Figure 2: Adaptive standard for naturally ventilated buildings.
Although both laboratory and field studies typically collect subjective data in terms of thermal sensation, Standard 55 presents temperature limits in terms of acceptability (with the goal of achieving 80% acceptability in the field). To create the link between 80% acceptability and measured thermal sensation, we accepted one of the underlying assumptions of Fanger’s PMV/PPD indices: namely, that a group mean thermal sensation (PMV) between the limits of ±0.85 corresponds with 20%of the group being dissatisfied (PPD). To apply a more stringent level of acceptability to the adaptive model, or if a building is expected to present greater than normal thermal asymmetries, an acceptability criteria of 90% might be chosen, corresponding to a mean thermal sensation falling within the limits of ±0.5.
For comparison, the 80% acceptability comfort zone in Standard 55 actually is based on a 10% general dissatisfaction criterion for the body as a whole, corresponding to tests performed in the laboratory under uniform conditions. It then allows for an additional average of 10% dissatisfaction that might occur because of local thermal discomfort. Since the adaptive model is based on field measurements, where people are naturally integrating whole body plus local sensations, field votes already account for both sources of discomfort.
A proposed adaptive standard for naturally ventilated buildings is shown in Figure 2. To make it easier for engineers to use, the regressions in Figure 1 (originally using ET*) have been recalculated based on mean monthly outdoor air temperature. At the time this article was written, the exact form and applicability of this proposed revision to Standard 55 were still being discussed. This comfort standard could be applicable to buildings in which occupants control operable windows, and where activity levels are < 1.2 met. As the outdoor temperature extends beyond the outdoor temperature limits included in the RP-884database, the acceptable indoor temperature limits could remain constant at the maximum and minimum levels.
To use this standard, engineers simply calculate the average of the mean minimum and maximum air temperatures for a given month, and then use Figure 2 to determine the acceptable range of indoor operative temperatures for a naturally ventilated building. During the design phase of a building, these numbers could be compared to the output of a thermal simulation model of the proposed building to determine whether the predicted indoor temperatures are likely to be comfortable using natural ventilation, or if air conditioning would be required. The figure also could be used to evaluate the acceptability of thermal conditions in an existing building by comparing the acceptable temperature range obtained from Figure 2 to indoor temperatures measured in the building
 
Figure 1: Observed and predicted comfort temperatures.
Conclusions
The research has demonstrated that occupants of buildings with centralized HVAC systems become finely tuned to the very narrow range of indoor temperatures presented by current HVAC practice. They develop high expectations for homogeneity and cool temperatures, and soon became critical if thermal conditions do not match these expectations. In contrast, occupants of naturally ventilated buildings appear tolerant of - and, in fact, prefer - a wider range of temperatures. This range may extend well beyond the comfort zones published in Standard55-1992, and may more closely reflect the local patterns of out-door climate change. Further analysis of research findings established that behavioral adaptations, such as changes in clothing insulation or indoor air speeds, could account for only half the observed variance in thermal preferences of people in naturally ventilated buildings. Since it has been established that physiological adaptation is unlikely to play much of a role in relation to indoor office environments, this suggested the rest of the variance was attributable to psychological factors. Chief among these was a relaxation of thermal expectations, possibly be-cause of a combination of higher levels of perceived control and a greater persity of thermal experiences in the building. Such research suggests that accounting for these broader adaptive mechanisms allows mechanical engineers to design and operate buildings in ways that both optimize thermal com-fort and reduce energy use. In many climatic settings, the practice of maintaining a narrowly defined, constant range of temperatures in fully air-conditioned buildings is unnecessary, and carries a high-energy cost. Unfortunately, the thermal comfort standards embodied in Standard 55 do not present alternative approaches to building conditioning. One reason is that the heat-balance models, on which the standard is based, were developed in tightly controlled laboratory conditions. In this process, people were considered passive subjects of climate change in artificial settings, and little consideration was given to the broad ways they might naturally adapt to a more wide ranging thermal environments in realistic settings. The laboratory context in which Standard 55 was established is similar to that of buildings with fully centralized HVAC systems. A historical connection exists between the two, since the standard originally was intended for application by the HVAC industry to the creation of “artificial climates” in “controlled spaces.”8 Therefore, it is not surprising that this research demonstrated that the PMV model could accurately predict people’s patterns of thermal preference in fully air-conditioned buildings. However, the research showed that the PMV model could not predict people’s thermal preferences in naturally ventilated buildings. This would seem to indicate the PMV model is an unsuitable guide when deciding whether air conditioning is even necessary in a particular building. On the strength of this research, we argue that an adaptive model of thermal comfort may usefully augment laboratory-based predictive models in the setting of thermal comfort standards. Furthermore, it appears that such an approach is essential to account for additional contextual factors and inpidual experiences that appear to modify people’s expectations in naturally ventilated buildings. As part of the next round of revisions to Standard 55, adoption of an alternative “adaptive” standard for naturally ventilated buildings may serve as a practical first step towards allowing engineers to adopt a more complex, socially and environmentally responsive approach to evaluating and designing indoor climates. It would reflect growing awareness among researchers that factors beyond the mere passive experience of a body’s thermal balance may play a significant role in determining human thermal preferences 自然通风英文文献和翻译(5):http://www.youerw.com/fanyi/lunwen_1947.html