Project description
Novel AI-based air quality forecasting technology
Emissions, climate change and other effects have created a need for improved forecasting, be it for weather phenomena or air quality predictions. Current forecasting technology struggles to predict air quality accurately. However, AI technology, utilised successfully in weather forecasting and similar contexts, has shown great promise for delivering improved data on air quality. Funded by the European Research Council, the AQplus4 project aims to study recent AI forecasting models to develop a novel AI technology suitable for processing large quantities of data and making efficient predictions. The project will also collaborate with stakeholders to increase their commitment and acquire the necessary equipment and training.
Objective
AQplus4 will develop the first scientifically sound operational air quality forecasting system based on innovative deep learning and IT technology. Based on the successful development of AI air quality forecasting models in the IntelliAQ advanced grant, we will explore the combination of several deep learning models into one coherent concept, test the transferability to new air pollutant species and other world regions. Furthermore, the grant shall cover the necessary technical developments to prepare the data processing and deep learning software for operational use and we shall set-up a dialogue with two identified stakeholders (UBA Germany and NIER Korea) to discuss the data processing and forecasting requirements as well as the deployment and maintenance options. The stakeholder exchange will also include training activities including extended training of a Korean researcher. Timely and reliable air quality forecasts are important to issue health warnings and prepare mitigation measures. IntelliAQ has demonstrated higher accuracy forecasts compared to conventional chemistry transport model results. The AQplus4 system will therefore constitute an important breakthrough innovation that may later be adopted at several environmental monitoring agencies around the world.
Fields of science (EuroSciVoc)
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques.
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques.
- natural sciencescomputer and information sciencessoftware
- engineering and technologyenvironmental engineeringair pollution engineering
You need to log in or register to use this function
Programme(s)
- HORIZON.1.1 - European Research Council (ERC) Main Programme
Funding Scheme
HORIZON-ERC-POC - HORIZON ERC Proof of Concept GrantsHost institution
52428 Julich
Germany