These half-hourly benchmarks have been created for different building-use types on campus-style sites. The benchmarks are produced as a result of the development of a consistent approach to assessing decarbonisation for scaled delivery. It provides a standard set of benchmarks that can be used to split aggregated site-energy consumption into a building-by-building profile for representative days of the year. The benchmarks have been created from data made available to Energy Systems Catapult whilst working with various departments within the public sector.

This dataset is compiled using two sets of benchmarks. Firstly, benchmarks created during the early stages of the Modern Energy Partners programme, based on available fiscal data at the time, and secondly, benchmarks created at the end of the project using energy consumption submeter data from sites.

The original benchmarks were created using energy consumption data from fiscal meters which were manually collated from sites. The data range in general covered the calendar year 2019, however, where multiple years of historic data were available, average consumption over the period was used. Electricity and gas consumption data from sites was available as half-hourly profiles. To create the benchmarks, the methodology employed followed the site energy demand estimator guidance, which is available on the Public Sector Decarbonisation Guidance webpages. 

The new benchmarks provided (which are highlighted on the Data Sources tab for reference) have been generated as a result of a programme of data capture from new and existing energy consumption submeters installed on sites. Benchmarks have been created through the compilation of multiple submeters on various campus-style sites. Data capture has been done automatically and covers a period from set-up/installation (varying by site, but generally from summer 2021) through to the end of February 2022. Therefore it was necessary to generate proxy values for the missing 'spring' months. This has been done using the actual consumption data from the 'winter' season and scaling using the heating degree day (HDD) values for the seasons based on the location of the sites used (refer to https://www.degreedays.net for further information). Winter data has been selected as a proxy for spring as the heating profiles and HDD for the period were seen to be similar across all site locations considered.

Data for the available seasons was assessed for quality, ensuring that there are no erroneous readings and the consumption profile is as expected. Representative seasonal half-hourly consumption profiles have been generated as an average of consumption across the available days in the season per meter/building and use case.