Adaptive controller based on flexi-structure neural network for dissolved oxygen control

Abstract

This paper reports on the design of an adaptive controller based on flexi-structure neural network (FSNN) for dissolved oxygen (DO) in an activated sludge wastewater treatment plant (WWTP). The proposed FSNN incorporates a structure variable feedforward neural network (FNN), where the FNN can determine its structure on-line automatically. The structure of the FNN is adapted to cope with operating character change, while the weight parameters are updated to make the control accuracy. The special feature is that the control accuracy is maintained during adaptation and, therefore, the control performance will not be degraded when the model character changes. The performance of the proposed FSNN is illustrated with simulation. As a result of the performance evaluation of the proposed control structure, which is compared with the fuzzy and fixed structure FNN approaches; it is particularly satisfactory for the DO concentration in the WWTPs.