, we allocate their MW of capacity into the correct corresponding buckets. For instance, a 100-MW coal plant that burns 90% coal and 10% gas would simply allocate 90 MW to the "hard coal" plant type and 10 MW to the most-appropriate natural gas plant type (NG peaker, steam turbine, combined cycle, etc). This would be a change in the input data, especially BHRaSYC Start Year Capacity.
• We set MPPC Minimum Power Plant Capacity, which we use to quantize new additions and retirements, to a small value like 1 MW or 0.01 MW for any fuel types that can be used for co-firing. This ensures the quantization doesn't interfere with the ability to build or retire plants that use co-firing.

Optionally, if we have time:

• We add a fuel shifting section to the code. This refers to fuel shifting without buying new equipment (i.e., not retrofitting). Based on the costs of the fuels and other factors, we determine how much shifting from one fuel type to another occurs during each modeled year. We change the MW from one plant type to another, without changing the vintage of those MW. Shifts should be distributed proportionately across vintages.

We had fuel shifting before (where it was added to help model KSA plants shifting between crude oil and heavy fuel oil based on the prices of those fuels), so we had planned to re-implement it anyway. Co-firing just gives that capability more relevance in more geographies.

Retrofitting (like a coal-to-gas retrofit, or a non-CCS-gas to gas-with-CCS retrofit) would be handled in a separate "Retrofitting" section of the code, which would update vintage. Retrofitting should not be commingled with fuel shifting in the model structure; they should be two separate, small mechanisms.

[robbieorvis]

Thanks for this.

Intuitively, it makes sense. There are at least two issues I see with the proposed approach though:

1. We cannot lower the quantization unit the way proposed, because we need those high values for coal and gas combined cycle plants, which are the types of plants we are talking about that might cofire with biomass or hydrogen. So that gives me pause. There might be a new variable or something that we add downstream, but this variable serves an important purpose as designed. Without the high values, the model would incrementally build small amounts of coal power plants and gas power plants, for example, which is not realistic (and would not be a tradeoff I'd make to enable co-firing).
2. Breaking apart the power plant types vastly increases the costs per MW, because it assume an entire new power plants of X MW needs to built for the co-fired technology, when in reality we might just be talking about either a different burner (e.g. I believe H2 cofiring just needs special fuel burner tips, and I'm not sure about biomass cofiring but I suspect it might be able to use existing boiler technologies) using the same generator and accompanying power plant equipment. It might add $1000/MW, for example.

The desire for cofiring in the model is actually less about historical plant types, where cofiring is mainly about oil and gas, and more about compliance options in the future, FWIW.

Anyway, maybe we don't need to tackle this right now, but I'd like to have a longer conversation about it at some point, and I don't think the proposed structure here is quite right.

From: Jeff Rissman @.> Sent: Tuesday, June 13, 2023 2:27 PM To: EnergyInnovation/eps-us @.> Cc: Subscribed @.***> Subject: [EnergyInnovation/eps-us] Handle power plant co-firing and fuel shifting (Issue #278)

Megan and I discussed the most tractable approach to handle co-firing (plants using different fuel types, such as a start-up fuel and a main fuel, or splitting their generation between two fuels). This generally involves using different equipment (i.e., a fluidized bed coal unit and accompanying natural gas equipment), which can have different efficiencies, emissions profiles, etc. We can't just divide up fuel types at the end (i.e., the "hard coal" plants across the fleet are burning 97% coal and 3% natural gas) because this wouldn't reflect the different heat rates and other properties of the fuels used.

The approach we thought of is along these lines:

• We already track power plants by type as a quantity of MW of generation (at a given vintage). We already aren't tracking specific plants or units.
• For co-firing plants, we allocate their MW of capacity into the correct corresponding buckets. For instance, a 100-MW coal plant that burns 90% coal and 10% gas would simply allocate 90 MW to the "hard coal" plant type and 10 MW to the most-appropriate natural gas plant type (NG peaker, steam turbine, combined cycle, etc). This would be a change in the input data, especially BHRaSYC Start Year Capacity.
• We set MPPC Minimum Power Plant Capacity, which we use to quantize new additions and retirements, to a small value like 1 MW or 0.01 MW for any fuel types that can be used for co-firing. This ensures the quantization doesn't interfere with the ability to build or retire plants that use co-firing.

Optionally, if we have time:

• We add a fuel shifting section to the code. This refers to fuel shifting without buying new equipment (i.e., not retrofitting). Based on the costs of the fuels and other factors, we determine how much shifting from one fuel type to another occurs during each modeled year. We change the MW from one plant type to another, without changing the vintage of those MW. Shifts should be distributed proportionately across vintages.

We had fuel shifting before (where it was added to help model KSA plants shifting between crude oil and heavy fuel oil based on the prices of those fuels), so we had planned to re-implement it anyway. Co-firing just gives that capability more relevance in more geographies.

Retrofitting (like a coal-to-gas retrofit, or a non-CCS-gas to gas-with-CCS retrofit) would be handled in a separate "Retrofitting" section of the code, which would update vintage. Retrofitting should not be commingled with fuel shifting in the model structure; they should be two separate, small mechanisms.

- Reply to this email directly, view it on GitHubhttps://github.com/EnergyInnovation/eps-us/issues/278, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AK5N6SIGVGP56UNJECTY7L3XLDLL5ANCNFSM6AAAAAAZFOSQTU. You are receiving this because you are subscribed to this thread.Message ID: @.**@.>>