The main goal of this research is to develop an ANN (Artificial Neural Network) model with simple structure and ample accuracy. In the first step, an appropriate ANN model with 24 input variables is developed using back propagation neural network by Levenberg-Marquardt algorithm to optimize the network parameters for the prediction of phosphate, total hardness and turbidity concentration in Batlagundu, Tamil Nadu. Subsequently, principal component analysis (PCA) is used to reduce the number of input variables. Finally, comparison amongst the operation of ANN-PC24 and ANN-RPC models is made. Findings indicated that the ANN-RPC models have more effective results than the ANN-PC24 model.

Keywords: Principal Component Analysis; Neural Network; Levenberg-Marquardt Algorithm; Varimax Rotation