Advanced Machine Learning Technologies for Energy Efficient Buildings

ADVANCED MACHINE LEARNING TECHNOLOGIES FOR ENERGY EFFICIENT BUILDINGS Réf ABG-103050 Stage master 2 / Ingénieur Durée 5 mois Salaire net mensuel 000 11/02/2022 CESI/LINEACT/IRDL Lieu de travail Brest Bretagne France Champs scientifiques * Sciences de l’ingénieur * Electronique * Informatique Mots clés Building energy systems, occupancy behavior, IoT, machine learning, and intelligent controllers. Date limite de candidature 15/03/2022 ÉTABLISSEMENT RECRUTEUR SITE WEB : https://lineact.cesi.fr/ LINEACT (EA 7527 ) - Digital Innovation Laboratory for Business and Apprenticeships for Regional Competitiveness is CESI's research laboratory, the activities of which are implemented on thirteen sites or campuses spread across France, grouped into six regions, as well as within a national setting at the General Management based in Paris La Défense. The specificity of LINEACT lies in the organization of its research into two interdisciplinary scientific themes and two applicative sectors, focusing on the major and increasingly fast changes in the industry and the city. The "Learning and Innovation" theme is primarily related to Cognitive Sciences, Humanities and Social Sciences, and Management Sciences, while the "Engineering and Digital Tools" theme is primarily related to Digital Sciences and Engineering. These two themes develop and cross their research in the two application sectors of the Industry of the Future and the City of the Future, and, more broadly, support companies and learners in the revolutions that our societies are currently undergoing. DESCRIPTION Buildings, in general, consume a significant amount of energy during their operating phase. A deep understanding of energy usage in buildings is essential to manage and control the buildings to achieve energy-efficient buildings [1]. Integration of renewable energies leads to reducing the overall energy consumption. However, the flexibility of these energies is limited, thus, it is essential to adapt these optimally while managing indoor comfort. Due to the availability of low-cost sensors and high-capacity digital storage systems, the use of digital technologies and IoTs has increased significantly in the recent decade [2]. It is now possible to collect and process all relevant data about the occupants, the building, and the environment, and exploit this big data to simulate building performance and comfort conditions in real-time [3]. Occupant behavior is primarily influenced by indoor conditions and cognitive emotions; they interact with buildings by adjusting system set-points, making automatic control challenging for building managers and stakeholders [4]. Therefore, it is necessary to use this big data to facilitate occupancy-centric automatic controlling that optimizes energy use, while satisfying the indoor comfort requirements in buildings by using advanced machine-learning techniques [5 - 7]. The main objectives of this work are to study the new possibilities for the use of machine-learning-based occupancy-centric building control techniques and to address implementation challenges. This internship can lead to a 3-year Ph.D. thesis with the funding. they interact with buildings by adjusting system set-points, making automatic control challenging for building managers and stakeholders [4]. Therefore, it is necessary to use this big data to facilitate occupancy-centric automatic controlling that optimizes energy use, while satisfying the indoor comfort requirements in buildings by using advanced machine-learning techniques [5 - 7]. The main objectives of this work are to study the new possibilities for the use of machine-learning-based occupancy-centric building control techniques and to address implementation challenges. This internship can lead to a 3-year Ph.D. thesis with the funding. they interact with buildings by adjusting system set-points, making automatic control challenging for building managers and stakeholders [4]. Therefore, it is necessary to use this big data to facilitate occupancy-centric automatic controlling that optimizes energy use, while satisfying the indoor comfort requirements in buildings by using advanced machine-learning techniques [5 - 7]. The main objectives of this work are to study the new possibilities for the use of machine-learning-based occupancy-centric building control techniques and to address implementation challenges. This internship can lead to a 3-year Ph.D. thesis with the funding. it is necessary to use this big data to facilitate occupancy-centric automatic controlling that optimizes energy use, while satisfying the indoor comfort requirements in buildings by using advanced machine-learning techniques [5 - 7]. The main objectives of this work are to study the new possibilities for the use of machine-learning-based occupancy-centric building control techniques and to address implementation challenges. This internship can lead to a 3-year Ph.D. thesis with the funding. it is necessary to use this big data to facilitate occupancy-centric automatic controlling that optimizes energy use, while satisfying the indoor comfort requirements in buildings by using advanced machine-learning techniques [5 - 7]. The main objectives of this work are to study the new possibilities for the use of machine-learning-based occupancy-centric building control techniques and to address implementation challenges. This internship can lead to a 3-year Ph.D. thesis with the funding. PROFIL Skills: - The applicant must be from a Mechanical or Electrical or Computer Science engineering background. - Any relevant experience in building physics or HVAC systems would be a plus. - Experience in data mining, analysis, and management is preferred. - Experience in programming environments (eg. Python, R, etc). PRISE DE FONCTION Dès que possible