Thermal comfort optimization with occupant interaction in dynamic HVAC control

According to the Environmental Protection Agency (EPA), people spend 90% of their time indoors. Consequently, a significant number of study is performed in order to determine the effect of thermal environment on occupant health and productivity. The findings of these studies show that thermal environment has a significant effect on occupant thermal sensation and well-being. With the development of HVAC and building management systems, total control of the indoor environment becomes possible, and comfort bears a higher level of importance in order to maintain healthy indoor conditions. Thermal comfort is defined as "the condition of mind that express satisfaction with the thermal environment" and its depends on the different physical, physiological and psychological parameters. In order to assess thermal comfort, different models are proposed by a significant number of studies. The most widely accepted method is developed by Fanger, and it rests upon heat balance equations between indoor environment and the human body. This model is a function of six parameters, which split into environmental and personnel parameters; indoor air temperature, mean radiant temperature, relative humidity, air velocity, activity type and level, and clothing insulation of occupants. The focus of this thesis is research and evaluation of dynamic thermal comfort optimization methods for a shared spaces. The reason of that in single occupant offices, a thermally comfortable environment can be created simply based on occupant requirement. However, it is difficult to find an optimal thermostat setting temperature for multiple occupants sharing the same office. Studies show that most of the occupants have to stay at uncomfortable environments during the day because of the lack of proper control in buildings. It causes a decrement in performance and well-being of occupant. To overcome this issue, HVAC system should be operated dynamically based on time-varying temperature requirement of space and occupant thermal comfort conditions. However, it is hard to determined optimum temperature setting based on thermal comfort in practice. Most of existing Building Management Systems (BMS) have the issue of the absence of adequate equipment to assess thermal comfort conditions. Even with the proper equipment, it would be near impossible to meet absolute satisfaction because of the subjectivity of the matter and optimum thermal comfort conditions vary from person to person. In order to control HVAC system based on occupant requirement, The occupant participating approach has been developed. This method is bringing the humans in the loop by using their thermal perception feedback to improve mathematical models prediction, and it is started to utilize in a growing number of studies. In this study, thermal comfort optimization methods, which using the occupant participating approach, was examined to find simple and accurate optimization models. For this purpose, literature reviews and commercial products were reviewed, and three different optimization methods were chosen to evaluate. One of these methods uses PMV (Predicted Mean Vote), which is mainly used the method in studies and standards to assess thermal comfort, model to estimate initial thermal comfort conditions with temperature and humidity sensors outputs. In order to correct the PMV estimation, occupant participatory approach is used to collect real and continuous thermal sensation feedback of occupant via a smartphone application. Based on corrected PMV value, thermostat set point temperature is adjusted accurately. In the second method, only occupant feedbacks, which are collected in a similar way to the first method, are used to adjust the set point temperature of indoor HVAC system. The final method uses two-step model; the first step is calculation the optimal temperature for each occupant based on their energy expenditure level estimation and outdoor temperature and the last step is an adjustment the temperature based on occupant dynamic thermal sensation feedback via smart phone. Performances of the selected optimization models were evaluated via Design Builder and EnergyPlus simulation tools. The analysis was conducted in a case study zone, which is an open plan office and located in ARI 6 Technopark building, Istanbul Technical University. Model performance was analyzed regarding energy consumption and thermal comfort conditions. Besides these analyses, effects of thermal comfort on occupants' productivity were examined. For this purpose, humidity and temperature data were monitored and recorded for a one-week period, also during this process, occupant real time feedbacks about their thermal sensations were collected via website application. Measured data were used to calibrate and validate the simulation model of case study building. The evaluations of optimization methods were performed by using calibrated model. The findings of this thesis indicate that dynamic set point temperature control helped to optimize thermal comfort. In addition to this benefit of the control method, the productivity of workers increases under comfort conditions. On the other hand, the results proof the trade-off between thermal comfort and energy consumption of the building.