Advanced Evaluation Methodology for Water Quality Assessment Using Artificial Neural Network Approach

Abstract

The increasing rate of water pollution and consequent increase of waterborne diseases are compelling evidence of danger to public health and all living organisms. Preservation of flora and fauna by controlling various unexpected pollution activities has become a great challenge. This paper presents an artificial neural network (ANN)-based method for calculating the water quality index (WQI) to estimate water pollution. The WQI is a single indicator representing an overall summary of various water test results. However, selection of the weight values of the water quality parameters for WQI calculation is a tedious task. Therefore, the ANN approach is found to be useful in this study for calculating the weight values and the WQI in an efficient manner. This work is novel because we propose a methodology that uses a mathematical function to calculate the weight values of the parameters regardless of missing values, which were randomly decided in previous work. The results of the proposed model show increased accuracy over traditional methods. The accuracy of the calculated WQI also increased to 98.3%. Additionally, we also designed a web interface and mobile app to supply contamination status alerts to the concerned authorities.