Merit order gives (too) high values per kWh

[ThomBuijs]

I'm using this LES. When applying the merit order to a stakeholder I get an average price per kWh for the realised flex of €0.60/kWh. I think this is way too high. Also, I am unable to download the merit order price/load curve and it gives me tis error:

We're sorry, but something went wrong. If you are the application owner check the logs for more information.

So I couldn't see the actual merit order curves. So I checked the total amount of shifted flex which is equal to 74608 kWh and the amount payed by the system operator to customer AC1b which is equal to €44,764.61.

The LES is scaled from a SER 2023 scenario and no other changes have been made.

[jorisberkhout]

Could you look into this with priority, @grdw ?

[grdw]

We're sorry, but something went wrong. If you are the application owner check the logs for more information.

Yes, this is a caching problem as far as I'm aware.

Technical details:

No such file or directory @ rb_file_s_mtime - /var/www/etmoses/releases/20161026091620/tmp/networks/production/666/feature/electricity/49a08c0e397826d795e50f9a6a16b08f35708bd965875a2b22d9bb1db5e25bc6.load.current_week.tmp

Including @antw

[jorisberkhout]

@grdw and I looked into this. We managed to download the merit order curve and found that all 35040 entries in this curve are 0.60. The merit order curve of the SER 2023 scenario shows more variability in the price and the average price is 46 EUR / MWh or 0.046 EUR / kWh. Do you know what might cause the difference, @antw ?

[antw]

Do you know what might cause the difference, @antw ?

Not off the top of my head, sorry. I can look into this as soon as I've dealt with quintel/etmodel#2296.

[jorisberkhout]

Not off the top of my head, sorry. I can look into this as soon as I've dealt with quintel/etmodel#2296.

Thanks!

[antw]

The merit order curve of the SER 2023 scenario shows more variability in the price and the average price is 46 EUR / MWh or 0.046 EUR / kWh. Do you know what might cause the difference, @antw ?

The production park is completely maxed-out because there are far more households in the LES than the original scenario.

The scenario was scaled to 610 households, but the LES has 47500 households present (aggregated and individual). I think it's fair to say that a central production park scaled for a region of 610 households is incapable of meeting demand for 48k. Hence all the central producers are running all the time. €600EUR/MWh (€0.60/kWh) is the fallback price used by Merit in the case that all producers are fully-loaded.

[jorisberkhout]

Thanks, @antw . Does this answer your question, @ThomBuijs and @JeroenvdLogt ? If so, could you close this issue?

[JeroenvdLogt]

This is strange because I can't find any unit field with a value of 100+. Also in the topology all the unit values are 1 or not filled.

[antw]

This is strange because I can't find any unit field with a value of 100+. Also in the topology all the unit values are 1 or not filled.

You're completely correct, I don't know what I was thinking. I misread the "demand" fields; in my defence, the demand field and unit label are quite close. 😬

The explanation for the price is still true though: €0.60/kWh means that all the central producers are maxed-out. I will take a quick look into why this is the case.

[DorinevanderVlies]

The following might help the search. I just looked at my LES which is based on an ETM scenario with an overload of electricity. The merit curve consists solely of 0.60 euro/kWh.

[antw]

@ThomBuijs @JeroenvdLogt The incorrect pricing should be fixed now. A (pretty big) oversight caused dispatchable sources of energy to not be included when calculating the merit order price.

[MarjoleinSchwachofer]

I have the same problem with a LES that ties to this one: In https://moses.energytransitionmodel.com/testing_grounds/681/edit it looks good but in https://moses.energytransitionmodel.com/testing_grounds/683/edit the merit order gives a to high cost from supplier to producer. Why doesn't this work now? I need this for a presentation.

[MarjoleinSchwachofer]

looks fixed after some trying, but don't know why

[antw]

looks fixed after some trying, but don't know why

The business case probably had not been recalculated; you made a change to the market model or LES and this forced it to be recomputed; replacing the old and incorrect prices with the new, correct ones.