Quantitative Assessment of Indoor CO 2 Concentration of a Comprehensive Office Building

. Abstract. To better understand the extent of different ventilation strategies, a set of multi-zone models of a comprehensive office building in Harbin were conducted based on CONTAMW. The main purpose aims to assess various factors affecting the concentration of indoor CO 2 in real situations by building network models, in order to seek appropriate ventilation control strategies to improve indoor air quality. Firstly, under the combined action of stack effect and wind pressure, the model was verified by field measurement. Secondly, CO 2 concentration peak value, attenuation rate and air exchange rate were analysed in conference rooms on different floors under typical seasons with the doors and windows closed. Moreover, different doors and windows opening schedules of conference rooms were set during the meeting to reduce CO 2 concentration. Results demonstrate that CO 2 concentration in conference rooms are affected by many factors including different climatic parameters, height of building, building envelope leaking characteristics, occupant participated ventilation behaviour etc. and multi-zone simulation analysis shows the necessity of its application at comprehensive building. The study highlights the need for effective guiding significance of ventilation control to reduce the concentration of indoor CO 2 in areas where there are more people indoors in short period of time, which is also meaningful for the personnel health and building design.


Introduction
At present, the current indoor air quality standards use CO2 as the key control pollutants, and its content has become an important indicator for evaluating indoor air quality [1].Shendell et al. pointed out that too high CO2 content in the air of the classroom had a significant inhibitory effect on learning and working efficiency [2].R.Nagy studied the influence of indoor mechanical ventilation and natural ventilation on building energy consumption and indoor environmental quality.The results show that the CO2 concentration is higher than the recommended value, which will cause fatigue and other symptoms [3].Liu used the network model to simulated the influence of air purifier on indoor CO2 concentration, and then proposed suggestions for optimizing the quality of indoor air environment [4].
Occupants in inner rooms often sustain low work efficiency with drowsiness.Therefore, effective guiding significance of ventilation control to reduce the concentration of indoor CO2 in areas where there are more occupants indoors in short period of time has necessity.In this paper, the multi-zone model was used to analyze the air flow and pollutant diffusion of a complex building, and then reasonable ventilation strategies and suggestions were put forward.This study can be extended to study of air flow and pollutant diffusion in various types of complex buildings, which has important theoretical and applied significance.

Building description
In this paper, CONTAMW was used in a comprehensive office building in Harbin.The building has five floors with area of 10600 m 2 .The height of the building is 15m.The building also contains a three-story atrium.A schematic of the third floor is shown in Fig. 1.Case studies mainly focused on four conference rooms from the second to the fourth floor in the building.As shown in Fig. 1, the conference rooms (ROOM A and ROOM C) located on the third floor which are adjacent to an atrium, the ROOM B (below ROOM C) located on the second floor is the same, and ROOM D (above ROOM C) located on the fourth floor is connected with the outside.The volume of ROOM A is 294 m³ , the other conference rooms have the same volume 108m³ with three-meter-height.The size of inner door and window are the same in four conference rooms.

Model validation
In this study, some field measurements including air volume and pressure drop across a couple of airflow paths and rooms were tested to verify the effectiveness of multi-zone simulation in winter and spring.In order to ensure the consistency between the test values and the simulation values, model leakage characteristics settings were partially adjusted.
Moreover, measurement of CO2 was carried out in ROOM A at outdoor temperature (3℃), wind direction (SE) and wind speed(4m/s).The meeting lasted an hour from 19:00 to 20:00 with 20 participants, meantime, the doors and windows were closed.After the meeting, the doors and windows still maintained closed for one hour.Kept the doors opened from 21:00 to 21:15 until the measurement and simulation were finished.The simulation results agreed well with the tested CO2 concentrations (Fig. 2.) and airflow rates.A brief description of instruments used in measurements is included as Tab.1.

Simulation conditions
During the meeting, the CO2 concentrations in relatively small ROOM B, ROOM C and ROOM D were more likely to exceed the standard limit (1000 ppm), so indepth simulation studies were carried out in three conference rooms.The indoor and outdoor parameters of typical seasons are shown in Tab.2.According to measured data, it was reasonable to take the average value of 400 ppm for the initial concentration of CO2 in the region and outdoors.In this study, leakage model, orifice model, shaft model and stairwell model were selected for different paths.Since there are few data on the wall leakage characteristics in China, and the wall of the building has strong tightness.The wall leakage was ignored in this study.The leakage characteristic value of the building component is shown in Tab.3.
In the study, the meeting was from 8:00 to 9:00 for 20 participants.For each participant, the amount of CO2 produced is 0.0052 L/s [5].Both stack effect and wind pressure had been considered in research.

