A Numerical Analysis of the Air Distribution System for the Ventilation of the Crew Quarters on board of the International Space Station

Quality of life on the International Space Station (ISS) has become more and more important, since the time spent by astronauts outside the terrestrial atmosphere has increased in the last years. The actual concept for the Crew Quarters (CQ) have demonstrated the possibility of a personal space for sleep and free time activities in which the noise levels are lower, but not enough, compared to the noisy ISS isle way. However, there are several issues that needs to be improved to increase the performance of CQ. Our project QUEST is intended to propose a new concept of CQ in which we will correct these issues, like the noise levels will be lower, more space for astronaut, increased thermal comfort, reduce the CQ total weight, higher efficiency for the air distribution, personalized ventilation system in CQ for the crew members in order to remove CO2 from the breathing zone. This paper presents a CFD study in which we are comparing the actual and a proposed ventilation solution for introducing the air in CQ. A preliminary numerical model of the present configuration of the air distribution system of the Crew Quarters on board of the ISS, shows the need for an improved air distribution inside these enclosures. Lower velocity values at the inlet diffuser, distributed over a larger surface, as well as diffusers with improved induction would appear to be a better choice. This was confirmed through the development of a new model including linear diffusers with a larger discharge surface. In this new configuration, the regions of possible draught are dramatically reduced. The overall distributions of the velocity magnitudes displaying more uniform, lower values, in the same time with more uniform temperatures. All these observations allow us to consider a better mixing of the air inside the


Introduction
Space is one of the most extreme environments imaginable.Outside ISS, the wall temperature will vary from 121℃ (Sun facing side) to -157℃ (dark side) [1].Due to the fact that the time spent by astronauts on ISS has increased over the years, quality of life has become a matter that needs to be improved.Furthermore, future space stations and long range transportation space ships will demand for more comfortable life conditions for both crew members and travellers.Over the last two-and-a-half decades, The International Space Station's Environmental Control and Life Support System (ECLSS) has developed in size, complexity, and capability continuously from the beginning [2].ISS crew members are living and working in a high-risk environment, and their comfort plays a major role in ISS proper operation, because a rested crew member operating in a comfortable environment is less prone to human error.
Nowadays, on the ISS operates several personal Crew Quarters (CQ) for the crew members [3][4][5].The most advanced form all CQ are four of them which are installed in the Harmony Node (also known as Node 2) [6].Functionally, the role of a CQ is to provide a quiet private place in which crew personnel can sleep or relax in the free time (Figure 1 and Figure 2).However, there are several issues that needs to be improved to increase the performance of CQ.NASA and ESA admitted that on ISS, higher noise levels, CO2 and dust accumulation are frequently occurring issues [7,8].From our observations we believe that the current CQ design can be improved in order to reduce the noise levels, increase the personal space for astronaut, increased thermal comfort, reduce the CQ total weight, higher efficiency for the air distribution, personalized ventilation system in CQ for the crew members in order to remove CO2 from the breathing zone.These improvements are the objective of the QUEST research project developed by our research team.
One of the goals of our project is to understand the flows occurring in the actual design of air diffusion system of the existing CQ on the ISS.Our team has experience on studying the air flow through different types of diffusers.This paper presents a CFD study in which we are comparing the actual and a proposed ventilation solution for introducing the air in CQ [9][10][11].
In the last five decades, extensive investigations and experiments involving human subjects have resulted in methods for predicting the degree of thermal discomfort of people exposed to a certain environment.One of the most well-known and widely accepted method is Fanger's "Comfort Equation" and his practical concepts Fanger's hypothesis relies on the thermal balance between the heat produced, consumed and transferred to the environment when the human body finds itself in a state of neutrality.The most important parameters that can influence the thermal comfort state were considered to be the metabolic rate, the clothing thermal resistance, the air temperature, velocity, humidity, pressure and turbulence intensity [12][13][14][15].

MATERIAL AND METHOD
The CQ geometrical model along with the virtual manikin was built using the SolidWorks software in Figure 3. Unstructured tetrahedral meshes were generated for both studied cases.The actual ventilation solution study of the CQ (Figure 3a) was already presented in [21].This configuration of the air diffuser is composed by four rectangular regions, each of them having independent vertical and horizontal guiding vanes [6,19,20].The proposed ventilation solution is presented in Figure 3b.Ansys Fluent was the software used for the numerical modelling.For the pressure-velocity coupling, Fluent provides four segregated types of algorithms (SIMPLE, SIMPLEC, PISO, Fractional Step) and a coupled one.For this study we used coupled algorithm which offers some advantages over the segregated approach because it contains a robust and efficient single phase implementation for steady-state flows, with superior performance compared to the segregated algorithms [22].A second order upwind scheme was used to calculate the convective terms in the equations, integrated with the finite volume method.SST k-ω was the turbulence model used for the numerical simulation due to our experience in using this model for indoor air circulation [23].The reason is that in SST k-ω turbulence model, for flows, both the boundary layer as well as the free stream are considered.

