Research On Measurement Mothod Of The Concentration And Size Distribution Of Indoor Suspended Particulate Matters

According to the survey, the urban residents spend 90% everyday in indoor environment. The suspended particulate matters is one of the most important sources of pollution indoor, which is sucked into the human respiratory tract, and deposit in the lower respiratory tract deposition site under the throat. The smaller the particle size, the deeper the location of respiratory tract, and thus the greater the harm to humans, then the higher morbidity and mortality. The conentration of particles determines the inhaled dose of people, and the particle size carries the important physical and chemical information of particle which is related to its deposition, detention and elimilation in the human respiratory. By researching the concentration and size distribution of indoor suspended particulate matters ,we can not only understand the impact factorsto particles, but also take the positive measures to improve indoor air quality and reduce the harm to the human health.This dissertation researches a measurement method to the concentration and size distribution of indoor suspended particulate matters. The relevant theories and methods have been put forward, and validated by experiment.Firstly, a new image threshold segmentation method based on the genetic algorithm and Otsu method has been put forward, which can obtain the segmentation threshold by the global optimization. The experimental results show that the algorithm has strong robustness and self-adaptive characteristics, and preserves better image details. Because the filtering effect of median filter is affected by the noise level, and that of opening and closing morphological filter depends on the shape and size of structural elements seriously, a new adaptive filtering method has been put forward based on the binary morphological filtering and median filtering method to adjust the size and shape of the neighborhood template and the structural elements by genetic algorithm. This dissertation also has put forward the mathematical morphology edge detection methods with multi-scale and multi-structural elements combined morphological operations with set operations. It can achieve a variety of particle image edge detection with the edge detection operator of different scales structural elements.Secondly,in order to provide the experimental datas to categorize according to the particle shape, this dissertation researched a identification algorithm to the main morphological parameters of a single indoor suspended particulate matter such as size, shape coefficient and the fractal dimension, then analyzed morphological characteristics about the ductility, the multilateral morphology and boundary curve. The experimental results show that the size distribution of indoor suspended particulate matters appears continuous multi-peak curve in rough coarse mode mainly, and most particles with 10μm diameter below are the most dangerous. The overall shape of suspended particulate matters is similar to the block, not the sharp diamond and long rectangular.Aftr that, adopting the shape coefficient and the fractal dimension as observational data of each type, a discriminant analysis classified method of particles has been put forward which can estimate the shape, volume and surface area. By analysing the sources of particles and estimating the density of particles, the mass concentration, total surface area and volume concentration and particle size distribution of indoor suspended particulate matters can be obtained at the same time.Finally, according to the real-time changes , the concentration and size distribution of indoor suspended particulate matters are predicted and analysed. The dissertation Analysed the factors to effect the concentration and size distribution, established the mass balance model adopting the deposition velocity, penetration factor,as well as air flow indoor as input its parameters. Combining experimental measurement data and building envelope characteristics with the concentrations outdoor particles and other related datas, the concentrations and particle size distribution of indoor suspended particulate matter is predicted by simulation.