A collection of command-line tools (and Python functions) that can be used to work with data from the Amber Electric API.
Amber Electric is an innovative energy retailer in Australia which gives customers access to the wholesale energy price as determined by the National Energy Market. This gives customers the opportunity to reduce their bills and their reliance on fossil fuels by shifting their biggest energy usage to times of the day when energy is cheaper and greener.
Amber gives customers access to a LOT of their own data through their public Application Programming Interface or API.
This tool relies on you having access to Amber's API, which means you need to be an Amber customer, and you need to get an API token. But that's pretty easy. Start here.
If you're a programmer comfortable with Git, I'm sure you already know how to get this code onto your machine from GitHub.
If you're not familiar with Git, you can download this code as a Zip file by clicking on this link. Once it's downloaded, unzip the file, which will create a directory containing all the files of this project.
-
You'll need to know how to use a command-line interface / terminal / PowerShell.
-
You'll need an Amber API token. (See above)
-
You'll need Python 3.9 (or higher) installed.
If you think you already have Python 3.9+ installed, you can check the version of Python in your path by running:
python3 -V
If that doesn't work, try:
python -V
If neither of those works, or neither is Python 3.9 or higher, you'll need to download and install Python.
Using a terminal, in the directory of this project:
- Create a Python virtual environment with this command. (Note: if 'python3' didn't work above but 'python' did and is v3.9 or higher, just use 'python'):
python3 -m venv venv
- Install the required dependencies with this command on MacOS or Linux:
venv/bin/pip install -r requirements.txt
or this command on Windows:
venv\Scripts\pip install -r requirements.txt
The following options are common among all the tools.
Run the script with the -h option to see its help page.
On MacOS / Linux:
venv/bin/python SCRIPT_NAME -h
On Windows:
venv\Scripts\python SCRIPT_NAME -h
All scripts require an Amber API token to be provided.
There is an --apitoken/-t option available in each script.
If you'd prefer not to paste your API token into a terminal command, you can
save it in a file called apitoken (with no extension) in the project's
directory.
All scripts will look for this file and read the API token from it if it exists,
removing the need for the --apitoken option.
If you have multiple sites in your Amber Electric account, you'll need to select
one using the --site-id option:
venv/bin/python SCRIPT_NAME --site-id SITE_ID_YOU_WANT_DATA_FOR
This is a command line tool that produces a summary CSV report of an Amber Electric customer's energy consumption and cost data.
You simply need to provide your Amber API token, and the tool will output a CSV like this for the last 12 months:
CHANNEL , 2020-09-01, 2020-09-02, 2020-09-03, ...
B4 (FEED_IN) Usage (kWh) , 1.351, 0.463, 0.447, ...
E3 (CONTROLLED_LOAD) Usage (kWh), 2.009, 2.669, 2.757, ...
E4 (GENERAL) Usage (kWh) , 20.400, 20.965, 16.011, ...
Using a terminal, in the directory of this project:
Run the tool with this command, replacing YOUR_API_TOKEN with your own API
token.
MacOS / Linux:
venv/bin/python amber_usage_summary.py --api-token YOUR_API_TOKEN > my_amber_usage_data.csv
Windows:
venv\Scripts\python amber_usage_summary.py --api-token YOUR_API_TOKEN > my_amber_usage_data.csv
Using the above, your summary consumption data for the last year will be saved
to the file called my_amber_usage_data.csv in the same directory.
By default, the tool just outputs energy consumption data.
If you also want a summary of your cost data, add the --include-cost option:
venv/bin/python amber_usage_summary.py --include-cost
By default, the report includes the last 12 full calendar months of data, plus all of the current month's data up until yesterday. You can select what date range to include in the output by adding and start date and, optionally, an end date to the command.
venv/bin/python amber_usage_summary.py 2020-07-01 2021-06-30
This is a command line tool that produces a detailed CSV report of an Amber Electric customer's historical spot prices.
You simply need to provide your Amber API token, and the tool will output a CSV like this for the last month:
DATE +10:00, CHANNEL , 00:00:00, 00:30:00, 01:00:00, ...
2022-01-21 , GENERAL (c/kWh) , 17.456, 17.743, 17.736, ...
2022-01-21 , CONTROLLED_LOAD (c/kWh), 14.366, 14.653, 14.646, ...
2022-01-21 , FEED_IN (c/kWh) , -5.913, -6.174, -6.168, ...
