Effective machine learning techniques for detecting inflow and infiltration water in wastewater channels

Abstract

In this degree thesis, the research team analyses different ways of monitoring water flow in wastewater channels finds an effective way of processing data using machine learning on collected data from a specific area and sets up a system for finding correlations between rain levels, water usages and wastewater flow to detect infiltrations and inflow water in wastewater channels. The study concludes that a sensor network is most suited for monitoring wastewater channels and that machine learning could be used to detect infiltration and inflow water using different types of data.