From Building to Neighborhood: Investigating the Use of Artificial Intelligence and Machine Learning in Energy Management of Zero Energy Urban Neighborhoods Using a Bibliometric Approach (2020-2025)

نویسندگان

1 دانشگاه آزاد اسلامی

2

3

4 دانشگاه علم و صنعت ایران

doi
10.22109/jemt.2026.574220.1588
چکیده

Net Zero Energy Buildings (NZEBs) and nearly Zero Energy Buildings (nZEBs) have evolved from individual building concepts toward broader applications at the urban neighborhood scale. This paper presents a bibliometric analysis of energy management practices in zero energy urban neighborhoods, focusing on emerging research trends over the period 2020–2025. Using VOS viewer software, we analyze 584 peer reviewed articles retrieved from major scientific databases. The study maps the co occurrence of keywords, country level research activity, and author collaboration networks. Results indicate exponential growth in publications from 2023 onward, driven by the convergence of zero energy concepts with artificial intelligence, machine learning, deep learning, and digital twins. Key thematic clusters include energy management systems, HVAC optimization, renewable energy integration, smart grids, and the Internet of Things. Geographically, research leadership has shifted from Europe and the United States (2021–2022) to Asian countries (China, India, South Korea, Japan, Iran) by 2023–2025. A persistent research gap is identified: zero energy neighborhood development in hot and humid climates, particularly in developing economies, remains severely underrepresented due to high upfront costs and long payback periods. The review synthesizes design hierarchies passive strategies first, followed by efficient active systems, then on site renewable generation and discusses the limitations of current AI applications, including data availability, generalizability, interpretability, and the absence of standardized benchmarks.