2022-01-22 , GENERAL (c/kWh) , 118.627, 17.623, 18.280, ...
...
Using a terminal, in the directory of this project:
Run the tool with this command, replacing YOUR_API_TOKEN with your own API
token.
On MacOS / Linux:
venv/bin/python amber_spot_price_export.py --api-token YOUR_API_TOKEN > my_amber_spot_price_data.csv
On Windows:
venv\Scripts\python amber_spot_price_export.py --api-token YOUR_API_TOKEN > my_amber_spot_price_data.csv
Using the above, your spot price data for the last month will be saved
to the file called my_amber_spot_price_data.csv in the same directory.
By default, the report includes the last month of data. You can select what date range to include in the output by adding a start date and, optionally, an end date to the command.
venv/bin/python amber_spot_price_export.py 2020-07-01 2021-06-30
This is a command line tool that produces a CSV summary of an Amber Electric customer's historical solar exports, including total exports, earnings, and average and peak outputs by month.
You simply need to provide your Amber API token, and the tool will output a CSV like this for the last year:
, 2022-05, 2022-06, 2022-07, ...
Total kWh , 0.137, 6.171, 16.988, ...
Total Income $ , 0.000, 0.230, 0.840, ...
Average Daily kWh, 0.034, 0.213, 0.566, ...
Peak Daily kWh , 0.074, 0.738, 2.161, ...
Peak Period kW , 0.116, 0.438, 0.656, ...
Using a terminal, in the directory of this project:
Run the tool with this command, replacing YOUR_API_TOKEN with your own API
token.
On MacOS / Linux:
venv/bin/python amber_solar_export_summary.py --api-token YOUR_API_TOKEN > my_amber_solar_export_data.csv
On Windows:
venv\Scripts\python amber_solar_export_summary.py --api-token YOUR_API_TOKEN > my_amber_solar_export_data.csv
Using the above, your solar export summary data for the last year will be saved
to the file called my_amber_solar_export_data.csv in the same directory.
By default, the report includes the last 12 full months of data. You can select what date range to include in the output by adding a start month and, optionally, an end month to the command.
venv/bin/python amber_solar_export_summary.py 2020-07 2021-06
This is a command line tool that produces a text-based estimate of an Amber Electric customer's invoice for a particular month or months in the past, using usage and pricing data from Amber's API.
Once configured, the tool can output an invoice breakdown like this one for any past whole month during which you were an Amber customer:
Month: 2022-01
Usage Fees:
General Usage Wholesale 319.9 8.10 $ 25.91
Controlled Load Wholesale 39.9 8.07 $ 3.22
Network - Peak Energy 35.2 25.37 $ 8.93
...
The output has been designed to closely match the format of Amber's most recent invoice format.
By running the tool multiple times for the same time period but using different tariffs, it's possible to compare the estimates for what you would have paid under different tariffs for previous months.
I'm in no way attached to Amber Electric (except for being a customer). This tool has been developed as an amateur, non-vocational exercise effort. Amber doesn't endorse this tool, nor support it.
The tool deliberately has "Estimate" in the name because it is NOT, and cannot be, a 100% accurate re-creation of Amber's billing engine that creates your actual invoices.
The purpose for creating this tool was to enable a rough comparison of different DNSP tariffs against historical usage data, and in my own use it appears to be accurate enough to serve this purpose.
While no warranty is given about any aspect of the tools in this repository, I want to specifically call out that I'm offering ZERO guarantees that the output of this tool is accurate. Amber themselves have published information about some of the reasons that make reproducing bills from API data problematic. And in relation to this tool they've told me:
Billing is complicated, and there are a bunch of business rules that you probably haven’t modelled, so the output may not match a bill exactly.
If you find inaccuracies in the output, and know why they are happening, I'd be happy to hear from you in order to improve the tool (and even happier to have you develop and submit a patch (after chatting with me about it)). See 'Contributions' below.
The following limitations are known to exist with the tool:
- There is no handling for block tariffs.
- There is no handling for capacity pricing.
- The outputs of demand pricing have not been tested against real bills.
- There is no provision for mid-month changes, e.g. if a tariff changes or GreenPower is switched on or off in the middle of a month.
- There is no special handling for doing part of the current month (per-day charges will be calculated for the full month).
- Only Ausgrid residential tariffs have been encoded (and only non-closed ones).
- Only NSW charges have been encoded.
- Only public holidays in NSW from 2021-2023 have been comprehensively encoded.
There are likely other limitations that are unknown.
