How do you calculate the CO2e emissions from a unit of energy used?
Energy retailers and the government produce conversion factors that describe the typical carbon impact of different energy sources. These allow us to take the energy uses (in their respective units), and calculate the approximate carbon dioxide emissions, normally measured in kilograms of carbon dioxide equivalents (kgCO2e). Defra's UK conversion factors may be found at Defra's 2018 Guidelines.
How do you get these data from the buildings?
Getting these energy data out of some buildings is harder than others, but in general the buildings contain a small low-power computer which takes very frequent readings from the electricity meters and stores the data. Every few seconds, this computer sends the information it has collected to a server. Your browser will then ask this server for the data it needs in order to draw the real-time detailed graphs and website teasers. The energy impact of this process is very low, and it gets lower with each additional site that uses the system.
What do the colours on the graph mean?
For buildings, the colours in the graph show approximately how the current level of usage would lead to a given Operational Rating – as set out on a Display Energy Certificate (DEC) – if the performance for a given moment carried on for an entire year. This goes from dark green for ‘A’ to red for ‘G’. We calibrate this using input data used for generating the building’s DEC, together with information relating to 'normal' buildings of its type. If we do not have data for all of the utilities noted in the DEC then the graph will appear in a light-blue colour scale, to indicate that the usage displayed on the graph is not representative of the full energy use of this building. Graphs for communities also show in this blue colour scale.
Why are you using these units and what do they mean?
We provide three different measures of the energy used: the amount of energy, its monetary cost, and the carbon impact of the energy used.
Energy use is measured in kilowatt hours (kWh), which are the standard units of a home energy bill (1kWh is the amount of electricity used by ten 100W light bulbs in one hour).
For electricity this number represents the amount of energy that flows into a building through the meter, and excludes distribution losses. For gas it is the amount of energy that is theoretically available by burning all the gas in an imaginary ideal burner. For district heating it reflects a flow of temperature into the building over time (after the heat produced by burning the fuel has been transported to the meter, which involves other losses). So each of these numbers, while all being measured in kWh, mean very different things. This is one reason that we prefer to use 'units per hour' when combining them. In some ways it would be more correct not to combine them at all, because combining them implies that the measures are comparable. This is a global challenge though, and conventions have become established around combining kWh. So we'll have to fix that another day.
Monetary cost is calculated using the costs per 'unit' for each utility in every building. The figures used are noted below in the Notes section.
The carbon impact is measured in kg of CO2e (the e stands for equivalent) which takes other climate-affecting gasses into account in addition to carbon dioxide.
How much does this organisation pay for its energy?
Prices come from the latest energy bills for FCO, which for Gas average out at 3.27p per unit and for district heating average out at 2.9p per unit and for electricity average out at 7.74p per unit. The gas volumetric measurement is converted to kWh using the meter correction factors and calorific values supplied by the utility company. These may be subject to change.
Can you show data from the transport emissions of this organisation/ building?
Data of CO2e emissions created by transport used by organisations is very interesting and powerful data to show here. We are working on ways to display and reduce the transport impacts of different organisations, and you will see some of the products of this work on these pages very soon.
How are real-time data displayed?
As far as the widget is concerned, there are 12 distinct 5-second periods in a minute. The real-time data is for the five second period just ended, which means that sometimes the widget could display values that are nearly 10 seconds old.
Because we have used the pulse-outputs of electrical and gas meters there are certain assumptions we need to make in order to generate a real-time value. The pulses from the meters actually signify a volume of gas or an amount of energy, and we need to determine a flow of gas or electrical power from the pulses. Since at any given moment you are always between one pulse and the next, you have to guess, to a certain extent, when the next pulse will come in order to estimate the actual flow or power. The accuracy of the guess depends on how close together the pulses are, so at busy times you hardly need to guess at all. The pulses from the main electricity meter come every 100 watt-hours, which is enough to let us overcome the issue by counting the number of pulses in a five second period and applying a calculation to smooth successive readings into a rate, even at night. The main gas meter pulses once for each cubic metre consumed, which for properly variable loads could leave us guessing for quite a while before the next pulse comes. We use an ‘exponential moving average’ calculation to generate the real-time value, which allows us to display a value that is as close-to-right as possible; the values go up in time with increases in actual use, but lag behind sudden reductions. The downside of this is that if you added up all the real-time values that the teaser shows every five seconds, over time it would be shown to over-report slightly. This inaccuracy in the real-time data is strongest when high gas use drops quickly (as when the main boilers shut down, which happens several times a day). This distortion in the real-time data does not introduce any inaccuracies into the archive data, and we will report on the exact degree of error introduced if there is interest in this.
Are these data complete?
The figures currently being displayed for district heating do not represent the total amount of energy consumed for heating this building. This is due to some technical issues, which mean that not all of the heating data can be collected at this time. We are looking into this and aim to display the correct amounts soon.