Creating an adaptive model for office buildings in cold climate

. Japan's goal of becoming carbon neutral by 2050 requires both efficient use of facilities and maintenance of comfort for office workers in office buildings. In this paper, we clarify comfort temperatures from adaptive model in an office building in Hokkaido, as well as examine the factors that influence a more comfortable office environment. In the thermal comfort declaration, "Neutral" was the most declared in the MX mode (ventilated environment regardless of heating/cooling use), and the office workers were satisfied with the thermal environment. The comfort temperature was around 25°C in both MX and HT modes (heated environment), and a positive correlation was observed between the comfort temperature and the moving average outdoor temperature in MX mode, which led us to propose an adaptive model. Although the outdoor temperature in this model was about 8 °C lower than in previous studies, the comfort temperature was within the same range. There was no correlation between the amount of clothing worn by workers and their perception of cold and heat, but there was a positive correlation with the amount of activity.


Introduction
Natural ventilation and ventilation are attracting attention in Japan as the country to achieve carbon neutrality by 2050. The natural ventilation in offices not only saves heating and cooling energy, but also influences comfortable temperatures, which is an important factor in improving the work efficiency of workers. The recommended office temperature settings in Japan are 20°C in winter and 28°C in summer, but the recommendation is not based on field studies in Japan. Therefore, it cannot be said with certainty that this is a comfortable environment for office workers. Therefore, it is not clear whether the environment is truly comfortable for Japanese office workers. So, a domestic survey was conducted to clarify whether the current office environment is comfortable or not, as well as to create an adaptive model for office buildings that is expected to contribute to efficient use of heating and cooling and comfortable office space. In this study, we measured the thermal environment of office buildings in Hokkaido, a cold region in Japan.
This study examined an adaptive model and analysed the effects of the amount of clothing and behaviour on the adaptive model based on a survey of the measured thermal environment of office buildings in Hokkaido, a cold region in Japan, and a survey of subjective thermal reports by office workers. In addition, the correlation

Method of survey
In this study, data collection for both the thermal comfort surveys and the measurement survey was conducted using a method different from that used in previous studies. [1] First, thermal comfort reports were submitted at any given time using a Google form. Approximately 495 reports were obtained by this method. (165 respondents). The declaration survey asked the sex and age ,constitution, clothing and behaviour just before filling out the form, and mental and psychological state.
Next, in the measurement survey, one to six measuring instruments were installed in each office, depending on the size and seating position, to constantly measure room temperature, indoor glow temperature, carbon dioxide concentration, and other parameters. The daily average outdoor temperature data required for the adaptation model described below was obtained from the Sapporo Meteorological Observatory.

Thermal sensation scale
In the thermal comfort report, office workers were asked to rate their feeling of coldness and heat in the office environment at the time of the report on a 7-point scale. Table 2 shows the scale of the cold/heat report. Scale 4 was set to "Neutral" based on previous studies.

Activity
In this study, the behaviour that affects the comfort temperature was defined as "the activity that is taken by the office worker from 15 minutes prior to. The respondents answered the applicable items by indicating where they spent their time (indoors or outdoors) and under what conditions (sitting, standing, or walking around). The amount of activity was calculated using an activity conversion table. [2]

Clothing
The stewards were asked to respond to approximately 12 items related to upper and lower body undergarments, clothing worn over undergarments, and other clothing items. The amount of clothing was calculated by taking out the value of the amount of clothing from the conversion table for all of the workers' clothing. The total of these values was used as the amount of clothing worn.[3]

Comfortable temperature
Comfort temperature was calculated using the Griffiths method. In field studies, the calculation of linear regression by regression methods often does not work well, so this study was conducted using the Griffiths method. regression method often does not work well, this study used the following equation of the Griffiths method Therefore, as in previous studies, the following equation of the Griffiths method was used to calculate the comfort temperature The following equation of the Griffiths method was used in this study as in previous studies to calculate the comfort temperature. [4,9,10] Tc =T-(4-C)/a (1) Tc :Comfortable temperature [℃] T :Globe temperature [℃] C :Cold and heat senses a : Regression coefficients

Moving average of outside temperature
The moving average outdoor temperature is the temperature calculated by exponentially shifting the daily average outdoor temperature, and in this study the following The following equation was used in this study. [11] Trm =αTrm-1+(1-α) Tod-1 (2)  The moving average outdoor temperature on the first day is assumed to be the same as the daily average outdoor temperature on that day. As in previous studies, a reaction rate of 0.80 was assumed in this study.

Distribution of thermal sensations
The distribution of thermal sensations in Mixed-Mode and Heat-Mode is shown in Figures 1 and 2. Mixed-Mode is an environment in which natural ventilation was used with or without heating/cooling use. Mixed-Mode is an environment in which natural ventilation was used with or without heating or cooling, and Heat-Mode is an environment in which only heating is used. [5] This trend was similar in all offices. It indicates that there was no significant difference in the thermal sensation of the office workers.

Relationship between clothing and max and min temperatures
The factors that determine clothing is the maximum and minimum temperatures. In this study, we tested whether office clothing is dependent on outdoor temperature in offices where people spend a lot of time indoors. Scatter plots between the amount of clothing and the maximum and minimum temperatures are shown in Figures 3 and  4. Although there is no strong correlation between amount of clothing and the temperatures, a slight rightward trend is observed, with the clothing decreasing as the maximum and minimum temperatures increase. This indicates that many office workers may be adjusting their clothing to the outside temperature rather than the room temperature. However, we cannot be certain, and it is necessary to examine whether there is a correlation between room temperature and clothing.

