River Water Pollution Pattern Prediction using a Simple Neural Network

Abstract

Rivers are an important element of its environment; river water sustains and prospers living beings in its surrounding. When river water becomes polluted, though, it becomes useless or even harmful to its ecosystem. This Paper proposes an IoT (Internet of Things) based system as a solution to counteract river pollution. The system is composed of a hardware that measures pH, temperature, and turbidity of the water – then transmitting the data via LPWAN (Low Power Wide Area Network), more specifically LoRa (Long Range. Successfully transmitted data will be used to train an ANN (Artificial Neural Network) which is used to recognize and predict patterns of river water pollution. The monitoring and prediction results will be accessible via a web app. This Paper has successfully designed and built a system that implements an ANN for recognizing patterns in river conditions, to predict potential river pollution. Early detection of river pollution can serve as vital information to act in preventing or anticipating river pollution.