Roorkee: Researchers at the Central Building Research Institute (CBRI) in Roorkee have developed an Artificial Intelligence (AI) model to predict the “probability of transmission” of Covid-19 in an enclosed space in a building.
The model uses an electronic device to detect the concentration of carbon dioxide, the temperature and the humidity of a room. These and other input parameters are used to show the probability of the presence of the COVID-19 virus in an office, classroom, or any other enclosed space in a building.
After calculating the parameters, the software determines the transmission probability and displays the results as a text alert on the screen. The study’s findings, called “Probability of transmission of SARS-Cov-2 in the office environment using an artificial neural network,” was recently published by IEEE Access, an open access scientific journal in the United States. Anuj Kumar, lead researcher and one of the study’s authors, called the research “the first of its kind.”
According to the study, 11 input parameters were used to predict the R-value, which refers to the “expected number of new infections arising from any event occurring over a total time in any space.” Parameters are listed as: indoor temperature (TIn), indoor relative humidity (RHIn), opening area (AO), number of occupants (O), area per person (AP), volume per person (VP), CO2 concentration (CO2), Air Quality Index (AQI), Outdoor Wind Speed (WS), Outdoor Temperature (TOut) and Outdoor Humidity (RHOut).
“Using a data logger, we take readings for CO2 concentration, temperature and humidity. Other equipment gives us readings for wind speed and AQI. Some of the other parameters are calculated manually,” said Kumar, who he is also a consultant in Building Energy. Department of efficiency of the institute.
According to the study, “real-time office environment data was collected in spring 2022 in a naturally ventilated office room in Roorkee under compound weather conditions.”
The data was fed into two models, an artificial neural network (ANN), a computer-based mathematical model, and one using more traditional techniques, the curve-fitting model, a mathematical analysis tool.
“We determined the correlation coefficients for both models, 0.9992 for ANN and 0.9557 for curve fitting. As the value of these coefficients decreases (for example, 0.90992 for ANN), the chances of transmission of the viruses are increasing,” Kumar said. “We established a link between the CO2 concentration and the event R as a model for prediction purposes,” he added.
