Simultaneous Multicomponent Voltammetric Determination throngh a Method Based on Data Mining in Chemometrics

A novel method named WPTRBFN is presented in this paper. This method is based on radial basis function neural network (RBFN) with direct orthogonal signal correction (DOSC) and wavelet packet transform (WPT) as a pre-processing tool for the simultaneous differential pulse stripping voltammetric determination of Pb (II), Ni (II) and Cd (II). DOSC was applied to remove structured noise that is unrelated to the concentration variables. Wavelet packet representations of signals provide a local time-frequency description, thus in the wavelet packet domain, the quality of noise removal can be improved. Radial basis function network was applied for overcoming the convergence problem met in back propagation training and for facilitating nonlinear calculation. In this case, through optimization, the number of DOSC components, tolerance factor, wavelet function, decomposition level, the number of hidden nodes and the width (σ) of RBFN for the DOSCWPTRBFN method were selected as 1, 0.001, Daubechies 4, 3, 8 and 0.7 respectively. The relative standard error of prediction (RSEP) for all components with DOSCWPTRBFN, WPTRBFN, and RBFN were 4.40%, 5.87%, and 6.89% respectively. Experimental results showed the DOSCWPTRBFN method to be successful and better than others.