Real-Time Monitoring of Chlorate Concentrations, Integrating Electrochemistry Modeling, and Kalman Filtering

To monitor harmful chlorate stemming from electrochlorination in real-time, a soft sensor based on sensor data fusion (SDF) is proposed. It is based on a model of the electrochemistry that considers both electrode kinetics and mass transport due to bubble formation. An unscented Kalman filter updates the model, using a reference electrode, temperature probes, a ultraviolet (UV)-a absorption sensor and a pH probe. By considering bubble formation, the mean squared error (mse) of the hypochlorite current density estimate was reduced by 49%. Extending the Tafel model with a chloride-dependent parameter reduced the mse of the chlorate current density estimate by 79%. The results prove that it is possible to accurately monitor the chlorate concentration using SDF. The utilization of reference electrode data and an accurate, first-principles model make this monitoring system a more practical implementation than previous efforts, which relied on frequent calibration using data from sample analysis.