Processing of signals from an ion-elective electrode array by a neural network

Neural network software is described for processing the signals of arrays of ion-selective electrodes. The performance of the software was tested in the simultaneous determination of calcium and copper(II) ions in binary mixtures of copper(II) nitrate and calcium chloride and the simultaneous determination of potassium, calcium, nitrate and chloride in mixtures of potassium and calcium chlorides and ammonium nitrate. The measurements for the Ca²⁺/Cu²⁺ determinations were done with a pH-glass electrode and calcium and copper ion-selective electrodes; results were accurate to ±8%. For the K⁺/Ca²⁺NO^?₃/Cl^? determinations, the measurements were made with the relevant ion-selective electrodes and a glass electrode; the mean relative error was ±6%, and for the worst cases the error did not exceed 20%.