Monitoreo de la calidad del aire en un espacio concurrido aplicando internet de las cosas y redes neuronales artificiales

Abstract

Current communication technologies play an important role that can even be considered as essential in all areas related to the administration of information or data management. Air pollution is a considerable issue, so much so that attempts have been made to create solutions to help counteract the damage it causes to environmental health and that of mankind. The present project consists of monitoring the air quality in a crowded space to translate the analysis of the data obtained, and if the quality of this decreases, provide the user in what state it is in real time by applying technologies such as Internet of Things and Artificial Intelligence. Through the examination of works related to the topic, the different sensors that were used for the construction of the prototype in charge of measuring the air quality were identified, prior to the construction, the 3D modeling of the prototype and circuit diagram were made. Subsequently, the selection process of the neural network model for the classification was carried out, being Decision Tree Classification the chosen model. For the delivery of data to the user, a messaging bot was coded to transmit information on air quality through WhatsApp. The prototype was subjected to different field tests, exposing it to different environments such as: Carpentry, a home and a smoke chamber; exposing it 10 minutes in each scenario with 5 minutes of waiting before taking the data, notable differences were observed, however, the particulate matter was affected more in the smoke chamber. Therefore, the validity and reliability of the prototype was tested in these scenarios when expected variations in the data from the different test environments were observed.