Fant 872 publikasjoner. Viser side 4 av 88:
Accurate Lightweight Calibration Methods for Mobile Low-Cost Particulate Matter Sensors
<p>Monitoring air pollution is a critical step towards improving public health, particularly when it comes to identifying the primary air pollutants that can have an impact on human health. Among these pollutants, particulate matter (PM) with a diameter of up to 2.5 μm (or PM2.5) is of particular concern, making it important to continuously and accurately monitor pollution related to PM. The emergence of mobile low-cost PM sensors has made it possible to monitor PM levels continuously in a greater number of locations. However, the accuracy of mobile low-cost PM sensors is often questionable as it depends on geographical factors such as local atmospheric conditions. <p>This paper presents new calibration methods for mobile low-cost PM sensors that can correct inaccurate measurements from the sensors in real-time. Our new methods leverage Neural Architecture Search (NAS) to improve the accuracy and efficiency of calibration models for mobile low-cost PM sensors. The experimental evaluation shows that the new methods reduce accuracy error by more than 26% compared with the state-of-the-art methods. Moreover, the new methods are lightweight, taking less than 2.5 ms to correct each PM measurement on Intel Neural Compute Stick 2, an AI-accelerator for edge devices deployed in air pollution monitoring platforms.
2023
2023
2023
2023
Aerosols are an important constituent of the atmosphere both influencing the climate system and contributing to increasing pollution of the Arctic. At the same time, their adequate monitoring is a big challenge, as instruments on the ground only can sample aerosols in the lowermost atmosphere. For this reason, these measurements are complemented with observations of aerosol optical depth (AOD) which quantify the total amount of aerosols throughout the atmosphere from the attenuation of direct sunlight (and moonlight). This procedure requires extremely careful instrument calibration and removal of cloud contaminated data. In Svalbard, such measurements have been performed by several research groups with different instruments, mostly in Ny-Ålesund and in Hornsund, but also on research vessels offshore. In the framework of the SSF Strategic Grant project ReHearsol, all AOD data from the Svalbard region since 2002 have been collected and made available to the SIOS research community. They indicate that number and intensity of Arctic haze episodes occurring in late winter and spring have decreased consistently and significantly in the last 20 years, while pollution events in summer/early autumn, caused by boreal biomass burning, are on the rise, though not as consistently. Comparison between in-situ measurements at Gruvebadet Atmosphere Laboratory in Ny-Ålesund and AOD measurements indicate that most (more than 65%) of the episodes with high aerosol load are not captured by surface measurements. This finding does not change when one includes in-situ measurements at Zeppelin Observatory (475 m a.s.l.). Studying extensive high-AOD episodes such as those in summer 2019 requires a multi-tool approach including in-situ and remote-sensing measurements combined with model tools.
2023
2023