jingzhountu / bench_ev_agg_disagg

Code of the paper "Benchmarking Aggregation-Disaggregation Pipelines for Smart Charging of Electric Vehicles"

This repository contains all the implementations of the EV charging event aggregation and disaggregation methods and the benchmarking setup used in the paper "Benchmarking Aggregation-Disaggregation Pipelines for Smart Charging."

Prerequisites

• Python
• rustup (https://www.rust-lang.org/tools/install)
• maturin (pip install maturin)
• gurobi (https://www.gurobi.com/downloads/)
• Dataset "Mobilität in Deutschland 2017" (https://www.mobilitaet-in-deutschland.de/archive/index.html)
  • Alternatively, you can supply any data that can be converted into the Agent class in python_code/profile_generator.py. In that case, you need to replace the function prepare_run() in python_code/utils.py with a version that takes your data as input. Feel free to contact me if there are any issues with this process.
• Electricity price, generation, and demand data for Germany 2022 in 15 min resolution from https://www.smard.de/en/downloadcenter/download-market-data/ (Warning: Use the english version of the website!).
  • Alternative: Only use the SINE price signal in benchmark.run()

Running the benchmark

1. Change the paths in python_code/config.py to direct to the MID and price data.
2. Run maturin develop --release
3. Run python parameter_tuning.py for parameter tuning.
4. Run python benchmark.py for benchmark results.

The results will be located in results/.

Once you run the two Python scripts, you can use the notebook Visualize.ipynb to recreate the figures from the paper.

Project Structure

• The Rust implementations are found under src/.

• The raw Python implementations are found under python_code/raw_python_impls/.

• All benchmark along with all helper scripts is contained in python_code/.