BINARY ENCODED 2ND-DIFFERENTIAL SPECTROMETRY USING UV-VIS SPECTRAL DATA AND NEURAL NETWORKS IN THE ESTIMATION OF SPECIES TYPE AND CONCENTRATION

An approach to determining the type and concentration of a range of re presentative contaminants, chlorine, nitrate and ammonia in waste wate r, based on a three-stage scheme for processing data from ultraviolet and visible (UV-Vis) spectra, is described. In simulation in the labor atory, data for the study are derived from laboratory-based measuremen ts of such spectra from mixtures of common chemical pollutants in wate r at levels around their legal limits and from mathematical models bas ed on these measurements. Through the work, it is concluded that mathe matical procedures alone, i.e. self-learning, are not currently effect ive, while classification based on a model for absorption spectra with prior knowledge of the expected chemistry in a particular water syste m under study, is more likely to be successful.