Do indoor comfort temperatures change in relation to out-door weather and climate? Adaptive theory suggests that the thermal expectations of building occupants, and their subsequent expectations for indoor comfort, will be dependent on outdoor temperature. This relation may vary, however, based on the extent to which the indoor environment is connected to natural seasonal swings in outdoor climate. Figure 1 shows a regression of indoor comfort temperatures as defined earlier against an outdoor temperature index for centralized HVAC (left graph) and naturally ventilated (right graph) buildings. The outdoor temperature index used was mean effective temperature (ET*). Each graph shows the regressions based on both observed responses in the database and the PMV predictions.
Looking first at observed responses (dotted lines), the gradient for the naturally ventilated buildings was more than twice that found in buildings with centralized HVAC systems. One possible interpretation of this finding is that occupants of the HVAC buildings become more finely adapted to mechanically conditioned, static indoor climates. In comparison, the range in thermal comfort levels in naturally ventilated buildings showed a much larger variation, suggesting that occupants of these buildings preferred conditions that more closely reflected out-door climate patterns.
How do field-based measurements compare to lab-based predictions, and what does this say about adaptation? The ob-served and predicted lines within each graph in Figure 1 pro-vide insight into how adaptation may influence the relationship between indoor comfort and outdoor climate in the two building types. Recall that clothing insulation and air velocity both had a statistical dependence on mean indoor temperatures (and are probably related to outdoor temperature as well). Both are included as inputs to the PMV model. Therefore, one would expect to see that the indoor comfort levels predicted by the PMV model might also show some dependence on outdoor climate. In fact, as seen in Figure 1, they do. In the HVAC buildings (left-hand panel of Figure 1), the observed (dotted) and predicted (solid) lines appear very close together, demonstrating that PMV was remarkably successful at predicting comfort temperatures in these buildings. A corollary of this finding is that, in HVAC buildings, behavioral adjustments to clothing and room air speeds fully explain the relationship between indoor comfort temperature and outdoor climatic variation, and that these adaptive behaviors are, in fact, adequately accounted for by the PMV model.
However, the remarkable agreement between PMV and adaptive models in the HVAC buildings clearly breaks down in the context of naturally ventilated buildings (right-hand panel of Figure 1), where the observed responses show a gradient al-most twice as steep as the PMV model’s predicted comfort levels. By logical extension therefore, it appears that behavioral adjustments (clothing and air velocity changes) may account for only half of the climatic dependence of comfort temperatures within naturally ventilated buildings.
What explains the rest? Having accounted for the effects of behavioral adaptations, physiological (acclimatization) and psychological components of adaptation are left to explain the pergence. But, as noted previously, existing literature suggests that acclimatization is unlikely to be a significant factor. This leaves psychological adaptation as the most likely explanation for the difference between field observations and PMV predictions in naturally ventilated buildings. This means the physics governing a body’s heat balance must be inadequate to fully explain the relationship between perceived thermal comfort in naturally ventilated buildings and exterior climatic conditions
An Adaptive Comfort Standard Using Standard 55 to determine acceptable indoor temperature ranges requires one to know, or at least anticipate, the average metabolic rate and amount of clothing worn by people in a building, regardless of whether that building is already built or occupied. In contrast, an adaptive model relates accept-able indoor temperature ranges to mean monthly outdoor temperature (in this case, defined as the arithmetic average of mean monthly minimum and maximum air temperature). This is a parameter already familiar to engineers and can be found easily by examining readily available climate data, such as that published by the U.S. National Oceanographic and Atmospheric Administration (www.ncdc.noaa.gov). Because the adaptive model is based on extensive field measurements, the relationship between expected clothing and outdoor climate already is built into the empirical statistical relationship. 自然通风英文文献和翻译(4):http://www.751com.cn/fanyi/lunwen_1947.html