 | Dr. Behzad NajafiHe is an Associate Professor at the Energy Department of Politecnico di Milano and the head of Science at DataOptima Lab. His research activities have been focused on the application of artificial intelligence-based modelling and optimization in the energy sector. He has specifically worked on physics-inspired machine learning-based thermal behaviour modelling of indoor environments and HVAC systems to facilitate real-time optimization aiming at providing demand flexibility, PV self-consumption enhancement, and energy saving. He has also carried out activities on smart meter-driven remote auditing of buildings and HVAC load disaggregation, along with physical modelling and long-term performance optimization of energy systems. In the context of these activities, he has collaborated with several industrial firms and academic institutions, has supervised 2 PhD theses and over 40 M.Sc. thesis projects, and has published over 40 journal publications. |
Description:
In this seminar, key motivations and benefits of developing and deploying adaptive HVAC setpoint management strategies in buildings are presented. In this context, the implementation of these strategies to enable smart ramp-up procedure, which permits achieving energy saving, is first discussed. Next, employing these strategies to facilitate HVAC demand flexibility, with different applications (e.g. PV self-consumption enhancement) is explored. Machine learning (ML)-based pipeline development for predicting the thermal behavior of indoor environments, which is a necessary step in implementing the latter strategies, is also briefly discussed.
What will you learn?
- The importance of achieving both energy efficiency and demand flexibility in buildings
- Adaptive HVAC setpoint management strategies and their applications
- HVAC-driven PV self-consumption enhancement
- A brief introduction to machine learning-based pipelines for indoor thermal behavior prediction