Title: A quick prediction of hardness from water quality parameters by artificial neural network
Authors: Ritabrata Roy; Mrinmoy Majumder
Addresses: School of Hydro-Informatics Engineering, Department of Civil Engineering, National Institute of Technology Agartala, Agartala, Barjala, Jirania, Tripura (W), 799046, India ' School of Hydro-Informatics Engineering, Department of Civil Engineering, National Institute of Technology Agartala, Agartala, Barjala, Jirania, Tripura (W), 799046, India
Abstract: Hardness is an important water quality parameter to determine the suitability of the water for use in different purposes. The hardness of water is conventionally determined by EDTA titration method, which is fairly accurate but time-consuming. Sensor based analyser for hardness, on the other hand, is quite expensive and not easily available. Thus, the conventional methods are inconvenient for the systems, where quick estimation of hardness is essential. This study proposes a model to predict the hardness of water from a few quickly measurable water quality parameters, having high correlations with hardness. The model, developed by artificial neural network, was further validated by a different set of field data. Results indicate that the model is successful in predicting the hardness of water fairly accurately with a high correlation of 0.92, and low deviation (MAPE = 13.60, RMSE = 10.24) of the model predictions from the actual data.
Keywords: artificial neural network; hardness; water quality parameter; WQP; prediction of hardness; tripura; India.
DOI: 10.1504/IJESD.2018.094037
International Journal of Environment and Sustainable Development, 2018 Vol.17 No.2/3, pp.247 - 257
Received: 25 Mar 2017
Accepted: 13 Feb 2018
Published online: 13 Aug 2018 *