Analysis of the accuracy of the Fanger model in the intelligent building – a case study

Nowadays, people spend most of their lives indoors, depending on the season, climate zone, work or study. Considering the amount of time spent in a closed environment, you should be inclined to create the most favourable and comfortable conditions for the internal environment, i.e. thermal comfort. The study, which was carried out in an intelligent building of “Energis” of the Kielce University of Technology, was to bring the authors of the work closer to whether such a building meets the thermal expectations of people using the PMV (Predicted Mean Vote) and PPD (Predicted Percentage Dissatisfied) indicators calculated on the basis of surveys and compared with the indicators calculated on the basis of the ISO 7730 standard.


Introduction
People's awareness of the feeling of thermal comfort is currently constantly growing. Many people can determine that a certain temperature range is appropriate or unacceptable for them. In order to learn about people's thermal sensations, a seven-step scale is used, with the help of which people define their thermal preferences by marking answers in the questionnaires 'too cold' (-3), 'to cool' (-2), 'pleasantly cool' (-1), 'comfortable' (0), 'pleasantly warm' (+1), 'too warm' (+2), 'too hot' (+3) [1]. Such a method is called direct. The indirect method is primarily used to learn about the parameters of the microclimate inside the rooms, which are measured by environmental measures. With both of these methods [1], you can calculate the PMV, which is the Predicted Mean Vote, and the PPD, or the Predicted Percentage of Dissatisfied In the case of the PMV questionnaire, it is the average rating describing the thermal impressions, and the PPD calculation takes into account the respondents' answers 'too cold' (-3), 'to cool' (-2) and 'too warm' (+2), 'too hot' (+3) -because it is this range that determines dissatisfaction with the prevailing conditions. On the other hand, the method in which the measured environmental parameters play an important role are used to calculate PMV and PPD based on ISO 7730, based on the Fanger model. This makes it possible to compare the actual thermal sensations of people with the assumptions of the ISO 7730 standard and, above all, to check whether the standard reflects the perceived thermal sensations of people.
Thermal comfort tests and attempts to modify the Fanger model are carried out in many places around the world. The desire to learn and, above all, understand the real thermal sensations of people and compare them to standards, prompts many scientists to seek a solution and improve the temperature ranges adequate to the expectations set by people. A literature review provides a great deal of information about studies conducted in different climatic zones, such as, for example, Martinez-Molina et al. [2] who proposed a comparison of the thermal feelings of teachers and students aged six and seven from a Spanish primary school located in a building of historic importance. The authors of the study focused on understanding the values of PMV, TSV and PPD. The obtained results gave information that children definitely have a higher, but also more difficult, threshold of thermal acceptability in rooms. Krawczyk & Kapjor [3] conducted a study on thermal comfort model at the Kielce University of Technology based on the Fanger model. The research group that was examined was 98 people in the age range of 19-23 years. The main aim of the study was to understand the values of PMV and PPD and to compare them with the applicable standards. After reviewing the results, the authors found that the assessments of people and the standards imposed do not match, which means that Fanger does not reflect the actual assessment of the respondents. Therefore, the authors undertook the modification of Fanger. Similar observations regarding the comparison of PMV and PPD values and actual feelings of the people were made by Majewski et al. [4] The authors noticed considerable differences between the calculated and actually observed values. An interesting study was a study by Zhang et al. [5] who examined the crew in the cabin of the submarine. The time in which the tests were performed included six dives to a depth of 7000m. PMV was consistently in the range -2 to +3, with constant changes from warm to cold. Dębska & Krakowiak [6] analyzed the thermal feelings of 83 people, from three rooms, in three different educational buildings. It turned out that from the temperatures obtained, people felt the best at temperatures around 22.5 o C, while at higher temperatures 25.3 o C and 27.6 o C the worst. Moreover, after calculating the PMV from the questionnaires and the ISO 7730 standard, there were considerable discrepancies between the guidelines from the standard and people's feelings. Lu et al. [7] undertook the study of the Chinese island of Hainan because of its unique tropical climate, which influences the thermal comfort felt by its inhabitants. 1944 people living in various buildings were examined. The temperature which was found to be the most neutral was 26.1 o C. However, the acceptable range that resulted from the respondents' responses ranged from 23.1 o C -29.1 o C. Additionally, it turned out that moisture had no significant impact on the feeling of comfort by the surveyed inhabitants of the island. Another example of research on thermal comfort is the study by Garcia et al. [8], who analyzed the results obtained from the responses of office workers with natural ventilation, located in various parts of the city of Bogota, Colombia. A total of 790 questionnaires were obtained. It has been shown that for 96.58% of people the best temperature in this climate zone is 23.47 o C. In the tropical city of Makassar, Indonesia, Hamzah et al. [9] examined the thermal comfort felt by 1,594 students in eight high schools -in 48 classrooms. The temperature measuring range obtained for noon hours was 28.2 o C to 33.6 o C. According to the surveys received from students, almost 80% of them accepted such high temperatures. Staying in a tropical climate, the research proposed by Balbisa-Morejon et al. [10] aims to get to know the thermal feelings of students at one of the universities. The authors showed that for about 90% of the respondents the internal environment was acceptable in the temperature range of 23 o C -24 o C. Trebilcock et al. [11] conducted research in Chile in 19 buildings, located in Concepcion and Santiago. The research provided information that the population living there accepts and adapts to the prevailing temperatures. However, a neutral temperature range of 19.5 o C to 24.6 o C was achieved with 19.9 o C to 24.6 o C being preferred. Dębska et al. [12] examined 14 people in the lecture hall in the autumn period. The air temperature obtained during the test was 29.4. People who completed the questionnaires about their thermal feelings confirmed in 80% that such a temperature is unacceptable and the ventilation in this room did not meet their expectations in terms of feeling thermal comfort.
The main aim of the article is to check whether the 'Energis' intelligent building meets the thermal expectations of people and whether the values of PMV and PPD calculated on the basis of the questionnaires and the ISO 7730 standard coincide or are different.

