Detection of energy anomalies - Datacity

Detection of energy anomalies

Home > Challenges > Detection of energy anomalies



How can sound signal analysis technologies be used to assist energy managers in the detection of anomalies in facilities?

The technology that equips modern buildings creates new opportunities to control energy consumption, improve comfort and manage physical spaces and security​. Most buildings, however, are not equipped with this technology and have only basic electricity, water and gas meters. The aim of this challenge was to leverage data to develop solutions to make these “dumb” buildings intelligent, without changing their equipment.



  • Use case

  • Data sets and experimentation field

  • Business opportunities

  • Results

The use case


Paris City Hall, together with certain partnering companies, have access to buildings equipped with sensors. This opportunity for experimentation enables them to implement an approach that mixes instrumentation, modeling and data analysis.

The data sets and experimentation fields


In the scope of the challenge, the following experimentation from the partnering company Setec, has been selected:

– Setec headquarters in Paris


The following data set is provided by the partners:

– Central meter box (electricity, gas, water)

The business opportunities


The business of energy managers is extremely small as they require buildings that are sufficiently equipped with sensors and companies able to process the data


– The solution may help energy managers to handle a bigger client portfolio
– The solution can be scaled easily
– The solution may lead to automatic detection of anomalies

For Setec: part of their business is energy management ==> they can leverage a better position in this way.

Results of the experimentation


This study was done through the modeling of three buildings having complementary uses: a nursery (Crèche Druinot, in Paris’ 12th arrondissement), a service building from Nexity and a residential building).




During the bootcamp, Sensewaves:

– Interviewed different end-users​ to understand their specific needs

– Used Setec’s HQ electricity data to test the reliability of their modeling by crossing the results with complementary data and expertise of the building manager and Setec professionals




Sensewaves was able to validate the relevance of their model ​and technology, noting that:

– Many problems indirectly linked to electricity consumption were instantaneously identified without recourse to supplementary data

– Their solution minimized the time to identify and analyze problems, as well as the risk of overlooking them
– 100% of the behavior anomalies and consumption problems were identified by the system and validated by the Setec Energy Managers.

– The scale of “seriousness” proposed by the system was judged 75% reliable.

– The problems are detected in real time, at the moment they appear: the manager can be notified instantly.




Since the end of DataCity’s 1st season, Sensewaves has carried on their collaboration with Setec by integrating Sensewaves’ technology into Setec’s Building Management System (BMS). For Setec, this is a perfect opportunity to add a new expertise to their business line​, and modernize their service offering​. This was a game-changer for Sensewaves which is now replicating ​this collaboration in Denmark ​and Portugal​, and will launch a proper new offer for all BMS players beginning in 2017.

Partner's Experts

Cyril Sailly

Director and Founder

Anne-Gaelle Baptiste

Heritage and Architecture Division, Energy Transition and Innovation Officer
Paris City Council