PV Bushfire Data

This data set ranges from 2017-11 to 2020-09 in 30-minute sampling rate. The data set includes 710 residential sites across Australia.

Database Stricture & Data

Two tables were created for this dataset:

1. pv_bushfire_site_details: contains the site details similar to the csv file, hence the following columns:

   • site_id (big integer)

   • postcode (character (4))

   • state (character varying (3))

   • timezone_id (character varying)

2. pv_bushfire_data: contains all the PV energy data from all sites and times. The original data was in UTC; we localized timestamps using the time zone of each site and added a new column, datetime, to store the local time. It is referred to when accessing and filtering data via the webserver. The table has the following columns:

   • utc_timestamp (timestamp with time zone)

   • datetime (timestamp without time zone)

   • site_id (big integer)

   • monitoring_data_type (character varying)

   • value (real)

The volume of this dataset is 2.7 GB (one month energy around 100 MB).

Appendix A: Original Dataset

This data set ranges from 2017-11 to 2020-09 in 30-minute sampling rate.

The data set includes 710 residential sites across Australia from the Solar Analytics data set.

Each month's data was provided in a separate csv file. There were 35 data files in total. In the data csv files, each column represents:

• utc_timestamp: the time stamp in UTC

• site_id: the ID of this site

• monitoring_data_type: the data type of this monitoring

• values: the PV energy produced during the 30 minutes in w- att-hour

There was a site_details csv file as well, each row representing the details of one site as follows:

• site_id: the ID of this site

• postcode: the postcode of this site

• state: the state of this site

• timezone_id: the local time zone of this site. This information will be useful when converting the UTC time stamp to local time stamp.