PV_Live

A Python implementation of the PV_Live web API. See https://www.solar.sheffield.ac.uk/pvlive/

Latest Version: 1.3.0

About this repository

• This Python library provides a convenient interface for the PV_Live web API to facilitate accessing PV_Live results in Python code.
• Developed and tested with Python 3.10, should work with Python 3.7+. Support for Python 2.7+ has been discontinued as of 2021-01-15.

How do I get set up?

• Make sure you have Git installed - Download Git
• Run either:
  • pip install pvlive-api
  • pip install git+https://github.com/SheffieldSolar/PV_Live-API

Usage

There are three methods for extracting raw data from the PV_Live API:

Method	Description	Docs Link
PVLive.latest(entity_type="pes", entity_id=0, extra_fields="", period=30, dataframe=False)	Get the latest PV_Live generation result from the API.	🔗
PVLive.at_time(dt, entity_type="pes", entity_id=0, extra_fields="", period=30, dataframe=False)	Get the PV_Live generation result for a given time from the API.	🔗
PVLive.between(start, end, entity_type="pes", entity_id=0, extra_fields="", period=30, dataframe=False)	Get the PV_Live generation result for a given time interval from the API.	🔗

There are two methods for extracting derived statistics:

Method	Description	Docs Link
PVLive.day_peak(d, entity_type="pes", entity_id=0, extra_fields="", period=30, dataframe=False)	Get the peak PV_Live generation result for a given day from the API.	🔗
PVLive.day_energy(d, entity_type="pes", entity_id=0)	Get the cumulative PV generation for a given day from the API.	🔗

These methods include the following optional parameters:

Parameter	Usage
entity_type	Choose between "pes" or "gsp". If querying for national data, this parameter can be set to either value (or left to it's default value) since setting entity_id to 0 will always return national data.
entity_id	Set entity_id=0 (the default value) to return nationally aggregated data. If entity_type="pes", specify a pes_id to retrieve data for, else if entity_id="gsp", specify a gsp_id. For a full list of GSP and PES IDs, refer to the lookup table hosted on National Grid ESO's data portal here.
extra_fields	Use this to extract additional fields from the API such as installedcapacity_mwp. For a full list of available fields, see the PV_Live API Docs.
period	Set the desired temporal resolution (in minutes) for PV outturn estimates. Options are 30 (default) or 5.
dataframe	Set dataframe=True and the results will be returned as a Pandas DataFrame object which is generally much easier to work with. The columns of the DataFrame will be pes_id or gsp_id, datetime_gmt, generation_mw, plus any extra fields specified.

There is also a method for extracting PV deployment (a.k.a capacity) data:

Method	Description	Docs Link
PVLive.deployment(region="gsp", include_history=False, by_system_size=False, release=0)	Download PV deployment datasets from the API.	🔗

Code Examples

See pvlive_api_demo.py for more example usage.

The examples below assume you have imported the PVLive class and created a local instance called pvl:

from datetime import datetime
import pytz

from pvlive_api import PVLive

pvl = PVLive()

Example	Code	Example Output
Get the latest nationally aggregated GB PV outturn	pvl.latest()	(0, '2021-01-20T11:00:00Z', 203.0)
Get the latest aggregated outturn for PES region 23 (Yorkshire)	pvl.latest(entity_id=23)	(23, '2021-01-20T14:00:00Z', 5.8833031)
Get the latest aggregated outturn for GSP ID 120 (INDQ1 or "Indian Queens")	pvl.latest(entity_type="gsp", entity_id=120)	(120, '2021-01-20T14:00:00Z', 1, 3.05604)
Get the nationally aggregated GB PV outturn for all of 2020 as a DataFrame	pvl.between(start=datetime(2020, 1, 1, 0, 30, tzinfo=pytz.utc), end=datetime(2021, 1, 1, tzinfo=pytz.utc), dataframe=True)
Get a list of GSP IDs	pvl.gsp_ids	array([ 0, 1, 2, 3, ..., 336, 337, 338])
Get a list of PES IDs	pvl.pes_ids	array([ 0, 1, 2, 3, ..., 336, 337, 338])

To download data for all GSPs, use something like:

def download_pvlive_by_gsp(start, end, include_national=True, extra_fields=""):
    data = None
    pvl = PVLive()
    min_gsp_id = 0 if include_national else 1
    for gsp_id in pvl.gsp_ids:
        if gsp_id < min_gsp_id:
            continue
        data_ = pvl.between(start=start, end=end, entity_type="gsp", entity_id=gsp_id,
                            dataframe=True, extra_fields=extra_fields)
        if data is None:
            data = data_
        else:
            data = pd.concat((data, data_), ignore_index=True)
    return data

Command Line Utilities

pv_live

This utility can be used to download data to a CSV file:

