heat-demand-peaks

The project, commissioned by the EEE consortium and supported by Open Energy Transition (OET), aims to assess the impact of various energy efficiency measures on the European energy system. This study focuses on energy affordability, social and household impacts, and industry considerations in the context of the EU's target of 90-95% emissions reduction by 2040. Utilizing the PyPSA-Eur integrated energy system planning tool, the project evaluates the effectiveness of different renovation scenarios, energy management measures, and demand-side flexibility measures in reducing energy generation needs, flattening the peak demand curve, and influencing energy prices. The results provide insights into the benefits of isolated and combined efficiency measures, contributing to data-driven decision-making in energy policy and planning. The project commenced in January 2024 and is expected to conclude by July 2024.

The study is developed as an extension to the PyPSA-Eur model in this repository.

Repository Structure

• submodules contains the relevant submodules PyPSA-Eur and technology-data.
• configs contains all relevant configuration files. in EEE_study, there are all relevant config files for this study, while Zeyen_etal contains config files to reproduce previously published results; config.plot.yaml contains network parameters for plotting and network modifications.
• plots contains some scripts to generate a few plots for model evaluation purposes
• scripts contains scripts that modify existing results in order to create new scenarios

Installation and Usage

1. Installation

Clone the repository including its submodules:

git clone --recurse-submodules https://github.com/open-energy-transition/heat-demand-peaks

Install the necessary dependencies using conda or mamba:

mamba env create -f submodules/pypsa-eur/envs/environment.yaml

Activate pypsa-eur environment:

conda activate pypsa-eur

Navigate into the main Snakemake workflow directory of PyPSA-Eur:

cd submodules/pypsa-eur

2. Running scenarios

Before running the scenarios, it is important to know their short working names used in the code. The table below provides the mapping between official scenario names and correspoding short names utilized in the code. To run the scenarios, refer to the coding names.

Scenario name	Coding name
Optimal Retrofitting & Optimal Heating (OROH)	flexible
Optimal Retrofitting & Green Heating (ORGH)	retro_tes
Limited Retrofitting & Optimal Heating (LROH)	flexible-moderate
No Retrofitting & Green Heating (NRGH)	rigid
Business as Usual (BAU)	BAU

Note! Running the scenarios requires a high-performance computing environment, as well as a Gurobi license.

A. Running scenarios using automated workflow (the easy way)

To run all scenarios for all horizons, run:

python scripts/run.py 

To run the simulations for specific scenario for several horizons at once (e.g. Optimal Renovation & Optimal Heating (OROH) scenario), run:

python scripts/run.py -s flexible -c 2040

where -s is a mandatory flag used for the scenario selection. Use the coding name of corresponding scenario from the table to trigger the execution. -c flag (optional) speficies from which planning horizon the simulations should start. So if -c 2040 is given, then 2040 and 2050 horizons will be simulated consequitively; if not specified, then 2030 is used as a starting year by default. -c flag is useful when simulation is interupted and it needs to be re-run from certain horizon. For BAU scenario, using only -s BAU without horizon definition is sufficient.

To simulate a single horizon for the scenario (e.g. Limited Retrofitting & Optimal Heating scenario for 2050), use -y flag as follows:

python scripts/run.py -s flexible-moderate -y 2050

• Note! To run Limited Renovation & Optimal Heating (flexible-moderate) scenario, the simulation results for Optimal Renovation & Optimal Heating (flexible) scenario must be present.

The available flags and their details are presented in table below:

Flag	Default	Description	Status
-s, --scenario	n/a	Selects scenario	required
-c, --continue	2030	Selects horizon from which simulation starts (used in simulating multiple horizons). If 2030 is selected, then 2030, 2040, and 2050 is simulated consecutively. If 2040 is selected, then 2040 and 2050 is simulated.	optional
-y, --year	n/a	Selects a single horizon to be simulated. If both -c and -y are prodived occasionally, then priority is given to -y.	optional
-i, --improved_cop	true	Enables/disables improved COP workflow. By default, improved COP for heat pumps is used.	optional

2.2. Running scenarios manually using snakemake (the hard way)

A. Run the scenario

To run the scenario of a particular configuration file (e.g. configs/EEE_study/config.flexible-industry.yaml), navigate to pypsa-eur directory using:

cd submodules/pypsa-eur

Then, run the following comamnd to prepare the pre-network:

snakemake -call prepare_sector_networks --configfile ../../configs/EEE_study/config.flexible-industry_2030.yaml

To solve the network, run:

snakemake -call solve_sector_networks --configfile ../../configs/EEE_study/config.flexible-industry_2030.yaml 

Please follow the documentation of PyPSA-Eur for more details.

