Integrating Adaptive Thermal Comfort in HVAC Setpoint Temperature Control : : Basic and Advanced Artificial Neural Network (ANN) based Application

다국어 초록 (Multilingual Abstract)

The building sector is one of the major energy-consuming sectors, with its energy consumption share reaching up to 36% of the total final energy consumption worldwide. Among building energy end–uses, the energy used for air–conditioning in buildings accounts for a greater proportion of building energy consumption and has seen the most significant increase over the past decade. Research has been trying to tackle excessive energy consumption in buildings, mainly focusing on developing building energy efficiency measures such as the improvement of building thermal performance and energy efficiency of HVAC (heating, ventilation, and air-conditioning) systems. Nonetheless, research has shown that besides building thermal performance and efficiency of its systems, factors related to its operation, particularly HVAC operation, have significant impacts on the energy consumed by a building. There is a need for more operation-centered approaches towards energy efficiency in buildings. Primarily, HVAC systems are run to provide thermal comfort to occupants, and conventionally, the systems are operated with a fixed setpoint temperature across an entire air- conditioning period (cooling/ heating season). However, research has revealed that people’s thermal comfort is influenced by their thermal expectations that are linked to recent experience with the thermal environment. It was indicated that occupants of a building have the ability to adapt and feel thermally comfortable with their indoor environment at a wider range of temperatures, depending on recent changes in outdoor thermal conditions. The research carried out in this dissertation aimed at developing a method for defining optimal HVAC setpoint temperature that minimizes the energy consumed for air- conditioning while keeping indoor temperature within the adequate temperature limits defined based on variations in outdoor thermal conditions. The study analyzed the integration of adaptive thermal comfort in HVAC operation control at two application levels: a basic application suitable for existing buildings with no smart control system, and an advanced application ideal for modern buildings equipped with energy management systems. The developed basic application of adaptive thermal comfort for setpoint temperature control applies adaptive thermal comfort to define a daily setpoint temperature. To assess its energy saving potential, an energy simulation model of a case study building was developed and calibrated. The method was then applied to adjust the setpoint temperature in the calibrated model, and the energy reduction was evaluated. In this study, the estimated energy reductions were validated by applying the developed method in an actual building. During the validation experiment, both the method’s energy savings and impacts on the thermal comfort of building occupants were evaluated. In the case of the developed advanced application of adaptive thermal comfort for setpoint temperature control, the methodology consisted of two steps. As a first step, an artificial neural network (ANN) model was built and trained to forecast indoor temperature and HVAC energy consumption for the next hour, given the setpoint temperature, and current outdoor and indoor environment conditions. In the second step, once the ANN model was developed and its prediction accuracy was optimized, the model was then used to select a setpoint temperature that minimized air-conditioning energy use and maintained indoor temperature within the defined range. In addition, for the advanced application, an ANN model was developed separately for each air-conditioning mode (cooling and heating), and energy saving potential was calculated based on the collected energy data for cooling and heating. Validation of the method’s applicability, energy efficiency, and impact on occupant thermal comfort was done through in-situ application in a case study building. The results from the basic application of adaptive thermal comfort for setpoint temperature control indicated that 9.0 and 8.6% of energy used for air-conditioning could be saved during cooling and heating, respectively. The actual application of the method in a case study building indicated that an actual energy reduction of 7.7% was achieved from cooling and 8.3% in heating, validating the estimated energy saving potential of the developed method. Regarding the developed advanced application of adaptive thermal comfort for setpoint temperature control, the main findings are summarized as follows: 1) although the developed ANN showed good prediction performance, hourly predictions were more accurate for indoor temperature in comparison with the energy consumption; 2) the estimated energy saving potential for cooling and heating was 7.1% and 12.8%, respectively; 3) the in-situ application of the method indicated cooling energy reductions between 8.4 and 12.4%. The highest energy reductions were achieved from days with higher outdoor temperatures. Method validation results from heating showed energy reductions of around 10.4% and 11.2%. Occupants’ responses from the survey conducted during the validation period for both the basic and advanced applications indicated that the developed method did not negatively impact the thermal comfort of occupants. In fact, collected responses during heating indicated that occupants were slightly more thermally satisfied on days when the setpoint temperature was controlled using the developed method. The overall energy efficiency of the developed method indicated that an annual energy reduction of 8.4% and 11.9% could be achieved through basic and advanced application of adaptive thermal comfort for setpoint temperature control.

