Couple TES temperatures to district heating supply temperatures

[cpschau]

Changes proposed in this Pull Request

This PR implicitly couples the top and bottom layer of PTES with the forward and return flow temperature of the local district heating network. Assuming a direct integration of the storage system, the bottom temperature equals the return flow temperature. The top temperature equals the forward flow temperature in the domain up to 90°C. At forward flow temperatures above 90°C the top temperature of the storage is kept at this operational limit of 90°C.

The variations of the network temperature lead to a dynamic storage capacity, that is described by the e_max_pu parameter/attribute of the Store component. According to Equation (1) in Sorknaes (2018), the dynamic storage capacity linearly depends on the deltaT between top and bottom temperature. The DEA catalogue indicates a deltaT of 55 K (top temperature of 90 and bottom temperature of 35 °C) as reference for the model. These reference values can be varied in the config settings sector:district_heating:ptes:max_top_temperature and sector:district_heating:ptes:min_bottom_temperature . As the spread between forward and return flow is smaller for a majority of systems, especially in summer, the effective storage capacity is smaller than the optimal value for the decision variable e_nom_opt.

2050, all countries (39 nodes), 8H resolution

The dynamic variation of the pit storage capacity can be turned off by setting the config parameter sector:district_heating:ptes:dynamic_capacity to false. The comparison of the SOC between a storage with a static and a dynamic storage capacity from an aggregated European perspective shows, that in the latter case the maximum SOC does never come close to the nominal value of the energetic capacity. It remains below 50%.

2050, all countries (39 nodes), 8H resolution

The changes in the storage operation are also characterized by a higher amount of cycles. Overall, the reduction in average PTES capacity makes it less attractive, leading to a reduction in PTES investments and an increase in total system costs.

	System Costs [bn€/yr]	PTES Capacity [TWh]	PTES Charger Capacity [GW]	PTES Cycles [1/yr]
Static	668.998	136.687	911.246	4.0134
Dynamic	674.144	97.2345	648.23	5.93258
Difference [%]	0.769166	-28.8633	-28.8633	47.8193

Checklist

• I tested my contribution locally and it works as intended.
• Code and workflow changes are sufficiently documented.
• Changed dependencies are added to envs/environment.yaml.
• Changes in configuration options are added in config/config.default.yaml.
• Changes in configuration options are documented in doc/configtables/*.csv.
• Sources of newly added data are documented in doc/data_sources.rst.
• A release note doc/release_notes.rst is added.

[amos-schledorn]

Great feature and very useful plots! Striking by how much the storage capacity drops in summer.
Not sure I understand the time series plots. What does e_max_py (weighted) on the secondary y-axis mean exactly?

Code-wise, I would suggest to

• move this to a separate rule that either returns flat ones or the dynamic e_max_pu
• make 90/35C config settings or technology-data entries
• rename the first e_max_pu in the if-clause to delta_t or similar (or just remove it and use e_max_pu = (t_top - t_bottom) / (t_top_nom - t_bottom_nom)
• add the Sorknaes paper and other relevant sources as references

Could you also add release notes and update the docs?

[cpschau]

Great feature and very useful plots! Striking by how much the storage capacity drops in summer. Not sure I understand the time series plots. What does e_max_py (weighted) on the secondary y-axis mean exactly?

Code-wise, I would suggest to

* move this to a separate rule that either returns flat ones or the dynamic `e_max_pu`

* make 90/35C config settings or `technology-data` entries

* rename the first `e_max_pu` in the if-clause to `delta_t` or similar (or just remove it and use `e_max_pu = (t_top - t_bottom) / (t_top_nom  - t_bottom_nom)`

* add the Sorknaes paper and other relevant sources as references

Could you also add release notes and update the docs?

I addressed all of your points @amos-schledorn. In the plot you asked about, the e_max_pu timeseries of the pit components are weighted by their optimal capacity values to aggregate them to a value between 0 and 1.

[amos-schledorn]

Looks really good! I just have a few minor comments. Crazy by how much p_max_pu is reduced - we should look into validating DH supply temperatures.

[fneum]

Looks great! Just a few style comments.

[amos-schledorn]

Thanks for the changes! Can be merged once Fabian's comments are addressed, imo.