Routines for assessing demand response potentials

This repository contains routines for assessing demand response potentials. These comprise:

• The outcome of a meta analysis on technical potentials and
• Routines for creating parameter sets using a clustering approach

The final result is a parameterization for demand response which can be used to model it in the fundamental power market model POMMES.

Demand response technical potential meta analysis

This repository contains all files from a demand response meta-analysis for Germany in which 37 publications have been evaluated on demand response parameter information (technical potentials).

Demand response potential clustering

The results from the meta-analysis as well as some availability information and assumptions are used to define demand response clusters, i.e. a grouping of demand response categories, mostly at the level of applications resp. processes.

Contents and structure

The repository is structured as follows:

• "main level": The main level contains input data as well as all routines for extracting, manipulating, visualizing and storing demand response parameter data.
• "out": The folder out is the place where all results are stored. It contains some subfolders: availability, plots, stats, parameterization as well as sources. Availability input data for the clustering is already given as external input. The other input data for the clustering is stored as well but can be easily created from running the potential evaluation notebook as it is.+
• "inputs": Input files for the analyses. The most important ones are:
  • "Methodenvergleich_Potenzialstudien.xlsx": An in-depth method comparison of the analyses conducted in MS Excel
  • "Potenziale_Lastmanagement.xlsx": The raw data for the potential comparison
  • "Columns.xlsx": Information on the parameter column names used
  • "Zitationen.xlsx": A citation analysis
• "drpotentials": Python tools used in the analyses, among others
  • "clustering_funcs.py": User-defined functions which are imported by the notebook for the potential clustering
  • "evaluation_funcs.py": User-defined functions which are imported by the notebook for the meta-analysis
  • "evaluation_workflow.py": Python-wrappers bundling functionality for the meta-analysis of technical demand response potentials

At the main level, the following files exist:

• "DR_availability_timeseries_generation.ipynb": Data preprocessing routines for the clustering, collecting availability data
• "DR_citation_analysis.ipynb": A cross-study citation analysis
• "DR_potential_evaluation.ipynb": The notebook used for the potential meta-analysis
• "DR_potential_clustering.ipynb"; The notebook used for the potential clustering and creating input to be passed to pommesdata
• "DR_potential_comparison.ipynb"; A notebook for assessing the results compared to literature

Usage

• Install the dependencies specified, see pommesdata's requirements
• Start the main jupyter-notebooks "DR_potential_evaluation.ipynb" or "DR_potential_clustering.ipynb" or any of the others.
• Manipulate the parameter settings if necessary (i.e., the boolean parameters controlling what is done).
• Run the notebook and obtain the data output demanded.

Development

Feel free to manipulate the data according to your needs. If you want to contribute, please create a fork, open a new branch and issue a pull request if you want to have your changes integrated. Let me now, if you need other permissions you do not yet have.

For further information on git usage, please refer to the following sources:

• Quick and dirty main git commands: https://rogerdudler.github.io/git-guide/index.html
• Official git documentation: https://git-scm.com/docs
• Gitlab basic help: https://docs.gitlab.com/ee/gitlab-basics/README.html
• A very good and detailled book called ProGit: https://progit2.s3.amazonaws.com/en/2016-03-22-f3531/progit-en.1084.pdf