Methods
The research was carried out in an intelligent building of the Kielce University of Technology, called "Energis", which is close to the passive building standard due to proper thermal insulation [13]. This building was built in 2012 and its main assumptions were to be energy self-sufficiency, control of indoor lighting or temperature, limiting energy losses, reducing pollution emitted to the environment, etc. Energis is shown in Figure 1. The research was carried out in six classrooms using the Testo 400 meter and specially designed questionnaires for people participating in these studies -as already mentioned in the introduction using the direct and indirect methods. The questionnaires included, inter alia, questions about the feeling of temperature, humidity, general satisfaction with a given room, etc. The meter, on the other hand, provided information about the current conditions of the internal environment during the test (temperature, humidity, light intensity, etc.). The meter is presented in Figure 2. In order to present the way of recording such results, data from one room will be used. The graph of changes in the parameters of temperature, relative humidity, globe temperature and air velocity is shown in Figure 3. From the start of the study, it can be seen that the air temperature (red line) and the temperature of the black sphere (blue line) tended to increase. On the other hand, if you look at the relative humidity (black line), the trend is downward. The highest swing can be seen for air velocity (the highest point reached was 0.22 m /s).
The meter with the probes was placed in the center of the room at the height of sitting people. After about 15 minutes, when the parameters remained unchanged, they were saved. In turn, the people who were in the rooms during the measurements -were to complete the questionnaire, which were then collected for analysis. In total, 75 people aged 19 to 26 participated in the study, where women constituted 1/3 of the group and men 2/3 of the group.

