COVER STORY PREDICTIVE THERMAL COMFORT MODELS6 Thermal comfort models are powerful tools to help designers make informed design decisions that can impact comfort and energy performance. The perception of thermal comfort quality is formed at four levels, as illustrated with the pyramid in Figure 1: climate, physics, physiology and psychology. On the left side of Figure 1, a steady-state whole-body thermal balance model links the top three levels of the pyramid to predict thermal comfort using the predictive mean vote (PMV) metric. The PMV model, derived from laboratory experiments on human subjects, assumes that individual thermoregulatory and mental responses to the environment are fixed (i.e. unaffected by any personal individualities or background, or by any local, building, environmental or climate context), which lead to narrow ranges of indoor comfort temperatures. The model can be run under different personal and environmental scenarios, affecting the indoor physics (blinds, fans), personal heat balance (clothing) and physiology (occupant activity), allowing for wide variations of indoor temperatures providing thermal comfort. In reality, our thermal comfort perception indoors is also largely influenced by the prevailing local weather and climate. After a period of exposure to hot or cold weather, our thermoregulatory system physiologically adapts, and adjusts its response to those conditions (acclimatization). We also tend to expect and accept slightly warmer or colder temperatures indoors, depending on the weather. On the right side of the pyramid in Figure 1, a dynamic adaptive thermal comfort (ATC) model, derived from field studies on humans, considers that humans adapt physiologically and psychologically to given climate and building environmental contexts, which fine- tune their physiological (e.g. lower metabolism) and psychological responses (i.e. tolerate wider temperature ranges), alter thermal perception and expectations, and trigger behaviours consistent with the climate and the environmental controls available. A higher level of perceived environmental control relaxes our thermal expectations on the indoor environment. The adaptive theory views comfort as a main trigger for changes to occupant behaviour (discomfort leads to action). Such physiological and psychological fine-tuning is represented by the dots in the shaded area in Figure 1. In Figure 1, a thermally adaptive environment is able to respond gradually to outdoor temperature changes, maintaining indoor thermal fluctuations within the ATC model limits and responding effectively to occupants’ thermal requests. Further to Figure 1, our comfort perception is also tuned by local and personal contextual factors. Studies reveal that personal (social, economic and demographic) characteristics have a significant impact on household decisions to use their air conditioning system.7 In summary, it can be argued that the PMV model can be applied to homes in extreme cold or hot and humid climates or seasons, where dwellers need to rely almost entirely on the mechanical system to maintain comfort. Furthermore, given that the PMV model ignores feedforward (i.e. perceived control) and feedback (i.e. thermal response) human-environment loops that trigger behavioural decisions, the PMV model also seems to be more applicable where limited adaptive personal or environmental control opportunities exist. In contrast, the ATC model relates indoor thermal comfort to the prevailing outdoor weather; as such, it is suitable to evaluate the effectiveness of passive design strategies. Therefore, the ATC model is applied to naturally conditioned buildings in mild and warm climates or seasons where the prevailing mean outdoor temperatures remain between 10ºC and 33.5ºC, because those ranges of temperatures allow greater flexibility for building occupants to adapt to the prevailing weather conditions, and for buildings to be more effective in moderating the effects of the weather indoors. In theory, ATC models could also be applied to mechanically conditioned buildings, depending on the adaptive opportunities available. Limitations of thermal comfort models have been acknowledged by industry experts and researchers. Understanding the mechanisms of adaptation sketched in dotted lines in Figure 1 under different climate and personal contexts, and reflecting these in adaptive models, is an active area of research8 . Over the past few years a wealth of research on residential thermal comfort has emerged, motivated by the need to support thermally comfortable and energy-efficient residential designs. The research has focused on developing residential adaptive thermal comfort models for different climate contexts under the premise that FIGURE 1: SKETCH OF TWO MAIN THERMAL COMFORT MODELS. 18 BCBEC ELEMENTS A BCBEC PUBLICATION