Weather Monitoring and Air Quality Prediction using Machine Learning

Population growth causes a rise in human activity, which in turn increases pollution levels. Both human actions and rapid economic industrialization are likely to result in increasing air pollution. This became one of the strong reasons for deteriorating health index in India. The government must take the important steps towards predicting air quality and real-time weather monitoring to ensure good air quality index in country. Today the industries are focusing towards smart and efficient technologies like $\text{IoT}$ and Machine Learning to solve these real-life problems. In this paper sensor-based system to predict air quality and weather monitoring is proposed which uses Arduino RP2040 microcontroller board as the heart of system. It uses Artificial Neural Network (ANN) machine learning model and TensorFlow lite library for predicting the air quality and displaying the real-time weather information. The model is trained using dataset taken from IQAir website for one region and tested using real-time sensory collected data from multiple sensors like BMP180, DHT22 and PM2.5 sensor. The overall efficiency of systems comes out to be 85.27%. In future, model will be trained using dataset of multiple regions over a period to further improve the accuracy of the proposed system.