Results and discussion
From Fig. 3. and Fig. 5, it could be noted that the peak value of CO2 concentration in winter was less than in summer under the combined action of stack effect and wind pressure.After the meeting, the indoor CO2 attenuation rate in summer was smaller than that in winter.The reason was that stack effect in winter was stronger than that in summer.Both Fig. 3. and Fig. 4. showed that the attenuation rate of CO2 concentration in ROOM B was smaller than the ROOM C. In addition, it could be seen shown from Fig. 4, during the meeting in winter, the window was opened from 8:15 to 8:25, the peak value of CO2 concentration on ROOM D could reach 2500 ppm which was much lower than that without window opening.However, since there were no outer windows in ROOM B and ROOM C, only windows connected to the atrium, CO2 concentrations did not change significantly.Condition d was the most unfavourable condition from analysis.It could be shown from Fig. 6, in transition seasons, it took a long time to return to the initial value even if the doors and windows opened in ROOM B and ROOM C. It was better to adopt mechanical ventilation measures to decrease CO2 concentration.With the doors and windows of the conference rooms closed, the effect on the concentration of CO2 in rooms was more severe than other scenarios.
Under four conditions, the CO2 concentrations in ROOM D were shown in Fig. 7.In winter, the peak value of CO2 concentration in ROOM D could reach 3200 ppm, while in summer, was up to 3500 ppm.The CO2 concentration at 12:00 (three hours after the meeting) reduced to 1000 ppm in winter.After the meeting, the indoor CO2 attenuation rate in summer was smaller than that in winter.As shown in Fig. 9, the air exchange rates of three conference rooms in winter were all greater than those in summer.ROOM D connected with the outside，the air exchange rate was greater than others.The air exchange rates in ROOM B and ROOM C were smaller under four conditions as a result of being located in the inner area.In addition, the air exchange rate in ROOM B was smaller than that in ROOM C. The reason was that ROOM B was nearer neutral level.

Conclusions
This study focused on assessing the indoor air quality of different factors including climatic parameters, height of building, building envelope leaking characteristics, occupant participated ventilation behaviour etc. Building air tightness data were determined through field testing, which were then used to create a comprehensive building model.Through the simulation of four different conditions, the CO2 concentration peak value, attenuation rate and air exchange rate were analysed.Main conclusions are following: • With the doors and windows closed, in winter the peak value of CO2 concentration in ROOM D could reach 3200 ppm, while in summer, was up to 3500 ppm.The CO2 concentration at 12:00 (three hours after the meeting) reduced to 1000 ppm in winter.
• During the meeting in winter, the window opened for 10 minutes (8:15-8:25), the CO2 concentration in ROOM D was much lower than that without window opening.
• In the transition seasons, it took a long time to return to the initial value even if the doors and windows opened in ROOM B and ROOM C. It was better to adopt mechanical ventilation measures to decrease CO2 concentration.
• ROOM B and ROOM C were located in the inner area where was no external window connected to the outdoor, even if the ventilation strategy of opening the window for 10 minutes was adopted ， the CO2 concentration was still high.As a result, we should pay more attention to the ventilation problem of the inner area.

Fig. 9 .
Fig. 9. Air exchange rate in conference roomsUnder condition b, it could be indicated from Fig.7.andFig.8. that the concentration of ROOM D was significantly lower than that in ROOM C. The reason was that ROOM C was located in the inner area where had no external window connected to the outdoor.As shown in Fig.9, the air exchange rates of three conference rooms in winter were all greater than those in summer.ROOM D connected with the outside，the air exchange rate was greater than others.The air exchange rates in ROOM B and ROOM C were smaller under four conditions as a result of being located in the inner area.In addition, the air exchange rate in ROOM B was smaller than that in ROOM C. The reason was that ROOM B was nearer neutral level.

Table 2 .
The four simulation conditions.

Table 3 .
Leakage characteristic value of air flow path.