Development of detailed HVAC system models for effective simulation of building performance at varying levels of automation

In recent years, the urgency to curb energy demand in sustaining everyday activities has grown, with buildings playing a central role due to their intensive energy requirements. Facilities such as schools, hospitals, shopping centers, and corporate offices consume significant energy to ensure comfortable and safe environments. A major contributor to this high demand is the operation of HVAC (Heating, Ventilation, and Air Conditioning) systems, which are often less efficient than desired. This inefficiency is largely due to limited adaptability and automation, causing these systems to consume more energy than necessary to maintain comfort standards. Thus, enhancing the responsiveness and intelligence of HVAC operations presents a substantial opportunity for advancing building energy efficiency and reducing overall consumption. This thesis aims to quantitatively assess the impact of different automation levels, as defined by the UNI EN ISO 52120-1 standard, on indoor comfort and energy consumption, establishing a robust benchmark for various strategies. The study focuses on heating systems, specifically on operational functions related to generation and distribution networks. The analysis of the heat generation system considers two distinct automation levels: the first employs a fixed supply temperature, while the second utilizes a climate-responsive approach, adjusting supply temperature based on outdoor conditions. For the emission and distribution network, two control strategies were implemented: automatic control with a fixed time schedule and automatic control with optimal start and stop functionality. The case study examines two office spaces with different structural characteristics and occupancy profiles, located in Turin. The HVAC system serving these offices is a water-based system with radiator terminals. The impact of automation levels was analyzed using two types of heat generation systems, a condensing boiler and a heat pump, to determine which system offers greater advantages at higher levels of automation. To implement the various control strategies, a co-simulation was conducted between two different open-source software programs: OpenModelica and Python . The first, which employs the Modelica language, was used for modeling the physical structure of the building and the heating system in its entirety, encompassing the generation, distribution, and emission of heat. Python was utilized to implement the different control strategies, including the RBC (Rule Based Control) controller that manages the pumps and the office setpoints and the optimal start and stop. The co-simulation is carried out using the Functional Mockup Interface (FMI) standard. The proposed control strategies were compared using KPIs (key performance indicators) that assess indoor discomfort, energy consumption, and cost relative to the baseline cases, which correspond to the lowest level of automation. The findings of the analysis indicate that automation levels have a considerable influence on performance, representing a highly promising avenue for optimizing building energy consumption. Additionally, the study demonstrates that the intrinsic technical properties of the heat generator can significantly impact system performance as automation level varies.