Air quality sensor assessments
Low cost sensor (LCS) systems – manufacturer performance assessments
Rating performance
Ricardo is undertaking performance assessments of measurements of Nitrogen dioxide (NO2), and Particulate Matter (PM10 and PM2.5) by sensor systems from a range of different manufacturers. These assessments are carried out independently on a “pro bono” basis with the co-operation of the sensor suppliers. Our aim is to provide a transparent view of initial “out-of-the-box” performance, and how this improves with application of suitable QA/QC. We will report on the most interesting cases and what this means for manufacturers and end users alike, when it comes to suitable application and uncertainty measurement.
This work is intended to help suppliers’ understanding and improvement of their sensor performance, as well as assisting air quality practitioners and other sensor users in choosing the most suitable sensor systems for their future projects.
Review process
All sensor system data are carefully reviewed in both real-time and retrospectively using Ricardo’s long-established QA/QC systems for automatic air quality monitoring. Co-located reference monitors are used to assess and correct all results as required. Every sensor deployment for a specific project or network will require development of operational protocols and ongoing QA/QC review and data correction to ensure that the selected sensors continue to meet the required levels of performance, ideally with some ongoing comparison and traceability to reference measurements.
As we complete each system assessment, links to new reports will be periodically published on this page for information. Each report will provide details of the calculations for the system and the outcome of the assessment.
N.B The results from the performance assessments presented here are valid only under the specific conditions and for the duration of the test. The sensor system is evaluated on a “best-endeavours” basis within the pollution climate of the test period, and according to Ricardo’s understanding of type testing requirements. It is not a formal accreditation or endorsement of the sensor system performance.
Table 1: measurement uncertainty at the limit values for gases and particulate matter for each class of sensor system
| Class | Gases | PM |
| Class 1 | 25% | 50% |
| Class 2 | 75% | 100% |
| Class 3 | 200% | 200% |
Table 1 above shows the defined levels of performance for sensor units (European Standardisation Committee (CEN) Technical Specification TS 17660-1:2021), from Class 1 (suitable for indicative measurements) to Class 3 (suitable only for estimation).
Following our assessment of each sensor system, a Class is assigned to each pollutant measured based on the calculated measurement uncertainty after application of QA/QC. The suggested PM classifications here are expected to be confirmed in 2024.
Performance reports
 |
 |
|
| Kunak Air Pro | South Coast Science cube |
Ricardo routinely provides sensor manufacturers and users with technical evaluation, guidance and development support in terms of dataset management, quality assurance and quality control, and processing. For more information, visit our web page on Independent QA/QC of air quality sensor data.
For direct queries about Ricardo’s air quality sensor performance assessment, or to discuss how Ricardo could help your organisation to improve the quality of its sensor data, get in touch.