Sorour Esfandeh - Msc Student, Department of Environmental Sciences, Faculty of Natural Resources, University of Tehran, Iran

The hardness of water is very important for many domestic and industrial users, because since iteffects using of soap in laundry and formation of scales in boilers respectively for their applications.In this paper, a models based on wavelet neural networks are derived. This model can predict thevalue of hardness with respect to the corresponding values of chloride, fluoride, and calciumcontents of the groundwater sample. The simulation results shows that novel techniques such aswavelet Neural Networks can be used to predict the output from the data set with better accuracythan other neural networks such as MLP and Multiple Regression Model (MRM).

Artificial Neural Network Modeling, Wavelet, Water Quality Prediction, Hardness