Multisensor systems together with methods for multivariate data analysis allow to solve quite difficult analytical problems, including express identification of multicomponent mixtures.

In the present work a particular implementation of such approach based on voltammetric system of tubular electrodes and principal component analysis (PCA) is proposed. The method has been applied for investigation of voltammetric behavior of aromatic nitrocompounds in mineral water samples. Differential voltammograms of the reduction of nitrocompounds obtained at three different scan rates were used as a raw data. PCA was applied for exploratory data analysis. The identification and classification of samples was carried out using soft independent modeling of class analogy (SIMCA).

The results of classification shows, that the identification of investigated solutions with proposed method is possible for groups of mineral waters, and in some cases for individual samples.