Fault diagnosticability for an aerobic batch wastewater treatment process

Abstract

This work is concerned with the evaluation of detectability and isolability for analytical- and signal-based fault detection and isolation (FDI). Based on a case study, an aerobic sequencing batch reactor (SBR) for wastewater treatment (WWT), and its known analytical process model, it is first shown that analytical FDI approaches are not feasible through application of structural analysis. An FDI approach based on the respiration signal is then demonstrated to be viable, using information from a novel application of sensitivity theory. These findings are validated with an FDI procedure that uses extracted signal characteristics and classification to successfully diagnose deviations of some model parameters.