B. Setting nominal capacities of retrofitting for Limited Renovation & Optimal Heating (LROH) scenario

The nominal capacities of retrofitting for Limited Renovation & Optimal Heating (LROH) (flexible-moderate) scenario is set by running:

snakemake -call set_moderate_retrofitting

• Note! This and the following snakemake commands must be run in heat-demands-peak base directory (not pypsa-eur submodule). The command needs to be run after preparation of pre-network.

This command will set p_nom for moderate retrofitting network as a half of p_nom_opt of solved Optimal Renovation and Heating (flexible) scenario. The network parameters, such as clusters and planning_horizon, are defined in moderate_retrofitting section of configs/config.plot.yaml.

As an alternative, the nominal capacities for moderate retrofitting scenario can be set by running the command with wildcards (e.g. scenario for 2030 with 48 clusters):

snakemake -call scripts/logs/set_moderate_retrofitting_48_2030.txt --force

The resultant file in scripts/logs/set_moderate_retrofitting_48_2030.txt contains Success parameter which indicates the success status of executed rule. Here, --force flag is used to forcefully re-execute the rule.

C. Transfering optimal capacities to future horizons

After optimizing scenarios for one horizon, it is important to transfer optimal generation and store capacities into future horizons. To do so, configure planning_horizon in set_capacities section of configs/config.plot.yaml to horizon of interest. By default, 2040 is set as planning_horizon in set_capacities, which helps to transfer p_nom_opt values from solved networks of 2030 into p_nom_min of corresponding generators and stores of unsolved network of 2040. The optimal capacities are transfered to corresponding scenarios. To set p_nom_min for all scenarios of 2040, run:

snakemake -call set_capacities

• Note! The command needs to be run after preparation of pre-network, but before solving it.

To set minimum capacities for specific scenario (e.g. flexible scenario of 2050 with 48 clusters), you can run run:

snakemake -call scripts/logs/set_capacities_48_2050_flexible.txt

The resultant file in scripts/logs/set_capacities_48_2050_flexible.txt contains Success parameter which indicates the success status of executed rule.

D. Utilizing improved COP for heat pumps in scenarios with building retrofitting

To use the workflow of improved COP, solve the network regularly and determine the heat saved ratio by building retrofitting from the solved network. Use the ratio to compute sink temperature:

heat_pump_sink_T = (55 - 21) * (1 - heat_saved_ratio) + 21

Update the heat_pump_sink_T parameter in corresponding configuration file. Then, prepare the pre-network and set capacities from previous horizon. To set retrofitting capacities from the first run to current run, execute:

snakemake -call improve_cops_after_renovation

Finally, the updated pre-network can be solved.

3. Plotting

To visualize the simulation results, the following list of snakemake rules can be used.

Rule name	Description
plot_total_costs	Plots total costs and optimal capacities of technologies and provides the results in a tabular form
plot_electricity_bills	Plots electricity bills per household and energy price per MWh
plot_electricity_for_heats	Plots electricity and gas consumption profiles to cover heat demands
plot_electricity_generations	Plots electricity generation mix
plot_curtailments	Plots total curtailment and provides detailed curtailment for renewables in a tabular form
plot_hydrogen_productions	Plots hydrogen generation and electricity used for its production and provides the results in a tabular form
plot_heat_tech_ratios	Plots optimal capacities (GWel) of heat generation technologies, such as gas boilers, heat pumps, and resistive heaters
plot_COP	Plots coefficient of performance (COP) for heat pumps
plot_co2_levels	Plots CO2 balances and provides the results in a tabular form
plot_historic_generation	Plots historic electricity generation mix for 2022 from [1]
plot_transmission_congestions	Plots transmission lines congestion
plot_all	Plots all abovementioned figures
get_heat_pumps	Provodes estimated heat pumps amount in a tabualr form
get_infra_savings	Provides infrastructure savings interms of costs and capacities in a tabular form

[1] How is EU electricity produced and sold? URL: https://www.consilium.europa.eu/en/infographics/how-is-eu-electricity-produced-and-sold/

To demonstrate how to use the plotting rules to visualize the total system cost for various planning horizons (e.g., 2030, 2040, and 2050) as specified in the config.plot.yaml file, run the following command:

snakemake -call plot_total_costs

Note! The total costs will be plotted only for scenarios where the solved networks are present. --forceall flag can be used to regenerate and overwrite the already existing plots.

As another example, to estimate number of heat pumps and generate table for all horizons and clusters specified in config.plot.yaml, run:

snakemake -call get_heat_pumps

To generate all aforementioned plots and tables at once, run:

snakemake -call plot_all --forceall

Note! --forceall flag is used to force a rerun of rule.