Jump to Main Content
Model-based measurement error detection of a coagulant dosage control system
- Liu, W., Ratnaweera, H., Kvaal, K.
- International journal of environmental science and technology 2019 v.16 no.7 pp. 3135-3144
- coagulants, coagulation, computer software, drinking water treatment, models, monitoring, turbidity, wastewater treatment
- Online instruments are widely used in wastewater treatment plants and drinking water treatment plants for process monitor and control. Although maintenances of online sensors are important routine works, potential measurement errors of online sensors are challenging not only monitoring of coagulation process but also the coagulant dosage control system, what this paper is focusing on. In order to estimate and detect the potential measurement errors, this paper proposes a concept of model-based measurement error detection. Relying on the model of the outlet software senor, the differences between simulations and measurements of outlet turbidity can be used as indicator of inlet measurement errors. Based on the concept, this paper enables to quantify the measurement errors and build up a novel detection method. In addition, the paper compares the proposed detection method with a traditional method—the normal variation range. The results show that the proposed method has a better efficiency to detect the measurement error.