Performance and Uncertainty of Portable Emissions Measurement Systems (PEMS) in Laboratory and On-Road Tests for Improved Quantification of Vehicle Exhaust Particles
MOHSEN KAZEMIMANESH, Jorge Saturno, Kim Winther, Rasmus Pettinen, National Physical Laboratory
Abstract Number: 294
Working Group: Combustion
Abstract
Ultrafine particles (UFP) and gaseous pollutants emitted from vehicle exhausts are major contributors to air pollution in urban areas. UFPs are especially concerning for human health due to their deeper penetration in the lungs. In the EU, the number of solid exhaust particles larger than 23 nm is regulated for on-road type approval of vehicles by testing real driving emissions (RDE) using portable emissions measurement systems (PEMS). However, PEMS are likely to have larger measurement uncertainty due to simpler design, while metrological validation of PEMS is currently lacking. Thus, it is critical to understand this uncertainty to underpin the current and future conformity factors.
The performance and uncertainties of the PEMS devices were evaluated by setting up a test matrix consisting of both in-laboratory and RDE tests. The PEMS devices were evaluated between 2021 and 2023 with four Euro 6 emission class passenger cars and in two countries, Finland and Denmark. During this study four PEMS devices were evaluated, three commercially available PEMS devices and a ‘GoldenPEMS’ (which was validated metrologically by several European National Metrology Institutes, NMIs). The PEMS devices used condensation particle counting (CPC) and diffusion charging (DC) technologies for particle number (PN) measurement.
The results showed that a significant day-to-day variation exists in RDE tests, with the same vehicle, test route, and driver. Day-to-day variations are mostly larger than device-to-device variation. Measured PN can vary by more than ± 50% depending on the PN measurement system. The performance of the tested PN-PEMS analysers varied depending on vehicle type. The RDE results recorded with DC-based PEMS correlated better with laboratory results for tests producing lower PN emissions, while the CPC-based PEMS performed better for tests producing higher PN emissions. The results suggest that the two PN PEMS devices may have to be tuned for different emission matrixes.