A project of the Seneca Foundation at the University of Murcia aims to know the quality of the air in every street, park, school …
Air pollution is one of the great challenges of the 21st century, which society must face. According to the World Health Organization, air pollution causes more than seven million deaths a year, which places it as one of the main public health problems in recent times, with a high probability of increasing its morbidity in the next years. This fact has led to the development of ICT solutions that allow the implementation of more sustainable cities, where high-resolution monitoring, data quality and impact assessment play a crucial role.
Eduardo Illueca Fernández, researcher at the Seneca Foundation in the Department of Informatics and Systems of the University of Murcia, works on the ‘Development of Artificial Intelligence Systems for the Calibration and Improvement of the Signal in Sensors for the Analysis of Present Suspended Particles in the Air (ParticleMatter) ‘. However, the project – which involves his doctoral thesis – is eminently industrial and has been carried out in the company HOP Ubiquitous, located in Ceutí and dedicated to the development of technological solutions for Smart Cities.
Given that currently one of the key aspects within the digital transformation and the challenges of society –such as the fight against climate change– is the monitoring of air quality; broadly speaking, its final objective is the implementation of a System for the Measurement of Particles in Suspension (PM) for ‘Smart Cities’ based on Artificial Intelligence, which allows high resolution monitoring, in such a way that the contamination in every street, park, school, etc. from the city without resorting to the data from the official monitoring point, which may be several kilometers away and not be realistic.
One of the key aspects within the digital transformation and the challenges of society, such as the fight against climate change, is the monitoring of the air we breathe
To do this, it will develop algorithms and solutions based on Artificial Intelligence for smart cities, within the framework of an IoT (Internet of Things) architecture. «Basically, these improvements are focused on increasing the quality of the measurement of polluting gases, PM and larger particles, as well as predicting the evolution of pollutants and anticipating pollution episodes or estimating the impact produced by measures taken by public administration in air quality, such as the implementation of Low Emission Zones, “he explains.
The project therefore proposes an improvement in calibration algorithms for PM and larger particles, using Artificial Intelligence, Machine Learning and Deep Learning techniques that allow an improvement in the analysis and calibration of the signal. More specifically, says Illueca, “what you want is to correct the effect of humidity, the particle size and the chemical composition of the particles, among others (for PM and larger particles), as well as their evaluation and standardization with respect to the reference methods, so that a high resolution monitoring can be guaranteed that allows to face the aforementioned challenges ”.
“What you want is to correct the effect of humidity, the size of the particle and the chemical composition of the particles”
Likewise, to improve the interoperability of the data and results, the principles FAIR (acronym in English for Findable, Accessible, Interoperable and Reusable) will be applied. These principles are a series of good practices that scientists use in order that their results serve others, so that they are useful for the advancement of science, and semantic technologies (those that allow computers to give meaning to data and process their relationships as well as store, manage and retrieve information based on logical relationships) for which a network or data system will be designed to define the relationships between the concepts corresponding to the Air Quality domain.
Currently, the project is reaching the halfway point of the four years for which it has funding, within the framework of the Training of Research Personnel in Universities and Public Research Bodies of the Region of Murcia, in academic fields and areas of interest to the industry of the Seneca Foundation, which implies a co-financing of the project by the company HOP Ubiquitous. “Right now, we are validating the algorithms in devices deployed in the city of Madrid, in a very ambitious project that consists of monitoring PM values in the low-emission zone, which would imply meeting a major milestone in project planning” , considers Eduardo Illlueca.
The starting hypothesis on which it works is that the application of artificial intelligence algorithms will improve the quality of the data generated by IoT devices, allowing high-resolution monitoring and the development of predictive models that will surpass the current mathematical models used so far. in the state of the art. As a secondary hypothesis, it is proposed that the use of semantic technologies and the FAIRification of the data will increase the efficiency of the algorithms and guarantee the capacity of the computer systems to access the data automatically.
This doctoral thesis is being supervised by Antonio Jesús Jara Valera, by the HOPU company, and Jesualdo Tomás Fernández Breis, by the University of Murcia. In addition, it is being developed in collaboration with highly prestigious foreign institutions, such as the Dutch TNO (Netherlands Organization for Applied Scientific Research) to ensure that optical sensors are capable of measuring smaller particles, which are the ones that can affect health the most. due to its ability to cross the alveolar barrier.