Numerical Simulation Of Deposition Characteristics Of Fine Particles In Multi-stage Bifurcation Networks Of Lung

Multi-stage bifurcation is a kind of network structure widely existing in nature,such as gas ventilation and air conditioning pipelines,water supply and drainage networks,multi-stage rivers and human respiratory systems.Due to the characteristic scale of bifurcation networks ranges several orders,a large number of related studies have been carried out.However,due to the limitation of its numerous stages of bifurcations,in general,the flow characteristics of local pipelines can only be studied thoroughly.Once the flow and particulate matter are coexisted in the system,the number of particles that need to be simulated will increase exponentially with the number of levels.To solve this type of transport problem becomes more difficult.The lower respiratory tract of the human body is such a typical network with this bifurcation structure.It is inevitable to solve this dilemma when the transport and sedimentation of fine particles,such as PM2.5,is considered.Based on the symmetry of the network structure and the average statistical characteristics of particles,a doubling and halving algorithm through the particles' splitting and merging was proposed in this paper,which can greatly reduce the simulation cost and make it possible to simulate the transportation in pipe system with multi-stage bifurcation.The main work is as follows:First of all,on the basis of previous research on the flow field in the respiratory area by our research group,this paper focused the research on the characteristics of the particle transport in the respiratory area.It is found that to realize a full-scale simulation requires a very high computational performance.If the number of particles released from the inlet is limited,the number of particles that reach the end of the respiratoryzone cannot meet the requirement of statistics.Therefore,the characteristics of particle deposition across the whole respiratory zone cannot be studied.While,by increasing the number of particles released from the inlet,the number of particles that may reach the end of the respiratory zone can increase,the amount of particles is too small to reflect the statistical behavior and also impossible to characterize the sedimentation characteristics accurately.In order to solve the problem that the large number of particles is required for simulation,a new algorithm needs to be developed.Secondly,for the problem of particle deposition in a typical three-stage bifurcation network,this paper develops particle doubling and halving algorithms based on particle splitting and merging.When the particles move from the main channel to the downstream branch,a particle doubling algorithm is used.Compared with the method without doubling the number particles released from the inlet,only about half of particles in the new method needs to release from the inlet and can realize the simulation.At the same time,the number density of deposited particles in different regions is approximately the same,and the overall deposition rate obtained by two methods is almost same.For multi-stage bifurcated pipelines,the more particles need to be released at the inlet,the more advantage the doubling algorithm will have during the simulation.Correspondingly,a halving algorithm by particle merging is adopted when the particles move from the branches to the main road,where the particles merge in the upstream channel and the number is reduced to one half.Under the premise of satisfying the statistical requirements,the trend of the deposition rate of particles on the wall can be reliably analyzed by studying a small amount of particles.Finally,for the problem of pulmonary particle transport with rhythmic respiratory characteristics,the particle doubling algorithm proposed in this paper is used for the study.In the first half cycle of the simulation,a small amount of particles are released from the inlet,the particle doubling algorithm is used to obtain the deposition characteristics of the particles in the respiratory zone.Subsequently,the number of particles in the followed half period is kept unchanged by introducing the piecewise function.The results show that although only a small amount of particles need be released at the original stage of this algorithm,the number of particles that reach the end of the respiratory area can still meet the statistical requirements and the computingresources was saved.The particle doubling and halving algorithm based on particle splitting and merging is used as a new simulation algorithm to solve the dilemma between the requirement for the number of particles and the huge amount of computation in the simulation of multi-level bifurcation systems.This method can also be applied to various types of networks having multi-stage bifurcation.More number of particles is required in modeling,more obviously the advantages of the present algorithm will have.