Results
In order to better understand the PMV and PPD indices, the obtained parameters of air temperature and relative humidity, air speed, globe temperature for six rooms have been summarized in Table 1. The temperature range between the six rooms was from 23.0 o C -25.2 o C and showed no significant temperature differences among themselves. Relative humidity ranged from 34.5% to 45.49%, air velocity from 0.05 -0.08 m / s. Based on the analysis of respondents' responses, the PMV and PPD for each room were calculated and then compared with the PMV and PPD calculated on the basis of the current ISO 7730 standard.
In order to calculate PMV from the questionnaires, people marked their thermal sensation on the question "How would you rate your thermal sensation right now?" describing them as 'too cold' (-3), 'to cool' (-2), 'pleasantly cool' (-1), 'comfortable' (0), 'pleasantly warm' (+1), 'too warm' (+2), 'too hot' (+3). Then, from each research group, the respondents' answers were averaged and named as Thermal Sensations Vote (TSV). This resulted in six PMV values (as six rooms were tested) which were then compared to the PMV calculated on the basis of ISO 7730. An overall assessment of the well-being in the room will facilitate the analysis of PMV and PPD values, as it will determine if people are feeling well in an intelligent building. The respondents indicated such responses in the questionnaires as -2 -very bad, -1 -bad, 0 -indifferent, +1 -good, +2 -very good. Figure 4 shows the overall responses of respondents from all rooms.  Generally, respondents felt good (+1) in the room. As many as 58.67% of people selected this answer. 37.33% of people declared that they felt indifferent in the room (0). Only 4% of people considered their well-being as bad (-1). Nobody described their feelings as very good (+2) or very bad (-2). Consequently, Energis lived up to people's expectations as more than half felt comfortable in the halls. Referring to previous research, the room that was examined [12] showed that the selected room was too hot or too warm for 80% of the respondents, as opposed to the research in which 96% of people are satisfied with the prevailing microclimate conditions. In turn, for the research carried out in three different rooms [6], it turns out that only the best conditions prevail in room 1, where the temperature is the most optimal to the other 2.
In the further part of the work, the results of calculating the PMV value according to the polls, calculated on the basis of the average response from the thermal sensations vote (TSV) and the standard, separately for each room will be presented. The optimal range for PMV is -0.5 to +0.5. Figure 5 shows a comparison of the PMV values according to the questionnaires and the PMV calculated from the standard.  Figure 5 shows how huge discrepancies exist between the calculated PMV according to the questionnaires and the PMV from the standard. The values calculated on the basis of the standard are in accordance with the applicable standards. On the other hand, in rooms 1, 3 and 6, the PMV calculated from the questionnaires is more than one, which proves that the rooms were too warm and these people did not feel thermal comfort, despite the fact that the standard says that there is compliance, and the differences are consecutive were 0.93, 1.43, 1.20. Rooms 2, 4 and 5 differences were 0.51, 0.35 and 0.66. In rooms 2 and 4, the respondents showed that the rooms are slightly cool. On the other hand, the results obtained from the calculator showed that the feeling of discomfort is much higher. The responses of the respondents staying in room no. 5 turned out to be a surprise. It is the only room where PMV calculated on the basis of the questionnaires was positive, and according to the ISO 7730 standard, the result was negative. The respondents' responses showed, according to the feeling of a warmer environment, and PMV calculated according to the ISO 7730 standard showed a feeling of coolness. The actual feelings of the internal environment and the results from the norm differ from each other. Taking into account the PMV range from -0.5 to +0.5 (considered as preferable), according to Figure 5, it can be seen that the PMV calculated from the standard is met in rooms 1, 3, 5, 6 (purpure color). However, calculated according to the questionnaires, was met only in rooms 2 and 4 (purpure color). The rest  [12], the PMV for the questionnaires was 2.20, and for ISO 7730, from which the PMV is calculated, it is 0.96. There is a significant discrepancy between the responses of people and the norm, which exceeded the range of -0.5 to +0.5. However, for example, in the publication Dębska & Krakowiak [6], the only PMV that complies with the guidelines of the ISO standard is room 2, and room 1 and 3 exceed the permitted range. Taking this into account, in the latest research carried out in Figure 5, it can be noticed that out of 6 rooms, as many as 4 are within the acceptable range.
The next part of the analysis focused on PPDs calculated by means of questionnaires, taking into account the responses from the questionnaire, which included answers: 'too cold', 'too cool', 'too warm', and 'too hot'. This range represents the dissatisfaction of the subject with the conditions in the room. Dissatisfaction of the respondents -i.e. PPD values are presented in Figure 6. Referring to Figure 5, where PMV showed significant discomfort in rooms 1, 3 and 6, Figure 6 confirmed on the basis of the calculated PPD from the questionnaires that the percentage of dissatisfied people is very high compared to the PPD calculated from the standard, which assumes compliance according to the standard. In the case of room 1 and 3, the percentage of dissatisfied people was more than 50%, and in room 6 it was above 35%. On the other hand, the standard showed that only 6%, 5% and 6% of people are dissatisfied with the prevailing conditions. Taking room 2 into account, it can be seen that the norm assumes a higher percentage of dissatisfied people, equal to 13%, while according to the respondents it was 10%. In room 5, the situation is reversed because surveys assume that PPD is around 18%, and the norm is only 5%. On the other hand, in room 4, PPD was 0% according to the volunteers -which was not dissatisfied by the respondents, while the norm showed as much as 11%. Therefore, in the case of data from Figures 5 and 6, it can be concluded that the standard does not reflect the actual satisfaction and feeling of warm comfort by the respondents.
It should be noted, however, that the assessment of temperature and other indicators of the indoor environment is influenced by many factors such as clothing, fatigue, headache, body weight, pressure, etc. -therefore people may feel better in one room and others feel worse in another, because it is a very individual approach of each of these people to their subjective feelings. To better understand this process, we should do as much research as possible.

Conclusions
Summarizing the analysis of the results, 58.67% of people describe their well-being in the surveyed rooms as good (+1), thus meeting people's expectations of an intelligent building. Taking into account the PMV and PPD indicators calculated on the basis of the standard and questionnaires (Figures 5 and 6), it can be noticed that the assumptions from the standard differ from the real feelings of people. For example, in room 1, the PPD was 53.33% according to people, and the survey assumed that indoor conditions met with people's dissatisfaction in only 6%. The conclusion that comes to mind is unequivocal. The standard does not take into account the individual feelings of people, but only the parameters of a given room. Where people feel good, the standard will assume that they feel bad. Likewise, when people feel unwell, the standard will assume that both PMV and PPD will be normal. Therefore, the best solution is to work on the improvement of the standard.