Measurement Uncertainty Related to Sampling Position
David Graham, Uniper
This article is based on the results of a VGB Research Project which considers the overall measurement uncertainty that can be attributed to emissions data that are reported to the authorities for compliance assessment [1]. Positional uncertainty, relating to the sampling location, is considered in this article and an associated presentation at CEM 2022 [2], but this is only one element within the overall uncertainty budget.
Continuous Emissions Monitoring Systems (CEMS) are referred to as Automated Measuring Systems (AMS) within CEN standards. AMS are subject to the CEN standard EN 14181 [3] which defi nes three Quality Assurance Levels: QAL1 (certifi cation); QAL2 (calibration) and QAL3 (control).
QAL2 requires fi ve-yearly calibration of the AMS, by an accredited test laboratory, using a Standard Reference Method (SRM) for each pollutant, as defi ned in a suite of CEN standards for the regulated pollutants. The test laboratory is required to return each year to perform an Annual Surveillance Test (AST) in which the continuing validity of the original QAL2 calibration is checked.
The IED specifi es that the 95% Confi dence Interval of a single measurement (hourly or half-hourly average) shall not exceed a defi ned percentage of the Daily Emission Limit Value (ELV), e.g., 20% of the Daily ELV for NOx
10 mg/m3
assessments are based on the Maximum Permissible Uncertainty (MPU) which is the value of the Confi dence Interval at the Daily ELV concentration, e.g., for a NOx
and SO2 limit of 50 mg/m3 . , the MPU is
Importance of Representative Sampling
Obtaining a representative sample of fl ue gas is critically important for both the AMS (CEMS) and the SRM measurement systems. The spatial deviation in pollutant concentration across a measurement plane must be taken into account when assessing the uncertainty of a measurement. Deviation from the mean at the AMS (CEMS) sampling location is corrected by the QAL2 calibration established under EN 14181 [3], provided that the SRM sampling location is fully representative. Concentration deviations from the mean value at the SRM sampling location depends on the type of SRM employed. If the SRM specifi es a grid sampling approach, as is the case for most wet chemical methods and gravimetric dust determination, it can be assumed that this spatial variation is accounted for.
. Various Quality Assurance
some circumstances, even a wet chemical SRM does not require grid sampling, as is the case for mercury when the mercury content of the fl ue gas particulate can be neglected and sampling is conducted non-isokinetically from a single point [6].
2
SRMs, and the sampling location meets the requirements of the European stack gas sampling standard, EN 15259 [4]. This is also the case when a wet chemical SRM is replaced by an instrumental Alternative Method (AM) having fi rst established equivalency with the SRM under EN 14793:2017 [5], as is commonly the case, in many European Member States, for acid gases such as SO2
However, when the SRM sample is extracted from a single point, the uncertainty related to the SRM location must then be considered. This is typically the case when the SRM is defi ned as an on-line instrumental method, such as NOx
, CO and O2 . In
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48
