Research on Water Quality Monitoring System and Classification Model Based on Internet of Things

Abstract

Aiming at the problems of small coverage, high power consumption, low degree of automation and low accuracy of water quality classification prediction model of traditional wireless water quality monitoring system in industrial cities, a river water quality monitoring system based on Internet of things and neural network was proposed. Using low-power wide area network Lo technology to build low-power, long-distance, wide coverage of water quality monitoring information transmission channel, collect PH value, dissolved oxygen, potassium permanganate index, ammonia nitrogen and other water quality information, through the Internet of things to transmit the data to the cloud monitoring platform to achieve remote access to environmental information. At the same time, quantum genetic algorithm was used to optimize the initial weight and threshold, build G-BPNN water quality classification model, and improve the model. The accuracy and real-time performance of the system for water quality information collection and the effectiveness of the water quality classification model were verified by experiments. Compared with the traditional BP neural network model, the water quality classification prediction MSE was reduced from 0.793 to 0.069, which showed that the system constructed in this study can play the role of factory sewage discharge monitoring.