|GERONAZZO ANGELA||Cycle: XXX |
Section: Computer Science and Engineering
Tutor: SCIUTO DONATELLA Major Research topic
:Energy efficient buildings: data integration, analysis and exploitation for smart buildings and smart users
Advisor: BOLCHINI CRISTIANAAbstract:
In recent years, improving energy efficiency in buildings has emerged as an important societal issue and research area, motivated by a pressing quest to design, develop and implement effective and affordable energy demand reduction strategies. The ultimate goal is to optimize the trade-off between energy consumption and the occupants¿ comfort, aiming at reducing the high-energy demand and carbon footprint without compromising the users' quality of life. In this perspective, current research trends focus on ICT for energy efficient communities and cities. Buildings (particularly existing, non-residential ones) and occupants are the two key elements of the stated problem, to be opportunely monitored, controlled and made aware in order to pursue the desired optimization goal. Therefore, ICT is used to collect large amount of data and information from the field (e.g., building characteristics, environmental conditions, users' behavior) to be able to: i) define a model of the system, ii) extract from the available, multi-facet data the information that represents the system, iii) simulate how interventions can impact energy efficiency iv) define management strategies to optimize resource usage while maintaining a satisfying comfort. Each one of the identified actions is itself an interesting research challenge, which involves numerous themes. Our interest is mainly focused on data model, management and exploitation towards the achievement of energy efficiency in smart communities and smart cities. One of the first challenges consists in the modeling and management of all information that can be collected. In fact, a variety of independent solutions are currently being developed, from both the theoretical and technological points of view, also considering the multitude of application scenarios that span from existing buildings to new ones, from residential sites to universities and municipal buildings. Therefore there is the need for the integration of such a multitude of solutions, possibly heterogeneous, and with the aim to be as general and comprehensive as possible, to be exploitable in different application contexts. Still referring to the collected information, a second major challenge consists in the availability of large amount of data, coming from different, not necessarily reliable sources. Therefore, data cleaning and validation issues arise, considering that in order to use such data for analysis, information extrapolation, etc., it is necessary to master data pre-processing to overcome the complexity of data by reducing high dimensionality and cardinality of data streams without oversimplify them. Moreover, exploring and mining the large amount of data generated can help discovering correlations between the building functions along with its characteristics, geographic location, climate conditions and occupants' behavior. The main aim of buildings data management and analysis is to provide robust models that can be employed as a decisional aid by energy management system to improve short-term and long-term building administration. Thus, data summarization and exploration need to be investigated. Finally, while the first generation of "smart building" approaches considered the user's behavior as a limitation to be overcome rather than a resource, nowadays it is clear that it is necessary to empower the user and make it an active actor within the management of the building to help further improving energy building performance and saving. It is therefore essential to provide the users with tools that allow them to understand their consumption/footprint associated with context information, making them aware and engaging them to improve their energy saving profile. The expected outcome is a general framework and a set of technologies for developing holistic systems able to integrate all the information related to the analyzed scenario and extensively mining the collected data to pave the way towards a greener exploitation of energy in building management.