A novel hybrid framework based on the ANFIS, discrete wavelet transform, and optimization algorithm for the estimation of water quality parameters

Abstract Improving the performance of machine learning (ML) algorithms is essential for accurately estimating water quality parameters (WQPs). For first time, a novel hybrid framework, namely adaptive neural fuzzy inference system–discrete wavelet transform–gradient-based optimization (ANFIS–DWT–GBO), estimation electrical conductivity (EC) and total dissolved solids (TDS), used. Before WQPs, ANFIS–DWT–GBO proven by several benchmark data sets. In addition, three algorithms, including ANFIS, ANFIS–DWT, ANFIS–GBO, are used to demonstrate strength framework. The principal component analysis (PCA) method determines best input combination in EC TDS estimation. consequences show that produces very successful competitive results sets modeling WQPs compared other algorithms. This result due simultaneous use DWT algorithm proposed can process WQP before applying it GBO utilized optimize hyperfine ANFIS. highest accuracy Mollasani Gotvand stations, respectively. correlation coefficient (R) value station 0.99, 0.98.