>> pv_live -h
usage: pv_live [-h] [-s "<yyyy-mm-dd HH:MM:SS>"] [-e "<yyyy-mm-dd HH:MM:SS>"]
               [--entity_type <entity_type>] [--entity_id <entity_id>]
               [--extra_fields <field1[,field2, ...]>] [--period <5|30>] [-q]
               [-o </path/to/output/file>] [-http <http_proxy>] [-https <https_proxy>]

This is a command line interface (CLI) for the PV_Live API module

options:
  -h, --help            show this help message and exit
  -s "<yyyy-mm-dd HH:MM:SS>", --start "<yyyy-mm-dd HH:MM:SS>"
                        Specify a UTC start date in 'yyyy-mm-dd HH:MM:SS' format
                        (inclusive), default behaviour is to retrieve the latest outturn.
  -e "<yyyy-mm-dd HH:MM:SS>", --end "<yyyy-mm-dd HH:MM:SS>"
                        Specify a UTC end date in 'yyyy-mm-dd HH:MM:SS' format (inclusive),
                        default behaviour is to retrieve the latest outturn.
  --entity_type <entity_type>
                        Specify an entity type, either 'gsp' or 'pes'. Default is 'gsp'.
  --entity_id <entity_id>
                        Specify an entity ID, default is 0 (i.e. national).
  --extra_fields <field1[,field2, ...]>
                        Specify an extra_fields (as a comma-separated list to include when
                        requesting data from the API, defaults to 'installedcapacity_mwp'.
  --period <5|30>       Desired temporal resolution (in minutes) for PV outturn estimates.
                        Default is 30.
  -q, --quiet           Specify to not print anything to stdout.
  -o </path/to/output/file>, --outfile </path/to/output/file>
                        Specify a CSV file to write results to.
  -http <http_proxy>, --http-proxy <http_proxy>
                        HTTP Proxy address
  -https <https_proxy>, --https-proxy <https_proxy>
                        HTTPS Proxy address

Jamie Taylor & Ethan Jones, 2018-06-04

Using the Docker Image

There is also a Docker Image hosted on Docker Hub which can be used to download data from the PV_Live API with minimal setup:

>> docker run -it --rm sheffieldsolar/pv_live-api:<release> pv_live -h

Documentation

• https://sheffieldsolar.github.io/PV_Live-API/

How do I upgrade?

Sheffield Solar will endeavour to update this library in sync with the PV_Live API and ensure the latest version of this library always supports the latest version of the PV_Live API, but cannot guarantee this. To make sure you are forewarned of upcoming changes to the API, you should email [email protected] and request to be added to the PV_Live user mailing list.

To upgrade the code:

• Run pip install --upgrade pvlive-api

Notes on PV_Live GB national update cycle

Users should be aware that Sheffield Solar computes continuous retrospective updates to PV_Live outturn estimates i.e. we regularly re-calculate outturn estimates retrospectively and these updated estimates are reflected immediately in the data delivered via the API.

As of 2023-05-02, the first estimate of the PV outturn for a given settlement period is computed ~5 minutes after the end of the half hour in question and is typically available via the API within 6 minutes

e.g. the initial estimate of the GB PV outturn for the period 14:30 - 15:00 UTC on 2023-05-02 (i.e. 15:30 - 16:00 BST) would become available at ~16:06 BST on 2023-05-02 and will be labelled using the timestamp at the end of the interval (in UTC): '2023-05-02 15:00:00'

This outturn estimate is not final though - Sheffield Solar will continue to make retrospective revisions as

• more PV sample data becomes available
• better PV deployment data becomes available
• the PV_Live methodology is refined

Since the PV_Live model produces outturn estimates, they will never strictly speaking be final, as there will always be things we can do to refine the model and improve accuracy. That said, there are some notable/routine retrospective revisions to be aware of:

• Outturns are re-computed in near-real-time every 5 minutes after the end of the half-hour for 3 hours, to allow for late arriving sample data
  • If all data ingestion pipelines are running smoothly, this does not result in any retrospective revisions
  • If some near-real-time sample data is late arriving, the outturn estimate will be revised at the next update
• Outturns are re-computed on day+1 (typically between 10:30 and 10:35 and again between 22:30 and 22:35) to make use of sample data which only becomes available on day+1
• Historical outturns may be re-computed periodically whenever our PV deployment dataset is updated retrospectively (usually every 3 - 6 months)

In order to maintain a local copy of the PV_Live GB national outturn estimates that is as in sync with our own latest/best estimates as possible, we recommend the following polling cycle:

• Every 5 minutes, pull the last 3 hours of outturns from the API
• Every day, around 11am, re-download the previous 3 days of outturns
• Every month, re-download all historical outturns

Who do I talk to?

• Jamie Taylor - [email protected] - SheffieldSolar

Authors

• Jamie Taylor - SheffieldSolar
• Ethan Jones - SheffieldSolar

License

No license is defined yet - use at your own risk.