목차 (Table of Contents)

• ACKNOWLEDGEMENTS i
• DEDICATION ii
• Abstract iv
• List of Figures xi
• List of Tables xiv
• SECTION I 1
• STUDY OVERVIEW & LITERATURE REVIEW 1
• Chapter 1 2
• Introduction 2
• 1.1 Research background 2
• 1.2 Motivation and research objectives 4
• 1.3 Organization of the dissertation 5
• Chapter 2 6
• Literature review 6
• 2.1 Thermal comfort in built environment 6
• 2.1.1 Heat balance thermal model 7
• 2.1.1.1 Concept of heat balance thermal model 7
• 2.1.1.2 Limitations of heat balance thermal comfort model 8
• 2.1.2 Thermal comfort adaptation 9
• 2.1.2.1 Adaption from past thermal experiences 9
• 2.1.2.2 Adaptation from perceived control. 9
• 2.1.3 Adaptive thermal comfort models 10
• 2.1.3.1 EN 15251 adaptive comfort model 10
• 2.1.3.2 ASHRAE 55 adaptive comfort model 13
• 2.2 Energy saving through setpoint temperature control 15
• 2.2.1 Static setpoint temperature control 16
• 2.2.2 Dynamic setpoint temperature control 17
• 2.3 Model predictive-based HVAC system control 18
• 2.4 Summary of the literature review 19
• Chapter 3. 22
• Significancy and limitations of the study 22
• 3.1 Significancy of the study 22
• 3.2 Limitations of the study 23
• SECTION II 25
• STUDY METHODOLOGIES 25
• Chapter 4. 26
• Development of a basic application of adaptive thermal comfort for setpoint temperature
• control 26
• 4.1 Introduction of the chapter 26
• 4.2 Methodology 27
• 4.2.1 Case study building and simulation model 27
• 4.2.2 Simulation model calibration 31
• 4.2.2.1 Coincident hourly weather data 32
• 4.2.2.2 Building envelop 33
• 4.2.2.3 Building HVAC system 34
• 4.2.2.4 Building operation and schedules 35
• 4.2.3 Adaptive setpoint temperature: basic application 40
• 4.2.4 Method validation through actual application 40
• 4.3 Summary of the methodology 41
• Chapter 5. 42
• Development of advanced application of ANN and adaptive thermal comfort for setpoint temperature control 42
• 5.1 Introduction of the chapter 42
• 5.2 Methodology 42
• 5.2.1 Development of ANN model 42
• 5.2.1.1 Data preprocessing and model variables 43
• 5.2.1.2 Building ANN model structure 48
• 5.2.1.3 ANN performance optimization 51
• 5.2.2 Adaptive thermal comfort and ANN-based setpoint temperature control 59
• 5.2.3 Method validation through in–situ application 60
• 5.3 Summary of the chapter 60
• SECTION III 62
• STUDY RESULTS 62
• Chapter 6. 63
• Results of basic application of adaptive thermal comfort for setpoint temperature control 63
• 6.1 Results of simulation model of the case study building 63
• 6.2 Results of model calibration 64
• 6.2.1 Weather data calibration results 64
• 6.2.2 Calibration results from building envelope 67
• 6.2.3 Calibration results from building HVAC system 70
• 6.2.4 Calibration results from building operation and schedules 71
• 6.3 Results of basic application of adaptive thermal comfort for setpoint temperature control 75
• 6.3.1 Energy saving potential during cooling 76
• 6.3.2 Energy saving potential during heating 79
• 6.4 Results of method validation through actual application 82
• 6.4.1 Validation results for cooling 82
• 6.4.1.1 Validation of cooling energy reduction 84
• 6.4.1.2 Validation of impact on occupant thermal comfort 87
• 6.4.2 Validation results for heating 91
• 6.4.2.1 Validation results for heating energy reduction 93
• 6.4.2.2 Validation of impact on occupant thermal comfort (heating season) 95
• 6.5 Summary of the chapter 98
• 6.5.1 Summary of calibration results 98
• 6.5.2 Summary of energy saving potential of basic application 99
• 6.5.3 Summary of method validation results 100
• Chapter 7. 102
• Results of advanced application of ANN and adaptive thermal comfort for setpoint temperature control 102
• 7.1 Chapter introduction 102
• 7.2 Advanced application results for cooling season 103
• 7.2.1 Predictive ANN model for cooling 103
• 7.2.1.1 Developed ANN architecture 103
• 7.2.1.2 ANN performance optimization results 108
• 7.2.2 Energy saving potential of advanced application during cooling 111
• 7.2.3 Validation results from advanced application during cooling 114
• 7.2.3.1 Cooling energy reduction results 115
• 7.2.3.2 Occupant comfort results from advanced application during cooling 119
• 7.3 Advanced application results for heating 122
• 7.3.1 Developed ANN model for heating 122
• 7.3.2 Energy saving potential of advanced application during heating 129
• 7.3.3 Validation results from advanced application during heating 131
• 7.3.3.2 Occupant comfort results from advanced application during heating 136
• 7.4 Summary of the chapter 140
• 7.4.1 Summary of the advanced application for cooling 140
• 7.4.2 Summary of the advanced application for heating 141
• SECTION IV 143
• OVERALL ENERGY EFFICIENCY, CONCLUSIONS & FUTURE WORK 143
• Chapter 8 144
• Overall annual energy saving potential 144
• 8.1 Annual energy reduction from the developed basic application 144
• 8.2 Annual energy reduction from advanced application 145
• Chapter 9 147
• Conclusions and Future work 147
• 9.1 Conclusions 147
• 9.2 Future work 149
• Reference 150
• 국문 초록 154
• Appendix A. 156
• Abbreviations 156
• Appendix B. 158
• Prediction performance of the developed ANN models 158