Abstract:This work presents a numerical investigation targeting to simulate the slice of a small aircraft cabin as an experimental facility with a controlled environment, to assess passenger comfort when exposed to high volatile organic compound (VOC) concentrations. The mixing and transport of chemical species are evaluated using computational fluid dynamics for 800 s of in-cabin actual flow time and measurements are taken every 10 s from selected computational nodes close to the passengers' noses. The results are used to create a dataset that trains four different machine learning classifiers, namely, the Random Forest, Support Vector Machine, Logistic Regression and Naive Bayes, and their performance is compared. Moreover, an additional simulation of the cabin with a filtering system utilising high-efficiency particulate air and activated carbon filters is conducted, to evaluate the impact of the molecular weight of the compounds on their residence time, and compare it to the simulation without the filters. Results indicate that the model is insensitive to the inlet air mass flow variation and that the mass of the VOCs measured in the monitored computational nodes remains relatively unaffected, meaning that the impact of the air-conditioning system setting is minor. Additionally, a Boruta feature selection algorithm is used to determine the importance of each measurement of the simulation and to form a dataset that will train the four machine learning classifiers. Furthermore, the comparison of the two simulations, the one with and the one without the filters, indicates that the residence time (RT) of the compounds is independent of their molecular weight, as they all show equivalent percentile reductions, with the naphthalene and styrene showing a 28.5% and 28.3% reduction respectively, compared to the simulation without the filters. Finally, in-cabin flow irregularities are present, disrupting the flow symmetry and suggesting that not all passengers share the same traveling experience.

Prediction on Air-Nasal Mucus Partition Coefficients of Odor Compounds

The Experimental Research on Mimetic Optical Artificial Nose Sensor System and Detection of Respiration Odor

An in Vivo Bioelectronic Nose for Possible Quantitative Evaluation of Odor Masking Using M/T Cell Spatial Response Patterns.

Prediction of Methane Adsorption Isotherms in Metal–organic Frameworks by Neural Network Synergistic with Classical Density Functional Theory

Odor Prediction and Aroma Mixture Design Using Machine Learning Model and Molecular Surface Charge Density Profiles

Odor sensory tests vs. In-silico prediction for the high-definition quantification of olfaction dynamics

A Local Weighted Nearest Neighbor Algorithm and a Weighted and Constrained Least-Squared Method for Mixed Odor Analysis by Electronic Nose Systems

Prediction of Octanol–air Partition Coefficients of Semivolatile Organic Compounds Based on Molecular Connectivity Index

Prediction of biphasic separation in CO2 absorption using a molecular surface information-based machine learning model.

Development of a computational fluid dynamics model for mucociliary clearance in the nasal cavity

Olfactory Perception Prediction Model Inspired by Olfactory Lateral Inhibition and Deep Feature Combination.

Comprehensively assessing priority odorants emitted from swine slurry combining nontarget screening with olfactory threshold prediction

Prediction of Air Quality Perception in Aircraft Cabin Based on Psychophysical Model and Artificial Neural Network (Ann)-Based Model

Prediction of mass spectrometry ionization efficiency based on COSMO-RS and machine learning algorithms

Modelling the octanol-air partition coefficient of aromatic pollutants based on the solvation free energy and the dimer effect

A Novel Electronic Nose Using Biomimetic Spiking Neural Network for Mixed Gas Recognition

Advanced analysis of the adsorption mechanism for sweet odorant on mouse and human olfactory receptors via statistical physics theory

Molecular Partition Coefficient from Machine Learning with Polarization and Entropy Embedded Atom- Centered Symmetry Functions

Odour-sensitive passenger comfort in small aircraft cabins

Determination of Odor Intensity of Binary Gas Mixtures Using Perceptual Models and an Electronic Nose Combined with Fuzzy Logic

Kinetic model to predict the absorption of nasally applied drugs from in vitro transcellular permeability of drugs