CoLoC - Comfort of connected housings - Research Project

Boost plan for development of individualised predictive models of thermal comfort for connected housing

Group : Reliability Engineering and Decision-Making tools

Labelling: none

Duration: 2 years (2022-2024)

Funding: ANR

Staff involved from LARIS: Marie-Lise Pannier, David Rousseau, Valentin Hoyet (Engineer on project)

Project partners: Maxime Robillart (Kocliko)

Abstract:

Thermal comfort depends not only on the energy performance of buildings, but also on the sensitivity of individuals. Currently, thermal comfort is taken into account in a global way when developing energy management strategies for buildings. The goal is to satisfy a majority of occupants, without integrating the subjectivity of the perception of comfort and the feelings of the users. In order to improve the consideration of comfort, the CoLoC project aims to develop individualised predictive models of thermal comfort for connected housing.

Firstly, a bibliographical and technical study on how to evaluate thermal comfort and the feelings of occupants will be carried out. Based on the results of this state of the art, review a database combining objective data (ambient measurements) and subjective data (users' feelings) will be developed in a second phase. Some dwellings and residences, already instrumented by Kocliko, will be chosen to deploy the tools for collecting the residents' feelings. The database will be consolidated by dealing with the problems of missing data. Finally, the collected data will be analysed using statistical tools and machine learning methods to develop comfort models and be able to anticipate discomfort situations.

For more information on the company Kocliko