Research on the prediction model of UV spectral water quality parameters based on INFO-LSSVM

In order to meet the requirements of precision and rapid detection of water quality parameters, UV- visible absorption spectroscopy was used for the measurement and INFO-LSSVM was proposed for predic- tive modelling. The common nitrate nitrogen (NO 3 -N) and nitrite nitrogen (NO 2 -N) in water quality testing as the solution to be measured, the UV-visible absorption spectral data filtering, spectral data integration, the establishment of INFO-LSSVM nonlinear prediction model; comparison of GA-LSSVM, PSO-LSSVM and LSSVM algorithm models, the results show that the INFO-LSSVM prediction model is effective, and pro- vides a good solution for water quality testing. LSSVM prediction model is effective and provides new re- search value for water quality detection.