Re-producing a bill requires a lot more data than what is available from the API. At a high level, the tariffs you want to use for each channel, the prices that make up those tariffs, and other charges specific to your state, your network, and even your site's location, all need to be available and relatively accurate to make a good estimate.
So far, I've only entered this data for scenarios that I've wanted to test. This means if you are in NSW, on the Ausgrid network, have an IntelliHub smart meter, and are connected to the Sydney South TNI, the tool might produce good estimates for you! If you're not part of that small niche, you will probably need to do some data acquisition and entry in order to get the tool estimating your own bills.
If you add new data for your own purposes, and you'd like to contribute it for others to use, I'd be happy to hear from you. See 'Contributions' below.
Before using this tool, you'll need to configure it to have the correct tariffs and other data for the location and scenario you want to test.
At the very least, you'll need to either edit or create your own copy
(recommended) of data/accountConfigs/my_account_config_example.json5, and
review/update all the values in the file to ensure they're accurate for your
site.
If you're using tariffs that aren't already encoded in a tariff file, you'll
need to create a new tariff file.
The document data/tariffs/TariffsReadme.md explains the format for creating
a tariff file.
It will probably be easiest to copy an existing one for a similar type of
tariff (e.g. time of use, demand, controlled load) and modify it rather than
start from scratch. But it shouldn't be too hard.
See below for how to find the data to go in these files.
Lastly, if there isn't an "Other Charges" file for your location under
data/otherCharges, or it's out of date, you'll need to create one of those.
This file has a similar format to tariff files with a few small differences as
explained at the top of the NSW file.
There are links in the NSW file that should help in finding the right values.
Note that, except for the Amber monthly fee, all other tariffs and charges should be entered as exclusive of GST.
The first thing you'll need to do is figure out who your Distributed Network Service Provider (DNSP) is. This is sometimes just called your "network" as is the company who controls the poles and wires in your ares. It is not the same as your retailer (that would be Amber!).
If you're planning to use these tools, then the easiest way to find out who
runs your network will be to
perform a /sites query using the Amber API docs.
The name of your DNSP appears in the response as the property network.
Good news: The /sites query response (see above) also contains the codes of
your current tariffs!
Bad news: Tariffs are usually made up of a number of different rates/prices that apply to different metrics (e.g. energy used, or days in a month) and at different times (e.g. peak vs shoulder vs off-peak, or different months of the year).
Good news: Tariffs are often available online. If you google for your DNSP's name, "tariffs", and the current year, you should usually come up with a document that describes your tariffs.
Here's an incomplete list of known tariff documents. If you find another one, or one that points to an old link, it'd be great if you could add it and submit a pull request to this README!
Amber have kindly said they'll probably be happy to supply people with extra data they may need to run this tool if they contact their support team at [email protected].
They've also said that they often have busy periods and that in those times requests like this might take a long time for a response, or could even be refused.
I would expect the only data your might need to get from Amber would be:
marginalLossFactorandsmartMeterAccessChargeCentsPerDayfor your Account Config file- the six
centsPerKwhitems for your state's Other Charges file
I believe most other rates should be able to be easily found on the internet. Let's try not to bother Amber with requests for data that we can easily google. Taking the time to submit a pull request to this project with any data you get from Amber or from tariff documents that could be useful to other people should help greatly in reducing that load.
Using a terminal, in the directory of this project:
Run the tool with this command, replacing YOUR_API_TOKEN with your own API
token, and the account config filename with the path of your own account config.
On MacOS / Linux:
venv/bin/python amber_invoice_estimate.py --api-token YOUR_API_TOKEN data/accountConfigs/my_account_config_example.json5
On Windows:
venv\Scripts\python amber_invoice_estimate.py --api-token YOUR_API_TOKEN data\accountConfigs\my_account_config_example.json5
By default, the report generates an estimate of your bill for the last complete
month.
You can select what months to include in the output by listing one or more
months as arguments to the script in yyyy-MM format:
venv/bin/python amber_invoice_estimate.py ACCOUNT_CONFIG_FILE 2021-10 2021-11 2021-12
I'm open to accepting contributions that improve the tool.
If you're planning on altering the code with the intention of contributing the changes back, it'd be great to have a chat about it first to check we're on the same page about how the improvement might be added. It's probably easiest to create an issue describing your planned improvement (and being clear that you plan to implement it yourself).
All files in this project are licensed under the 3-clause BSD License. See LICENSE.md for details.