Water Quality Prediction using Artificial Intelligence and Machine learning Algorithms

Abstract

Artificial intelligence approaches may significantly lower the cost of water supply and sanitation systems while also ensuring compliance with drinking water and wastewater treatment standards. As a result, there has been a lot of study on modeling and predicting water quality in order to reduce water pollution. The suggested system's innovation is offered in order to establish an effective monitoring system for drinking water in order to maintain a sustainable and environmentally friendly green environment. To estimate the water quality index, the adaptive neuro-fuzzy inference system (ANFIS) method was created in this study (WQI). Water quality was classified using a feed-forward neural network (FFNN) and K-nearest neighbors. Although the dataset contains eight important parameters, only seven were found to have significant values. These statistical factors were used to create the suggested approach. The ANFIS model outperformed the others in terms of predicting WQI values, according to the data. Despite this, the FFNN algorithm has the greatest accuracy (100%) in classifying water quality (WQC). In addition, the ANFIS model correctly predicted WQI, whereas the FFNN model was more robust in identifying the WQC. Furthermore, the ANFIS model performed well in testing, with a regression coefficient of 96.17 percent for predicting WQI, while the FFNN model had the best accuracy (100 percent) for WQC. Water treatment and management might be aided by this suggested technology, which employs superior artificial intelligence.