Extending the dynamic range of the determination of copper by adsorption differential pulse stripping method using a principal component artificial neural network

A principal component artificial neural network (PC-ANN) has been applied f or the analysis of the voltammogram data of Cu(II) and Cu(II)-PAN complex f or extending the dynamic range of the determination of Cu(II) (5-550 mug/1) A three layer back-propagation network (6:5:1) was used with learning rate (r= 0.09) and momentum (m = 0.9), with sigmoidal transfer function in the hidden layer and one bias node in the input layer. Overall, the application of PC-ANN enables the extension of the dynamic range of the determination of Cu(II) from its narrow linear range (5-50 mug/l) to the high dynamic ran ge (5-550 mug/l). (C) 2000 Elsevier Science B.V. All rights reserved.