Test Rig Design Considerations to Detect Volatile Organic Compounds in Aircraft Cabins

Abstract

Abstract A numerical investigation for simulating the aircraft cabin as an environmental chamber is set to assist a test rig design assimilating passenger comfort, considering their exposure to high concentrations of Volatile Organic Compounds. Computational Fluid Dynamics is used to evaluate the flow inside the cabin for 800 sec of actual flow time, where the mixing and transport of chemical species are also evaluated. Measurements close to the passengers’ noses are used to create a Boruta feature selection-based dataset that trains four machine learning classifiers, namely, Support Vector Machine, Random Forest, Naive Bayes, and Logistic Regression, and compares their performance. Furthermore, the evaluation of molecular weight impact on residence time is explored, with an additional simulation including cabin filters. The model is proven to be insensitive to inlet air mass flow variation, indicating that the air-conditioning system mass flow has a minor impact on chemical species mass measurements. The Naive Bayes classifier shows the greatest performance with 96 % accuracy and is being selected to create a digital nose model. Moreover, when comparing simulation results between the models with and without cabin filters, results indicate that the residence time is independent of each compound’s molecular weight, with all showing equivalent residence time reduction. Finally, the observed cabin flow irregularities indicate that passengers may share different comfort experiences during the flight. This dictates the need to manufacture a full-scale test rig to quantify the impact of the flow asymmetry on the comfort of frequent travelers and aviation professionals.