Alternative particle compensation techniques for online water quality monitoring using UV–Vis spectrophotometer

Particles in the water can significantly affect UV–Vis absorption measurements. There is a need for the water industry to develop a reliable technique to eliminate particle impact on on-line water quality monitoring using UV–Vis spectroscopy. This study aims to develop and use digital techniques for particle compensation: single wavelength compensation, linear regression compensation and multiplicative scatter correction method for on-line UV–Vis spectrophotometers. Water quality data were collected from three selected water sources in water treatment plants which represent different water qualities in terms of particles and organic matters. UV254 measurements were determined with these three software compensation techniques in comparison with the proprietary instrument built-in compensation algorithm using Bland-Altman analysis. Linear correction methods were found to be able to adjust the three compensation techniques to achieve acceptable compensated UV254 results, particularly for raw waters. UV254 measurements using single wavelength compensation, linear regression compensation and multiplicative scatter correction techniques with the assistant of linear correction methods were confirmed to be comparable to the instrument built-in compensation method. Our results reveal that these particle compensation techniques can make the UV254 technology reliable for online water quality monitoring in water treatment network. This paper demonstrated the advantage of using software compensation method to establish local compensation and calibration models instead of relying on the predetermined global calibrations for online water quality monitoring.