University of Patras
Ανάπτυξη προσαρμοστικών συστημάτων για εξοικονόμηση ενέργειας σε κτίρια με εφαρμογή τεχνολογιών διαδικτύου των αντικειμένων (IoT) και μεθοδολογιών πληθοπορισμού
Abstract
dc:descriptionThis work examines issues related to the application of Internet of Things (IoT) technologies, with the aim of developing sensor systems (WSN) and crowdsourcing methodologies, to save energy in adaptive buildings. Modern Building Energy Management Systems (BEMS) are based on measurements and signals from sensors that are statically installed in a building and based on some scenarios developed by user to control the operation of the building's devices. All these are managed from a central system installed in the building. Wireless sensor networks are a rapidly evolving technology that is widely used in buildings as well. In a smart building, tiny sensors, fixed or moving, are integrated into all the devices that humans use in their daily lives, creating a distributed system. These sensors can communicate wirelessly with each other, as well as with an external network over the Internet, turning conventional devices into smart ones, and integrating them into the Internet of Things (IoT). In this way they communicate with any other device on the internet, making their management by users easier and more dynamic, while there can even be an initiative of their own actions. However, in addition to the sensors that need to be integrated in the building for monitoring, crowdsourcing technologies can also be used. More specifically, crowdsensing methods are studied, where used sensors for measurements that are not installed in the building specifically for the BEMS system, but are available on devices used by users in their daily lives. For example, a user's mobile phone, especially a smartphone, has a lot of capabilities both because of its sensors and because of its computing and storage features. This enables us to develop applications for smartphones that will provide valuable information about the user's location, environmental conditions close to the user, his body condition (sleeping, lying down, sitting, walking, running, driving, etc.), his interactions with other users, the occupancy of other people in the place where the user is, etc. Such applications could find significant application in places where there are many users, and provide as much information as the system needs without the need to install permanent sensors. For example, at an airport, users' mobile phones could provide information on the temperature in each room to adjust the air conditioning system locally. Such crowdsensing applications can be developed in many cases, but there are several challenges such as: how to endorse the participation of users, how to properly manage the battery and other resources on the user's device, how to ensure users' privacy, etc. As part of this dissertation, we designed prototype building automations that are part of the Internet of Things (IoT) and use wireless sensor networks with IPv6 technologies, while being flexible and compatible with a wide range of building facilities. The automations developed include wireless communication using the 6LoWPAN protocol, based on the IEEE 802.15.4 standard and the IPv6 protocol. Also, using the CoAP protocol, they follow a RESTful architecture that makes access from the outside world very easy with GET/PUT requests. The automations we designed were distinguished in the international IPSO innovation competition, where their important aspects were recognized with the new technologies they incorporate. In the smart room of the laboratory where the automations were installed, we developed combined scenarios that control lighting, air conditioning, ventilation, etc., and we managed to save at least 40% energy, while the thermal and visual comfort of the users was maintained at very satisfactory levels. Also, also been integrated virtual light sensors that we developed, which are based on algorithmic methods that estimate the brightness in places where there are no sensors, using the light measurements from sensors in other points of the room. Then we studied the human factor in building environments. We developed methods for fine-grained indoor localization of a user, based on communication threshold effect between two nodes in a wireless sensor network. Then, we developed feature extraction methods to identify a user's activity and environment by applying machine learning algorithms with data from their smartphone. Finally, we studied the federation of buildings that their facilities and the IoT equipment follow different architectures. In the platform that was developed, users can also participate (crowdsourcing) through a mobile application on their smartphone, in order to feed the system with their opinion (subjective input), but also to provide measurements from the sensors of their smartphones (crowdsensing). In the laboratory, scenarios for the lighting control were implemented with active participation of crowd and different incentives policies for the users were studied, to encourage their participation.
Degree
thesis:*- Grantor dc:publisher
- University of Patras
- Year dc:date
- 2020
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Filios, Gavriil
- Φίλιος, Γαβριήλ
Subjects
dc:subject × 22- Διαδίκτυο των αντικειμένων
- Κτιριακοί αυτοματισμοί
- Προσαρμοστικά κτίρια
- Τεχνολογίες πληθοπορισμού
- Αναγνώριση δραστηριότητας ανθρώπου
- Εικονικοί αισθητήρες
- Συστήματα διαχείρισης κτιρίων
- Εξοικονόμηση ενέργειας
- Internet of things
- Building automation
- Adaptive buildings
- Crowdsourcing
- Feature extraction
- Virtual sensors
- Building management systems
- Energy saving
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Επιστήμες Μηχανικού και Τεχνολογία
- Αυτοματισμοί και Συστήματα ελέγχου
- Electrical Engineering, Electronic Engineering, Information Engineering
- Engineering and Technology
- Automation and control systems
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/49963