Skip to content
English
Sign in
Access to metron.energy
  • There are no suggestions because the search field is empty.

1-EnergyLab: the energy data science studio

Introduction: What is EnergyLab?

EnergyLab is a tool that enables energy professionals to carry out data analyses, from the simplest to the most complex, without any prior knowledge of Data Science.
It enables you to explore your data in depth, so you can go even further in your analyses. 
With EnergyLab, you can create your own machine learning models, to model phenomena (consumption, vibrations, flow rates...) in a reference situation. 

The aim is to know at all times whether or not your real-time data matches the model, and therefore to know at all times whether or not you are drifting away from the reference situation. 

A reference model can thus be used to :

  • easily detect consumption drift,

This type of drift detection model generally takes into account all known parameters, in order to control the impact of unknown parameters. This makes it possible to react in real time to hazards that could have a negative impact on production. 

  • quantify savings achieved over a period,
  • quantify losses due to an incident.

These 2 reference models for quantifying savings or losses take into account parameters that are “fixed” or “under control” (i.e. not within the operator's control). This makes it possible to determine the impact of control on performance. 


The models created with EnergyLab are based on automatic learning algorithms (linear regression, Random Forest, kNN, etc.).

Access to EnergyLab

If you have access to EnergyLab on your platform, you can access it via the navigation bar on the left of your screen:
EL1

If you do not yet have access to EnergyLab, please contact your METRON project manager.

Creating a baseline with EnergyLab

There are 2 ways to create baselines with EnergyLab:

  • Automatic baseline
This is the quickest and most efficient way of modeling a reference situation, based on consumption data and influential parameters.

  • Customized baseline
The customized baseline is the most precise and adaptable way of defining and modeling a reference situation. It allows you to explore the data to best define the reference period and influential factors.

To find out more, see Creating a baseline on EnergyLab.