Establishment and Operation of a Simulation Stack for Proficiency Test of Particulate Matter Emissions

Abstract Accurate quantification of particulate matter (PM) emissions from stationary sources remains challenging due to the dynamic behavior of airborne particles. Although standardized PM sampling methods exist, a formal proficiency testing (PT) framework for validating measurement results has been lacking. To address this gap, a simulation stack with controllable flue gas velocities (3–18 m s −1 ) and dust concentrations (3–300 mg Nm − 3) was developed, enabling a comprehensive evaluation of the entire sampling procedure. Portland Type I cement (count median diameter: 2.6 μm; geometric standard deviation: 1.8) was used as the challenge aerosol. This study statistically evaluates PT-PME (Proficiency Testing for Particulate Matter Emissions) data from 2019 to 2023 to assess system stability and long-term trends. The dust delivery system achieved an adjustable feed rate of 1.1–16.7 g min −1 with a coefficient of variation (CV) of only 2.51% at its maximum rate, indicating high transport precision. CVs of stack gas velocity and temperature were consistently below 3%. Regression analysis revealed strong linearity between the measured flue gas velocity and the set flue gas velocity controlled by the PT-PME system (R 2 = 0.985), as well as between the measured particulate mass concentration and the normalized dust delivery rate (DS) (R 2 > 0.96). Z-score evaluation revealed that 7.01% of velocity and 2.3% of mass concentration measurements were unsatisfactory. Long-term analysis indicated reductions in the root mean square error (RMSE) since 2003, with values decreasing from approximately 20% to 8% for flue gas velocity and from approximately 30% to below 20% for mass concentration. Overall, the PT-PME program significantly improves stack emission measurement accuracy and provides a reliable reference for air quality monitoring and regulatory compliance. Graphical abstract