Development of Energy Disaggregation Algorithms for Commercial Buildings

Energy management in commercial buildings requires accurate measurement of enduses, which is crucial for sustainability and cost reduction.However, a common challenge is the lack of high-resolution energy submetering, hindering effective energy management.To this end, a suite of load disaggregation algorithms suitable for buildings with and without building automation systems (BASs) is developed in this study.The thesis consists of four parts.The Ąrst part uses multiple linear regression models estimated by the genetic algorithm and a least-square solver to disaggregate heating and electricity data using BAS data as predictors.The method was tested on data from an Ottawa office building and showed accurate disaggregation into end-use categories, such as occupant-controlled loads and distribution loads for electricity, air handling unit (AHU) heating coils, and perimeter heating devices for heating.The second part explores the impact of submeter density and conĄguration on the disaggregation strategy using BAS trend data as predictors.Findings highlighted factors affecting the minimum number of submeters needed for accurate disaggregation, providing insights for building design codes, energy consultants, and policymakers.The third part proposed a time-series decomposition-based method to disaggregate major energy end-uses without BAS trend data.Data from ten office buildings in Ottawa were used to break down total energy use into three major end-uses.Disaggregation results were compared with actual submeter data, and insights into lighting and thermal scheduling were assessed.The proposed method showed promising performance, suggesting potential for quick and low-cost auditing of commercial buildings with limited submeter data access.The fourth part of this thesis utilized the developed disaggregation algorithms integrated into a visualization tool for end-users to interact with disaggregated data.Additionally, preliminary recommendations for thermal submetering requirements were incorporated based on the Ąndings presented in the second part of the thesis.This thesis enhances energy management in commercial buildings by developing practical energy disaggregation iii I extend my gratitude to Dr. Burak Gunay and Dr. Araz Ashouri, my Ph.D. supervisors, for their outstanding guidance and dedicated support.Their enthusiasm, motivation, patience, and expertise signiĄcantly contributed to the enhancement of both my research and technical writing abilities.IŠd also like to acknowledge the examination committee-Ian Beausoleil-Morrison, Dr. Elie Azar

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.