Go to contentGo to menuGo to searchGo to the news list

Laboratoire Angevin de Recherche en Ingénierie des Systèmes

Main navigation




    User-centric optimisation and predictive control approaches of the performance of smart buildings

    • Share this page on social networks
    • E-mail this page

      Send by mail

      Separated by coma
    • Print this page

    Research Project BIoT - Building & Internet of Things

    User-centric optimisation and predictive control approaches of the performance of smart buildings



    Team: Reliability Engineering and Decision-Making tools



    Term: 3 years (01/01/2020 - 31/12/2022)


    Funding: AAP Attractivité du RFI WiSE


    LARIS staff involved: Marie-Lise Pannier, David Bigaud,Chuhao Jiang (Phd student)

    Project partner : 



    The technical dimension is no longer the only indicator of the building performance during its use phase. Building Management System (BMS) have already been integrated in buildings, but a new vision is required to redefine the building performance. The concept of smart building relies on this new vision, and does not only cover the building technical dimensions but also the human dimension, including all building actors and occupants.

    In addition, the concept of Cognitive Building Management (CBM) is a promising application field of Internet of Things (IoT). A building will be equipped with sensing and actuating devices to monitor it (e.g. the windows opening, the indoor air quality, the user presence, the position), to predict the energy loads and to measure real-time energy consumptions. All data collected by sensors must them be analysed using specific algorithms in order to improve the user experience and comfort, to optimise the building management and to reduce costs.

    Research problem:

    The PhD takes place in the frame of the BIoT (Building Internet of Things) project founded by the RFI Wise within its Attractivity call for proposal. The BIoT project involves four mains topics: the integration of cognitive (and statistical) occupation model, the modelling of occupancy and use based on deep learning, the simulation and optimisation of the building performance, and the control of the building systems and components using predictive control model.

    The main focus of the PhD will be on the occupation modelling using deep learning techniques (agent-based modelling and reinforcement learning) and on optimal predictive control model.

    - Stochastic and robust modelling of the occupancy and the building use, using deep learning methods: although an increasing number of studies are carried out on the user behaviour, the role of the occupants remains rarely considered during the building operation phase. This can be explained by the lack of available model to represent the occupancy in a realistic way. The identified occupancy patterns are often insufficiently dynamic (e.g. the patterns often have a temporal dynamic, but not a spatial dynamic). If this problem can easily be solved in the case of residential buildings with a small number of occupants and building zones, it is more difficult to solve it for more complex building having different uses and where different profiles meets and may interact.

    Occupancy scenarios are strongly dependant of the season, the weather, the time of the day, the user preferences, its individual and collective behaviour (e.g. single person or family in residential buildings, individual office or open-spaces in office building)… At least, stochastic models (e.g. Markov chains) should be used to create realistic scenarios taking the temporal variability into account. In the PhD, machine learning methods or reinforcement methods will be applied to get more robust occupation model that can take uncertainties into account. The LARIS laboratory has expertise in the use and development of deep learning methods such as decision trees, artificial neural networks, or Bayesian network (that have recently been applied in order to identify deviation in the building performance). We would like to explore the concept of multi-agent modelling, which is new for the laboratory, in order to predict occupants’ behaviour as well as their habits and preference regarding thermal, air, visual and acoustic comforts.

    The occupancy and use model will be implemented in a digital twin of a building, coupled to building dynamic energy simulations, in order to calibrate the energy model more precisely and to have a more precise basis for the optimisation studies. The models will also be used in the digital twin for the monitoring and control in real time of the building energy systems.


    - Control of the building systems using predictive control model: the second topic within BIoT is the user-centric control of the building systems. Occupants play an important role in the building performance and have strong interactions with the building. Their needs and actions should therefore be taken into account and the user-centric approach appears to be a viable alternative to an automated control approach. After having performed learning technique on previous observations, the user behaviours and habits are well known and the building techniques equipment (HVAC, lighting, blinds, …) can them be control in real time, (depending on these known users preferences), using a set of sensing and actuating devices.

    Two research themes are linked with the user-centric control. The first theme concerns the definition of optimal control strategies, according to the occupants’ activities, and their applications to the building in real time. The second theme concerns the predictive control model based on the occupant’s habits, the forecasted meteorological data and the building dynamic to predefine the next optimal strategy.


    Required skills and experience:

    The aim of the PhD is to model the building user behaviours in order to improve the monitoring and control of the buildings global performance (including thermal, air and visual comfort). The modelling based on data gathered from connected sensors and devices. We are seeking a candidate with a background in building energy and skills in data processing and optimisation (e.g. agent-based and multi-objectives optimisation), as data mining and machine learning techniques will be used to develop occupation models and to predict the occupancy and activities within the building. The candidate will have to install sensors on the Polytech Angers building; basics knowledges of instrumentation would therefore be appreciated.

    The candidate holds a Research Master Degree or is a graduate engineer with a research experience.

    Host laboratory:

    The work will be carried out in the LARIS Laboratory (Systems Engineering Research Laboratory of Angers) at Polytech Angers – University of Angers. The LARIS (EA7315) includes three research teams, focusing on: dynamic systems and optimisation; information, signal and image processing and life science; and the dependability and decision aids. The staff includes 55 teacher-researchers (27 of them have the accreditation to supervise research activities), and 32 PhD candidates.

    The PhD will be conducted in the area “evaluation model for the operational performance of complex systems” of the research team on dependability and decision aids. One of the scientific issues of the team involves guarantying and maintaining the performance of complex systems during the operation phase. The works of the team aim at developing monitoring and forecasting strategies, based on models or data, to be integrated in complex decision-making process. The built environment is one of the application fields for 15 years.