The main difference between the presented solutions in
For the boundary conditions, we imposed three mass flow rates corresponding to the three positions the fan controller: low, medium and high, respectively 108 m 3 /h, 138 m 3 /h and 196 m 3 /h [20].The Reynolds numbers at the exit of the air vents based on streamwise mean velocity and on the equivalent diameter (De1= 1.144m) were Re1= 4462, Re2=5692, and Re3=8098.The inlet turbulence intensity was imposed at: 5.6%, 5.43% and 5.19%, being calculated using the empirical relation proposed by Jaramillo [24].The present configuration of the air diffuser is presented in [6,19,20] and is composed by four rectangular regions, each of them having independent vertical and horizontal guiding vanes.Inspired by [20] we considered that the air flow is deflected equally towards right and left in the horizontal plane with a 45° angle respectively, and in the same time the air flow is deflected equally towards the 'up' and 'down' in the vertical plane with a 45° angle as well.
For the cases 1, 2 and 3 we have considered that the air flow is deflected right and left on the horizontal and up and down on the vertical with a 45° angle [20].Case 1, 2 and 3 are corresponding to the flow rates of 108 m 3 /h, 138 m 3 /h and 156 m 3 /h.For the cases 4 and 5 the air was deflected 45° angle down, through an improved linear diffuser.This case is associated with an idea of a first level improvement of the air distribution system in the CQ.This way, we propose to keep the actual configuration of the inlet and the exhaust ducting parts but renounce to the ducting and abatements that are situated along the door of the CQ.This will be possible using transverse flow (crossflow) fans.The inlet and the outlet grilles inside the CQ cabin will be kept at the same locations.The inlet air grille to the Node 2 cabin will be placed on the 'upper' chamfered part of the bump-out while the exhaust air grille will be kept at the same position on the lateral flat side of the bump-out.
In the case 4 we tested the medium flow rate of 138 m 3 /h, and in the case 5 we tested the high flow rate of 156 m 3 /h.We considered to present here only the two highest flow rates that were recorded to produce problems in terms of local air speed discomfort.The inlet air temperature was in all cases 18°C corresponding to the air taken from Node 2. The heat flux from the walls was equivalent to 150 W (imposed wall temperature was 22 °C).On the different zones of the virtual manikin inside the CQ we imposed the temperatures from Table 1.

CONCLUSIONS
From our preliminary numerical study representing the testing of a model of the present configuration of the air distribution system of the Crew Quarters on board of the ISS, it appears that the need for an improved air distribution inside these enclosures is obvious.This way, lower velocity values at the inlet diffuser, distributed over a larger surface, as well as diffusers with improved induction would appear to be a better choice.This was confirmed through the development of a new model including linear diffusers with a larger discharge surface.
In this new configuration, the regions of possible draught are dramatically reduced.The overall distributions of the velocity magnitudes displaying more uniform, lower values, in the same time with more uniform temperatures.All these observations allow us to consider a better mixing of the air inside the enclosure.
Other further improvements, like a supplementary PV ventilation system, can aid in diminishing the issues with CO2 accumulation during the sleeping hours.A combination between the general ventilation system in the CQ module and a personalized ventilation system with an adjustable air diffuser will allow us to obtain good air quality near the crew member in any conditions and further improve thermal comfort by covering the need for fresh air without having to lower the overall temperature inside the CQ module.This configuration has to be tested next.
This work was supported by a grant of the Romanian space agency ROSA, QUEST -Advanced air diffusion system of the crew quarters for the ISS and deep space habitation systems, STAR-CDI-C3-2016-577

Figure 2 :
Figure 2: CQ in its actual structure [3] When the heat balance equation proposed by Fanger is satisfied, the heat generated by the human body is dissipated without having an increase or a decrease in the human body [12].PMV index values are between -3 and 3.These values quantify the average opinion of a group of subjects on the state of comfort.Associated with this parameter, is the index PPD, indicating the percentage of occupants in thermal discomfort.Fanger and Christensen [17] proposed correlations for the local turbulence intensity of the indoor air flow with the thermal discomfort sensation in an index called "Draught Rate" (DR).The Draught Rate is an index that depends on the mean velocity and temperature but also on the air turbulence intensity value.DR values are divided in four regions of values corresponding to a classification of the indoor ambiance regarding its comfort level.This way, a values of DR≤15% are associated to the zones with a high standing quality of the ambiance, values of 15%≤ DR≤20% are corresponding to a good quality of the ambiance, values 20%≤ DR≤25% are corresponding to an acceptable quality of the ambiance and values of DR>25% correspond to unacceptable conditions to regions where the sensation of draught might create a serious discomfort.More information on thermal comfort

Figure 3
is the replacement of the axial fan solution with a tangential fan solution.a. b.

Figures 5 ,Figure 5 :Figure 6 :Figure 7 :Figure 8 :Figure 9 :
Figures 5, 6 and 7 are presenting the distributions of the velocity magnitude, air temperature, of the in-plane vectors, and of the predicted percentage of dissatisfied regarding the draught sensation in the coronal plane of the virtual manikin in the first three cases (cases 1, 2 and 3).These three cases are corresponding to the current situation of the air distribution system in the CQ on board of the ISS.