Aristotle University Of Thessaloniki (AUTH)
Βελτιστοποιημένη λειτουργία συστημάτων ηλιοπροστασίας σε συνδυασμό με θερμοστατικό έλεγχο στα κτήρια
Abstract
dc:descriptionThis work suggests a new methodology of a whole building simulation-based control of building shading systems. The thermostat control is also considered synchronously to study the effect of adding new control systems. The problem is focusing on the users’ viewpoint by describing a “user satisfaction” function, which is attempted to be optimized. The suggested approach allows users to choose their comfort factors with their limits and the allowed energy spent to provide view to the outside. Users can grade all shading positions regarding view and the suggested approach seeks for minimizing energy consumption and maximizing view to the outside taking notice of visual and thermal comfort satisfaction. The problem described as an optimization problem and new tools has been built and new functionality to current tools has been added. The initial typical approach used optimization algorithms such as the simplex method of Nelder and Mead, the pattern search method of Hooke and Jeeves, the genetic algorithm, the Particle Swarm Optimization and their hybrid and parallel versions. Results show that such algorithms cannot achieve satisfactory performance, because of the needed long runtime, which restricts real-time use of them. The provided solution produced an innovative optimization strategy by adding domain expert knowledge to the problem. This strategy cut the problem into parts, which can be solved synchronously with great performance. The suggested strategy outperforms the typical approach both on runtime and results reliability. This work produced several new tools to the research community. New optimization algorithms and packages have been added to GenOpt, such as the ECJ and JGAP. The programming code of the genetic algorithm of JGAP has been altered to improve parallel performance. The pattern search of Hooke and Jeeves has been modified to a synchronously parallel version. A hybrid version of the genetic algorithm and the pattern search has been produced. Code to couple GenOpt with Radiance has been provided as well. A new methodology of shading systems control has been produced. This methodology has included thermostat control as well and can be extended to other building systems. Results gathered and provided some simple rules for shading control, which follows the logic of the “user satisfaction” approach. A new viewpoint of building design and control optimization has been provided and showed the directions to future building optimization tools.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2015
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Machairas, Vasileios
- Μαχαιράς, Βασίλειος
Subjects
dc:subject × 19- Λειτουργία συστημάτων ηλιοπροστασίας
- Λειτουργία θερμοστάτη
- Λειτουργία σκιάστρων
- Βελτιστοποίηση
- Αλγόριθμοι βελτιστοποίησης
- Μέθοδοι βελτιστοποίησης
- Shading
- Solar slats
- Solar blinds
- Thermostat control
- Control
- Optimization
- Optimization algorithms
- Optimization methods
- Simulation based control
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Πολιτικού Μηχανικού
- Engineering and Technology
- Civil Engineering
Rights
dc:rights- Statement dc:rights
-
- BY_NC_SA
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/36624