Interval predictor models for effluent BOD of wastewater treatment

Abstract:

Biochemical oxygen demand (BOD) is an important index for evaluating water quality, and a variable directly controlled in the wastewater treatment process. To improve the performance of wastewater treatment, it is necessary to find out an effective method for measuring BOD. This paper presents a new soft measurement which can provide guaranteed estimation of the effluent BOD. The principal component analysis is utilized to select the secondary variables for the soft sensor. In virtue of its simple topological structure and universal approximation ability, the radial basic function neural network (RBFNN) is utilized in the soft sensor modeling. Considering the bounded modeling error, linear-in-parameters set membership identification algorithm is used to obtain a description of the uncertain set of the output weights after the determination of centers of the RBFNN. The RBFNN model with uncertain output weights can predict the upper and lower bounds of the effluent BOD during the wastewater treatment. Besides, a bundle of soft sensors is constructed and the intersection of the results given by the soft sensors is used to lower the conservatism by using a single sensor. Experiment results show the satisfying performance of the proposed method.