In this article, you will learn how to use the Bulk Baseline Generator (BBG), available in the Apps Market upon request.
Contents of this article:
Introduction
This application allows you to calculate baselines for multiple locations under a deployment (account) using the same multiple linear regression method that the ABC uses and apply them to create an M&V Project for each.
Requirements
For the BBG to work, we need to have a minimum of configuration properly set in our account.
- Reference meters: for electricity, gas, thermal, water…
- Degree days data (HDD, CDD): by having a reference meter for outdoor temperature (weather data) and the location configured with summer and winter reference temperatures
How does it work?
The baselines are created by automatically retrieving the following data:
- Energy consumption data for the selected energy source and reference device (Electricity, Gas, Thermal…)
- Degree days data (HDD, CDD) linked to each location
And with the user's inputs for:
-
Mandatory data:
- Resolution: D, W, M
- Use Day, Month as variables
- Reference period
- Demonstration period
- Target
- Error
- Project type (Energy Management, Lighting…)
- Energy Sources/Ref.Devices for which to calculate baselines
-
Nice to have:
- Squared parameters / Interactions on/off
- Period to discard
- R2 threshold
Example
In the image below we show an example of a BBG execution. Here we select Electricity as an Energy Source, and the Main Supply as the reference device type that will be taken per each location under our hierarchy.
The reference and demonstrative periods will be as seen below by default, but we can change them to suit our project. The frequency and project type can also be selected here. The user can select a target and uncertainty, in the same way that it is currently done for M&V projects.
Advanced options give the possibility to use extra inputs for the baseline such as squared parameters, between parameters or to use the weekend, weekday or the month in order to define the baseline. This allows the creation of baselines even if there is no historical weather data and Degree Day data is not accessible, as long as there is a year of historical energy consumption data.
In the same way as ABC, the user can discard periods for the baseline calculation, allowing to ignore periods like COVID lockdown.
The KPI used to evaluate the generated baselines will be the R2adj which is the default used in ABC. The user can define a threshold for that KPI, so that any baseline calculated that has an R2adj below the threshold is ignored and no projects are generated for those. In the example below, a minimum of 0.6 R2adj is necessary for the project to be created.
The locations for which M&V Projects are generated are all the locations under the current hierarchy level. So if the user is sitting at the top of the hierarchy (Global level), the new feature will attempt to calculate a baseline and M&V Project for every location in the account. Tags and hierarchy filters apply, so customers using those features will be able to target specific locations to create their baselines.
Every time that the process is launched, a new execution is stored, with the timestamp corresponding to the moment it started.
In order to access our generated Projects, we can find them directly under "M&V Projects".
We find all of our created projects from the BBG named as:
- [location name]_BULK_[frequency]_[ref devicetype]_[date and time of creation]
We can delete those that do not have enough data, or that we consider not representing the location. Any projects created with that execution can be deleted with 2 clicks. This allows the users to remove any projects created by mistake in bulk.
Pricing
This application is free with enterprise licenses upon request.