Final Year Project

Project Title

Predictive and Monitoring Analysis of Outdoor Air Quality

Description

In this project, I designed and developed an air quality monitoring system using CO, CO2, PM10, PM2.5, dust, temperature & humidity sensor. These sensors were interfaced with an Arduino Uno board for data collection. LoRaWAN module was interfaced with the Arduino to send the collected data wirelessly to the cloud storage (TheThingsNetwork). Another LoRa module was interfaced with a Raspberry Pi which acted as a single channel LoRa gateway to receive the sensor data & store it on cloud. Future air quality prediction analysis was done based on the collected data. In order to do that I used python to first clean the data from anomalies and then built 5 different machine learning models such as linear regression, SVM, random forest, gradient Boost and MLP to predict the future air quality. Based on my results I was able to compare the models based on their cost functions and RMSE.