Data-driven prediction and optimization toward net-zero and positive-energy buildings: A systematic review

Recent advances toward sustainable cities have promoted the concept of near-zero energy consumption. A Positive Energy Building (PEB) model has been developed by the European Union as part of Horizon 2020 to contribute to a cleaner neighborhood environment. To achieve PEB goals, a variety of factors must be optimized, including occupant comfort, building efficiency, economic benefits, and clean energy provision. Building modeling simulation combined with data-driven tools such as machine learning and artificial intelligence can be used to predict energy production and optimize passive and active systems. Based on these findings, this study evaluates studies from the past decade that include data-driven approaches, which accelerate different aspects of PEB, including supply and demand. These aspects include renewable energy supply prediction with the local context, optimizing comfort control with IoT, and reducing demand by optimizing building envelope design, materials selection, and active systems. While there are a few surveys regarding renewable energy management and energy efficiency in buildings, none simultaneously classified the algorithms in a PEB framework. Hence, this work inherently creates a technical framework for future researchers and building engineers to apply the appropriate data-driven approach for achieving net positive energy performance in residential, educational, and commercial buildings. Finally, comparing different applications suggests future research problems that can be addressed by integrating optimization algorithms and machine learning approaches, as well as data gaps that can be resolved to improve prediction accuracy.