<p dir="ltr">In U.S. residential buildings, space heating makes up about 43% of total energy use, with natural gas fulfilling 45% of this demand. As climate change concerns escalate, moving away from fossil fuel heating systems to more sustainable options are essential, especially in cold climates where heating needs are significant. Air-source heat pumps are a promising alternative, but their capacity and efficiency decrease as outdoor temperatures drop, impacting comfort due to lower supply temperatures (e.g., 32°C/90°F). This can lead to potential discomfort, as such temperatures feel cooler than skin temperature. Additionally, defrosting cycles pull heat from indoor spaces to clear outdoor coils. Conversely, gas furnaces provide steady heat at higher temperatures (over 49°C/120°F) without defrosting issues. Research shows that discomfort prompts occupants to raise thermostat setpoints and increase energy use.This study aims to investigate the influence of operational characteristics of various comfort delivery systems in cold weather on occupants' thermostat adjustment behaviors, identify the limitations of current heat pump systems, and develop a novel wall-embedded micro heat pump (WEMHP) radiant heating system that enhances comfort and reduces energy consumption, supporting the electrification of residential buildings. To achieve this goal, the research focuses on the following specific objectives: 1) develop a controlled laboratory testbed to emulate different thermal comfort delivery systems, including convective air and radiant systems; 2) investigate occupant setpoint preferences and thermostat adjustment behaviors under different operational modes using a residential community testbed; 3) study occupant thermostat adjustment behaviors for different types of heat pump systems through laboratory experiments; 4) develop and evaluate a novel wall-embedded micro heat pump for radiant heating in buildings; 5) design and test a prototype of the wall-embedded micro heat pump as a proof-of-concept demonstration.</p><p dir="ltr">This study first introduces the Human Building Interaction Laboratory (HBIL), a new facility with a modular design that includes reconfigurable thermally active panels for walls, floors, and ceilings. Each panel’s surface temperature can be independently controlled via a hot and cold water hydronic system, allowing the simulation of various climate zones, building conditions, and different heating/cooling systems. This setup facilitates research on localized comfort delivery, occupant comfort control, active building materials, and more.</p><p dir="ltr">Subsequently, a residential community test-bed was established within a newly built residential community in Indianapolis. A study was conducted in 30 homes to collect data on occupants' thermostat adjustments under two different operation modes: 1) a baseline mode featuring a heat pump paired with an auxiliary heater controlled by default thermostat heuristic rules, and 2) a comparison mode where the auxiliary heater was activated to provide the majority of heating. The findings showed that 8 out of 13 units preferred lower setpoints in the comparison mode, where higher supply air temperatures were utilized. Four distinct setpoint-increasing behaviors were identified, contributing to the observed setpoint differences between the two modes. Notably, two of these behaviors were closely linked to the operational characteristics of heat pumps in cold weather, specifically cases of insufficient and sufficient HP capacity.</p><p dir="ltr">To further explore the differences in setpoint preferences and the motivations behind setpoint adjustments, two scenarios were designed, and 32 experiments with human test-subjects were conducted in a controlled laboratory (Human Building Interaction Laboratory). The first case, with a single-stage heat pump and auxiliary heater, replicated the operational characteristics observed in the field study. The second case, using a variable-speed heat pump with enhanced comfort features, aimed to investigate participants' comfort preferences and provide insights for future heat pump design improvements. According to the thermal comfort survey results, 19 out of 32 participants increased their setpoints in the single-stage heat pump case, even though the heat pump had sufficient capacity to warm the indoor space. Cold air movement and indoor temperature fluctuations due to the heat pump cycling on/off were the main reasons participants reported increasing their setpoints in this case. In contrast, participants felt more comfortable with the variable-speed heat pump in the laboratory study, attributing their comfort to stable indoor temperatures and the absence of cold air movement.</p><p dir="ltr">Finally, a novel wall-embedded micro heat pump (WEMHP) was developed as a new distributed comfort delivery approach with several distinct advantages compared to alternatives: (1) A WEMHP eliminates the need for a secondary water loop and does not require separate indoor and outdoor units. Instead, a WEMHP unit operating in heating mode directly absorbs heat through an embedded heat exchanger (evaporator) at the outside wall surface and then conditions the indoor space using an embedded heat exchanger (condenser) at the indoor surface. (2) This packaged solution eliminates the need for extensive HVAC installation and on-site refrigerant charging. (3) The interior surface temperature of the exterior wall section empowered by the micro heat pump is independently controlled, allowing for distributed space conditioning and delivery of radiant heating to meet diverse occupant needs in different zones. The system performance was studied thoroughly based on energy simulation and experimental comfort study. Moreover, a prototype WEMHP was designed, assembled, and tested in a laboratory environment as a proof-of-concept demonstration. The test results demonstrated that the heating capacity under condition H1 reached around 190 W at a compressor speed of 4000 RPM with a COP of 1.67. Additionally, the system exhibited a fast thermal response, with a time constant τ₆₃ (the time it takes for the surface temperature to reach 63% of the difference between its final and initial values) of less than 0.5 hours and a τ₉₅of approximately 1.5 hours.</p>