Transient particle transport prediction based on lattice Boltzmann method-based large eddy simulation and Markov chain model

Fast and accurate prediction of particle transport is essential for the determination of as-needed mitigation strategies to improve indoor air quality. Several methods have been proposed to achieve this goal. However, they mainly based on the Reynolds-averaged Navier-Stokes (RANS) approach, which may affect the accuracy of particle calculations. Considering the lattice Boltzmann method (LBM) can execute high-speed large eddy simulation (LES) while Markov chain model performs well for particle calculations. This study proposed an LBM-LES-Markov-chain framework for indoor transient particle transport simulation. The performance of the proposed framework was investigated in a two-zone ventilated chamber, and compared to the CFD-LES based models. Results show that the proposed framework is as accurate as but faster than the CFD-LES based models. The mean normalized root-mean-square deviations of the proposed model is 12%, similar to the CFD-LES-Lagrangian (15%) and CFD-LES-Eulerian (13%) models. The computing time of the proposed model is 5.66 h, shorter than the CFD-LES-Lagrangian (153 h) and CFD-LES-Eulerian (15.03 h) models. Furthermore, we further compared the framework with CFD-RNG based Markov chain model, CFD-RANS based models, and FFD-Markov-chain model and found that it is an alternative for the fast prediction of indoor particle concentration.