Study On The Variation And Prediction Of Ultrafine Particle Concentration In The Micro-environment Vehicle

With the rapid development of China's automobile industry,automobiles have become an indispensable means of transportation for people,but their exhaust emissions also cause air pollution.The exchange of air inside and outside the car and the release of toxic components in the interior decoration materials cause the driver and passengers to be exposed to high concentrations of pollutants,affecting attention and mood,and causing illness or even death in severe cases.In particular,ultrafine particles,due to their small particle size,can not only deposit in the lungs for a long time,but also can enter the blood through the alveoli,which is more likely to cause respiratory diseases and cardiovascular diseases.However,the current research on ultrafine particles in the car mainly focuses on concentration monitoring and analysis of influencing factors,and can only provide some suggestions for reducing the exposure of people in the car.The prediction of the concentration of ultrafine particles in the car is beneficial to the driver to accurately grasp the pollution situation inside the vehicle and take effective control measures at an appropriate time to reduce the exposure level of the personnel inside the vehicle.In this study,real-time dynamic monitoring of the ultrafine particle concentration of the particle size range of 20 nm to 1?m,vehicle ventilation mode,vehicle interior temperature and relative humidity was carried out to analyze the variation characteristics and exposure degree of the ultrafine particle concentration in the vehicle.Then,the multivariate linear regression model is used to determine the significant factors affecting the concentration of ultrafine particle in the car and its contribution rate.Finally,two support vector regression models with and without factor lag are established to predict the concentration of ultrafine particle in the vehicle,and the prediction accuracy of the two models is compared.The results show that the concentration of ultrafine particle in the vehicle has obvious spatial and temporal characteristics,which vary with the road environment,surrounding vehicles and traffic conditions.Moreover,the average ultrafine particle concentration(4.87×104±790.0 cm^-3)in the vehicle during working hours is higher than that during the off-duty period(3.21×104±489.9 cm^-3),and the exposure of the in-vehicle personnel is more serious.The multiple linear regression model determined the concentration of ultrafine particle outside the vehicle,the interior temperature,relative humidity and vehicle ventilation were the significant factors affecting the concentration of ultrafine particle in the vehicle.The contribution rates were 10.7%,35.8%,3.3%and 33.5%,respectively.Among them,the concentration of ultrafine particle in the car and the concentration of ultrafine particle outside the car,the relative humidity in the car is positively correlated,and negatively correlated with temperature.The inner circulation open air conditioning ventilation mode has the best purification effect,and the outer circulation open air conditioning ventilation mode has the fastest purification speed.Finally,by comparing the two prediction models of in-vehicle ultrafine particle concentration,it is found that the average relative error and the decision coefficient of the prediction model considering the lag period are 3.13%and 0.91,respectively.The prediction accuracy is higher and the generalization ability is stronger.Therefore,the support vector regression model considering the factor lag period has a better predictive effect on the concentration of ultrafine particle in the vehicle.The research results are helpful to grasp the exposure degree of the current micro-environmental in-vehicle ultrafine particle concentration.It also provides a feasible method for predicting the concentration of ultrafine particle in the car.The driver can take effective control measures to improve the air quality inside the vehicle according to the real-time information on the quantity of ultrafine particles in the car.