Analysis of activity
The relationship between the activity and the thermal sensations is shown in Figure 5. The graph shows that the activity was 1 to 1.5 met within the comfort range(Declaration 3 to 5), because most of the activities in the offices are done while sitting. A trend of increased thermal sensation with increased activity was observed, although the correlation was small. The trend was observed not only in the office but also in the home in previous studies. This may be since room temperature changes in offices are small, and many office workers feel hot as their metabolism increases with an increase in the activity. Although the activity was limited in this study, information on the number of steps taken by office workers and their energy consumption was used in the wearables. However, if more detailed information on the amount of activity is reflected using wearable devices, such as the number of steps taken and the amount of energy consumed by the workers, a strong correlation may be obtained.

Analysis of clothing
The relationship between the clothing and the thermal sensation is shown in Figure 6. The amount of clothing in the comfortable range is 0.6-0.8 clo. The difference in clothing by each person is small, because most people wear suits or uniforms in the office. The results also indicate that there is little correlation between the clothing and thermal sensation. As mentioned earlier, room temperature changes are small in an office setting, and it is unlikely that the environment would be extremely hot or cold, which is why we predict that the association between the two is weak.

Adaptive model
The relationship between moving average of outdoor air temperature and comfort temperature in Mixed-Mode and Heat-Mode is shown in Figures 7 and 8 . Figures 7 and 8 below show the adaptive model of the relationship between moving average of outdoor air temperature and comfort temperature in Mixed-Mode and Heat-Mode, respectively. The adaptive model for Mixed-Mode shows a small correlation but a slight rightward trend, indicating that the comfort temperature is expected to increase as the moving average outdoor air temperature increases. This result is similar to previous studies. In addition, this model did not show a large difference in the range of comfort temperatures even though the moving average outdoor temperature was about 8°C lower than in the previous study. On the other hand, the heat-mode adaptive model showed a slight rightward trend. This means that as the moving average of outdoor temperature increases, the comfort temperature decreases. We assume that this is because offices in the Hokkaido area have higher insulation performance than those in other areas, which makes it easier for heat to accumulate during heating and makes people feel hotter. However, since the number of reports is small and not reliable, we would like to continue further analysis of the results. [6,7,8,9]

Factors that may affect comfort temperatures
The declaration of heat and cold is also influenced by factors other than the activity and the clothing .
In this chapter, we will examine the relationship between the thermal comfort surveys collected from the thermal comfort report of office workers (sex, constitution, air flow, and moving average of outdoor temperature) and thermal sensation. In this study, multivariate analysis was used. [10]

Sex
In the multivariate analysis, males were used as the standard, with males as 1 and females as 2.

Constitution
In the thermal comfort surveys, the participants were asked to rate their own constitution on a scale of 4. Table  4 below shows the scale.

Air flow
In the thermal comfort surveys, the respondents were asked to rate their own perception of airflow on a scale of 7. Table 5 below shows the scale.

Multivariable analysis
The analysis was conducted using EZR, with the objective variable being thermal sensations, and the explanatory variables being sex, constitution, feeling of airflow, and moving average of outdoor temperature. Table 6 shows the regression coefficients and P-values calculated in the analysis Table 6. Regression coefficients and P-values Figure 9 shows a box-and-whisker diagram of thermal sensations and gender. The graph shows that there is no significant difference between male and female respondents in terms of their perception of coldness and heat, but more women report being on the "cold" side. The correlation coefficients in Table 6 also indicate that more women than men report on the cold side. This is thought to be because women tend to feel the cold more easily in an office environment, where they are less active and have less muscle strength.  Figure 10 shows a box-and-whisker diagram of thermal sensations and constitution. From the graph, most of the body types other than hotness The graph shows that there is no significant difference in thermal sensations between the two types of constitution. Previously, the hot constitution had a wide range of responses, but the increase in the number of data may have changed this difference. Table 6 shows that those who are either cold or both are more likely to report the "cold" side of their body constitution. However, the results were not significant enough to be definitive.   Figure 11 shows a box-and-whisker diagram of Thermal sensations and air flow sensations. The graph shows that the more strongly people feel the wind, the colder they feel. We also found that most of the neutral respondents also reported neutrality in their cold and heat sensations. However, since we cannot conclude from the P-values that there is a significant difference, we will continue to examine this issue.

Thermal sensations and moving average of temperatures
The regression coefficients in Table 6 show that the increase in the thermal sensations with an increase in the moving average of outdoor temperature is larger than for the other items. The P-values indicate that the two correlations are highly correlated with each other, with little significant difference. The P-values indicate that the two correlations are highly correlated. This indicates that the indoor temperature may be affected by the outdoor temperature, even though the temperature in the office is set independently of the outdoor temperature. This factor and the fact that clothing decisions are likely to be determined according to the outside temperature can be considered.

Conclusion
The findings of this study are as follows.
1) The clothing decreases as the maximum and minimum temperatures increases. 2) As the activity increases, the thermal sensations increases, and the comfort temperature decreases.
3) There is no correlation between the clothing and the thermal sensation 4) In Mixed-Mode, the range of comfortable temperatures did not change from other regions, while in Heat-Mode, there was a downward trend to the right. 5) Thermal sensations may be greatly influenced by air currents and moving average of outside temperature.