Context-based Control Strategies for Improving Energy Efficiency in Multi-zone Buildings

Abstract

With the increasing concern about energy saving, demand-side management has been widely implemented during the building life cycle. In the design phase of buildings, technical improvements in the insulation of its envelope and the efficiency of mechanical and electrical equipment significantly contribute to energy saving in building. Also, the energy-efficient operation of mechanical and electrical equipment is a permanent solution to achieve energy saving in the operation phase of buildings. Among these alternatives, recent efforts have focused on the operational solution due to a significant energy saving potential with relatively less effort.

Due to the importance of control strategies in demand-side management, a substantial amount of studies has been conducted to investigate the amount of energy saving from heating, ventilating and air conditioning (HVAC) scheduling techniques in buildings. Unfortunately, despite the previous achievement, the two main problems still exist in establishing optimal HVAC control strategies in multi-zone buildings. First, almost all studies have scheduled the operation of HVAC systems at the zone level. Although the zone-based HVAC scheduling provides a significant reduction in energy consumption, there still remain limitations in maintaining occupants thermal comfort. This is because a zone may consist of multiple rooms having different energy use patterns. Furthermore, its practical applications are limited due to a time-consuming process to set up control parameters in multiple zones. Second, little is known regarding the effect of contextual variables on energy saving from HVAC scheduling techniques in multi-zone buildings. Although only few studies have investigated how dynamic environment-related variables affect energy saving from HVAC scheduling techniques, they assumed that all zones have identical energy use patterns. Thus, it is impossible to establish optimal control strategies that can maximize energy saving from HVAC scheduling techniques in in multi-zone buildings.

As an effort to address these problems, this research aims to investigate how temporal and weather variables affect energy saving from HVAC scheduling techniques in multi-zone buildings. In order to achieve this objective, representative end-user groups (EUG) are identified in dormitory buildings of a university in Seoul, South Korea. Then, the following two models are developed depending on their purposes. First, a data mining-based model is constructed to predict baseline energy consumption for EUGs. Second, a thermodynamic model is developed to simulate post-retrofit energy consumption under controlled conditions by HVAC scheduling techniques. Then, based on the developed models, energy performance simulation is conducted to evaluate the amount of energy saving from HVAC scheduling techniques in different temporal and weather contexts.

From the results of energy performance simulation, the two key findings are summarized as follow. First, global temperature and on/off control of HVAC systems, as HVAC scheduling alternatives, produce the significant amount of energy saving in the case buildings. This implies that the buildings exhibit characteristics of low energy efficiency in heating seasons. Second, the HVAC scheduling techniques produce different energy saving potentials depending on outdoor temperature and course period. In other words, this indicates that establishing optimal control strategies is dependent on the given contexts. However, there is not always a consistent relationship between contextual variables and optimal HVAC control strategies in multi-zone buildings.

The main contribution of this research is to improve our understanding of the effect of temporal and weather variables on energy saving from HVAC scheduling techniques in multi-zone buildings. More specifically, this research provides a first look into the contextual behavior of representative end-user groups in multi-zone buildings. Additionally, this research contributes to an enhancement of the knowledge about how the characteristics of representative end-user groups affect the performance of building energy use prediction. Lastly, the developed models enable facility managers to schedule the energy-efficient operation of HVAC systems without compromising occupants thermal comfort in multi-zone buildings.