NEM (Re)Bidding Analysis, 2012-2021

(Re)bidding analysis of the Australian National Electricity Market (NEM).

This repository is a companion to the pre-dispatch forecast analyses in this respository.

Results

Bidding zip file size, 2012-2021

Monthly bidding data zip file sizes have increased exponentially since the middle of 2021. See this plot.

Share of rebids by technology type in June, 2013-2021

This plot look at the share of rebids across the years 2013-2021 by technology type.

Examples of battery energy storage system bidding behaviour

This plot shows how BESS participate in a risk-averse (strictly, loss-averse) manner in the energy market.

Usage

Installation

Install poetry and run the following install this repository to create a poetry virtual environment:

poetry install

Running analysis and plotting scripts

This repository uses a Makefile to automate a series of Python scripts used for the analysis.

Processed data has already been committed to this repository. To plot the results in the plots directory, run:

make create_plots

If you wish to change what is being analysed (e.g. different years, different months), you will need to process the data yourself.

To get bidding zip file size data, run the following:

make create_data_for_bid_zip_file_plot

To get rebid data with a certain ahead time for a particular month across years, run the command below.

The scripts within this pipeline use polars to manage memory, but it will still need a machine with 20-25 GB RAM.

make create_data_for_rebid_plots

Tooling

Analysis in this repository uses NEMOSIS, NEMSEER, mms-monthly-cli and nem-bidding-dashboard.

Rapid analysis of bidding data was made possible by using polars.

Author & licenses

This repository contains work by Abhijith (Abi) Prakash, PhD Candidate at the UNSW Collaboration on Energy and Environmental Markets. If you are interested in extending or using this work, please get in touch. The source code is currently light on documentation.

The source code from this work is licensed under the terms of GNU GPL-3.0-or-later licences. It includes modified source code made available by Dylan McConnell under the terms of GNU GPL-3.0-or-later.

The results (generated plots) are licensed under a Creative Commons Attribution 4.0 International License.

Acknowledgements

Nicholas Gorman for assistance in developing pipeline to analyse bidding data