• 关键词：PM_2.5旋风切割器  性能测试  切割粒径  几何标准偏差  模拟

• 基金项目：北京市科委项目（No.Z161100000716004）；国家自然科学基金（No.41575119，41775121）；国家重点研发计划（No.2017YFC0211404）

• 阮兵
• 北京航空航天大学空间与环境学院, 北京 100191

• 李兴华
• 北京航空航天大学空间与环境学院, 北京 100191

• 谢岩
• 北京航空航天大学空间与环境学院, 北京 100191

• 王威
• 北京航空航天大学空间与环境学院, 北京 100191

• 李垚
• 北京航空航天大学空间与环境学院, 北京 100191

• 摘要：切割器是PM_2.5采样器的关键部件，其性能的优劣将直接影响到采样结果的准确性，对其性能进行测试有助于评估PM_2.5采样器采样结果的准确性和可比性.因此，本研究选取两种应用广泛的BGI公司和URG公司生产的PM_2.5旋风切割器，分别采用多分散气溶胶和单分散气溶胶标定装置对其性能进行了测试.结果表明，两种装置的测试结果基本一致.BGI和URG切割器的切割粒径D₅₀为（2.5±0.2）μm，能够满足环境空气PM_2.5采样要求（HJ 93-2013）；BGI切割器的捕集效率几何标准偏差δ_g为1.2±0.1，满足要求；而URG切割器的δ_g超出规定范围.同时，本研究获得了切割器采样流量Q与D₅₀的关系式，可通过该关系式改变切割器的Q获得所需的D₅₀；为使BGI切割器的D₅₀在规定的范围内，其流量偏差需限定在设计流量（即16.7 L· min^-1）的-1.0%~16.7%之间，而URG切割器的流量偏差需限定在设计流量的-1.3%~16.2%之间.其次，获得了切割器流动雷诺数Re_f与无量纲切割粒径D₅₀/D_C的关系式，利用该关系式可在保持设计流量不变的情况下，通过改变切割器的出口管尺寸，得到所需的D₅₀.最后，获得了切割器Q与压力损失△P的关系式，采用5种模型对两种切割器的无量纲阻力系数K_L进行模拟计算，并与测试计算结果进行比较，发现Casal&Martinez模型、Coker模型和Demir模型更适用于BGI旋风切割器的△P预测，而Shepherd&Lapple模型和Casal&Martinez模型更适用于URG旋风切割器的△P预测.

• Abstract：The size selectors is the key component of PM_2.5 sampler and its performance directly affects the accuracy of sampling results, and measurement its performance is helpful to evaluate the accuracy and comparability of sampling results from PM_2.5 sampler. In this study, two widely used PM_2.5 cyclones, produced by BGI company and URG company, were selected, and their performances were tested by using polydisperse aerosol and monodisperse aerosol calibration systems, respectively. The results from two systems were consistent with each other. Cutting diameter (D₅₀) of BGI cyclone and URG cyclone is in the range of (2.5±0.2) μm, and meets the requirement of ambient air PM_2.5 sampling (HJ 93-2013). Geometric standard deviation (δ_g) of collection efficiency from BGI cyclone is in the range of 1.2±0.1, and meets the requirement. However, δ_g of URG cyclone exceeds the range of the requirement. The relationship between sampling flow rate (Q) and D₅₀ were determined. The specific D₅₀ can be obtained by changing Q based on their relationship. When D₅₀ of the BGI and URG cyclones meets the requirement, deviation ranges of their Q should be in -1.0% to +16.7%, -1.3% to +16.2%, respectively. The relationship between flow Reynolds number (Re_f) and dimensionless cutting diameter (D₅₀/D_C) were determined. The specific D₅₀ can be obtained by changing cyclone outlet tube diameter but maintaining the same Q based on their relationship. The relationship between Q and cyclone pressure loss (△P) were determined. Five models were applied to simulate dimensionless loss coefficient (K_L), and the simulation values and test results were compared. Casal & Martinez model, Coker model and Demir model show a good prediction on △P of BGI cyclone, while Shepherd & Lapple model and Casal & Martinez model is more suitable for prediction △P of URG cyclone.

• Key words：PM_2.5 cyclone  performance testing  cutting diameter  geometric standard deviation  modeling

• 
  摘要点击次数： 1114 全文下载次数： 1779