Aristotle University Of Thessaloniki (AUTH)
Προσωποποιημένα μοντέλα προσδιορισμού εσωτερικής έντασης φωτισμού με χρήση τεχνικών υπολογιστικής ευφυΐας
Abstract
dc:descriptionThe scope of this thesis is to study the interaction of each individual with the artificial lighting and shading systems in an office and to develop techniques for modeling this behavior, predicting the daily actions of the user and finally determining the optimal illuminance on the work-plane of each user depending on external conditions and his unique preferences. The stochastic nature of human reactions and the ever-changing daylight conditions must be quantified, combined and forecasted. Thus, it is necessary to model the behavior of each user and to develop appropriate tools in order to achieve accurate, personalized and energy-efficient control of lighting and shading systems. Initially, the effect of external conditions and user actions on the internal illuminance is studied using illuminance measurements conducted in three university offices. These data are utilized in the formulation of models which capture the probability for each user to control the artificial lights or the blinds in order to create a comfort environment in his office. This data-driven modeling results in behavioral curves which provide the probability for each user to act, depending on his work-plane illuminance. In addition, two new separate models are developed and evaluated based on Artificial Neural Networks (ANN) that predict the work-plane illuminance, taking into account the daylight entering the workspace and the special requirements of each user. The illuminance curve due to the external daylight is predicted for each day of the year, depending on the sky type (clear-overcast), the building location and the office orientation. This prediction also takes into account the distance from the windows and the usage of blinds. Finally, a data-driven fuzzy-based model is provided, that implements the concept of user comfort in an office space, based on illuminance measurements and modeling of occupant behavior. The scope of this model is to determine the current optimal value of work plane illuminance under the respective external conditions, so that visual comfort is achieved for each user. This personalized model can also increase the user acceptability when integrated in a smart lighting and shading control system.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2020
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Katsanou, Varvara
- Κατσανού, Βαρβάρα
Subjects
dc:subject × 12- Προσωποιημένος έλεγχος φωτισμού
- Οπτική άνεση
- Συμπεριφορά χρηστών
- Υπολογιστική ευφυΐα
- Personalized lighting control
- Visual comfort
- User behavior
- Machine learning
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Engineering and Technology
- Electrical Engineering, Electronic Engineering, Information Engineering
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/48291