Mid-term Summit showcasing TRANSITION-funded Clean Air Discovery & Innovation projects:
-
Measuring Exposure in Different Transport Modes | Nick Molden (Emissions Analytics Ltd)
-
Characterising Changing Travel Patterns in the COVID-19 Era | Dr Fiona Crawford (University of the West of England)
-
Progressing Real-Time Source Identification | Gordon Allison (DustScan Ltd)
-
Minimising Public Exposure at the Roadside | Dr Fabrizio Bonatesta (Oxford Brookes University)
Progressing Real-Time Source Identification | Gordon Allison (DustScan Ltd)
Enabling real-time air quality management at high spatial coverage, Dustscan Ltd will develop statistical techniques for machine learning to differentiate between construction dust and non-exhaust vehicle emissions using its new DustScan Cloud ‘low-cost’ air quality sensor, including on the HS2 Curzon Street site.
Tiny particles in the air, known as particulate matter (PM) air pollution, are harmful to human health with some particles worse than others. PM comes in different size patterns that are typical of different sources, i.e. diesel smoke is smaller than construction/brake dust. Low cost ways to measure and understand the sources of PM pollution form a new field pioneered by the University of Birmingham (UoB) that has great commercial potential. This project joins the business and consulting expertise of DustScanAQ with the academic expertise of the UoB. The measurement and source apportionment of atmospheric pollutants is crucial for the assessment of air quality and the implementation of policies for its improvement.
Up to now for particle sizing, measurements have used bulky, expensive regulatory grade instruments costing approximately £100k, which makes it difficult to put many instruments around sources. Low cost sensors (£sub-5k) provide an affordable alternative, as evidenced by recent work by Bousiotis et al. 2021 (https://doi.org/10.5194/amt-2021-11), but their capability and reliability have yet to be tested with real world problems. In this project, low cost PM sensors will be used to differentiate particle sources from two policy relevant areas that relate to low carbon transport transitions: 1. Nuisance dust from railways infrastructure construction; and 2. Non-exhaust vehicle emissions.
Furthermore, the project will develop the technology so it can perform real-time source identification in the cloud, utilizing machine learning techniques to allow for source identification in real-time. This gives the potential for real-time identification of emissions from sources, helping drive the transition to a cleaner world. The project is in collaboration with HS2, which is building the next generation infrastructure for the UK’s high speed rail service.