The dissertation was defended at the public meeting of the Dissertation Defense Council of the Scientific Field of Civil Engineering in the SRA -I Meeting Hall of Vilnius Gediminas Technical University at 1 p.m. on 14 June 2024.
This dissertation addresses the energy consumption problems of ventilation systems, which are directly related to user behaviour. The main objects of the research are open-plan office spaces and their ventilation systems This dissertation aims to develop an artificial intelligence-based model for ventilation system control based on the prediction of user behaviour. The dissertation addresses several main challenges: first, to identify user behaviour parameters and assess their impact on building energy consumption; second, to carry out long-term monitoring of real buildings, to analyse data on occupancy and microclimate parameters and to prepare a data sample for the development of a mathematical prediction model; and finally, to adapt to machine learning-based models based on occupancy rates of premises and to validate the model. The dissertation consists of an introduction, three chapters, general conclusions, a list of literature, and a list of publications on the dissertation’s subject. The introduction discusses the research problem and the relevance of the work, describes the research object, formulates the aim and objectives of the dissertation, describes the research methodology, the novelty of the topic, the practical significance of the results, and the defended statements. The introduction closes with lists of the author’s publications and conference presentations on the subject of the dissertation and the presented structure of the dissertation. The First Chapter is devoted to a literature review. It discusses the factors influencing user behaviour and models for predicting user behaviour. The Second Chapter discusses the research objects and the behaviour of their participants, as well as the methods for predicting passive user behaviour and the concept of how to integrate them into the building ventilation system. The Third Chapter provides a sensitivity analysis of the prediction of machine learning models. A modelling algorithm for an open-type office ventilation system is proposed to show the performance of the system in predicting passive user behaviour. Each chapter closes with an outline of the general conclusions for the work covered in it. There is a separate chapter, of general conclusions of the research at the end of the dissertation. Six articles were published on the topic of the dissertation: two in peer-reviewed scientific journals with a citation index in Clarivate Analytics Web of Science databases, one in international conference publications referenced in the Clarivate Analytics Web of Science database Conference Proceedings, and three in other peer-reviewed journals. The results of the research were published in two scientific conferences in Lithuania and abroad.
Doctoral dissertation readers can search via VILNIUS TECH Virtual Library.
