ARTIFICIAL NEURAL NETWORK MODELLING FOR ESTIMATION OF CONCENTRATION OF NI (II) AND CR (VI) PRESENT IN AQUEOUS SOLUTION

Analysis of aqueous solution for determination of metal components is an important task for many professionals including chemical engineers, metallurgists, biologists, geologists etc. and can be done by using sophisticated analytical instruments. It is time consuming and expensive. In present work Artificial neural network has been applied to estimate concentration of Ni (II) and Cr (VI) simultaneously present in aqueous solution with its physical properties such as optical density and pH. Experimental observations for aqueous solutions in the concentration range of 0.247 to 49.38 mg/10ml and 0.353 to 17.67mg/10ml for Ni (II) & Cr (VI) metal ions respectively, have been used in developing ANN models NCO, NCP and NCPO. These are compared for their accuracy of predictions based on the RMSE for training and test data sets. The results are indicative that the ANN model NCPO has high accuracy of prediction for both the data sets .The % relative error for maximum data points predicted is between 5 to 40 using NCPO which is acceptable. The novel feature of this work is estimation of concentration of two heavy metal ions present in the aqueous solution with its physical properties in a single model using ANN.