Hospital Airflow Dynamics: A CFD Study on Ventilation Efficiency and Pathogen Dispersion

Project Description:

This project will simulate a patient and a doctor in a hospital room. Three cases are involved: In the first case, the patient is sneezing. In the second one, the patient is coughing; in the last case, he is just breathing. Note that the door is always closed.

There are two inlets in the room. The side inlet has a 100 cfm flow rate, and the air conditioner placed on the ceiling has a 700 cfm flow rate. Two lamps have a 40 w/m² heat flux. To satisfy mass conservation, we need an outlet. A negative pressure outlet type with a -25 Pa value is defined, as shown in the figure. Also, the door is closed.

Geometry and Mesh:

The geometry was created using SpaceClaim software, which has 2 lamps, a doctor, a patient, an air conditioner, an inlet, and an outlet. Then, the Geometry import was performed in Fluent Meshing software and generated polyhedral cells for the hospital.

CFD Simulation:

The simulation can be performed in one-way regarding the particle`s diameter (1 micrometer). Furthermore, the Saffman lift force, Thermophoretic & Brownian force should be considered. Additionally, the Rosin-Rammler diameter distribution is employed. The injection characteristics will be explained in the results section.

The simulation process has two significant steps. Firstly, we need to simulate the flow & turbulence equations in steady-state form. After convergence, we can start injecting particles (transiently) and turn off the flow and turbulence equations.

Results:

Sneezing

In this case, the particle injection lasts for 0.1 seconds, with a velocity of 32.8 m/s and a mass flow rate of 0.0028 kg/s. The mid-plane velocity contour illustrates how the air conditioner can produce a flow pattern in a steady-state condition that cannot carry particles through the outlet (see the figure below).

The weakness of the ventilation system becomes clearer when the pathlines are displayed. The injected particles almost trap inside a circular path, which is inappropriate considering our purpose. This is fully evident in the animations and pathline contours. Moreover, the “mass in the domain” plot indicates that after 3 minutes, more than 95% of the particles leave the room or get trapped. However, the ”escaped particles” plot proves that just 2-3% of the particles were trapped.

Coughing

In the case of coughing, however, the injection velocity decreases in contrast to sneezing (18 m/s), and the mass flow rate increases to 0.005 kg/s. It also lasts for 0.1 seconds. Actually, the flow pattern is the same as before, so it is highly expected to have similar behavior.

The “mass in the domain” report indicates that, once again, particles start escaping from the domain after approximately 30 seconds. After 3 minutes, 90% of the particles had either escaped or become trapped. Actually, just about 1% of them were trapped, and others escaped.

Breathing

In this case, we also need to consider flow equations in transient simulation due to the nature of breathing, which affects the domain consecutively. The patient exhales every 2.5 seconds, so we needed a User-Defined Function (UDF) to apply the condition. The following plot is a report of the velocity inlet from the mouth of the patient.

The pathline contours below clearly exhibit the flow pattern and several circulations through the outlet.

As can be seen in the previous plot, the particles were injected with a velocity of 0.25 m/s every 2.5 seconds and remained for 2.5 seconds. The mass flow rate is 0.000022 kg/s. As a result, the following “Mass in the domain” & “Escaped Mass” is achieved. Notice that the mouth boundary has an “escape” DPM condition, allowing particles to escape from the mouth as well. As a result, most particles escape from the mouth while the patient inhales. This leads to a maximum mass of 0.0012 kg in the domain, which occurs after 150 seconds.