Analysing the Effect of Additional Instrumentation on Prediction of COD Removal in the Hias Process

The aim of this article is to analyse the effect of additional instrumentation on prediction accuracy of COD removal in the Hias Process. The Hias Process is a novel moving bed bioreactor with enhanced biological phosphorus removal and simultaneous nitrification and denitrification (MBBR-EBPR-SND). A combination of data pre-processing methods and machine learning methods were developed to predict COD removal. Regression models including linear regression, trees, support vector machine, and neural networks models with single layer to multilayer networks were developed and tested. The prediction accuracy of the modelling methods with and without the additional electrical conductivity measurements were compared among the best performing models. The results show that the highest accuracy was achieved with support vector machine (up to R2 = 0.71). On average, the effect of the additional instrumentation was 12% increase in prediction accuracy. Further, the results can be used for monitoring and early-warning application detecting faults in sensors.