Dynamic Multiobjective Optimal Control with Knowledge-Decision for Wastewater Treatment Process

摘要

Optimal control system plays an important role in the stability and safety of wastewater treatment process (WWTP). However, because of the dynamic complex mechanism of WWTP, it is challenging to decide the suitable set-points of manipulated variables for improving optimal control performance in the dynamic complex environment. Therefore, a dynamic multiobjective optimal control with knowledge-decision (DMOC-KD) is proposed in this paper. First, a dynamic multiobjective optimal control scheme is presented to adapt the dynamic complex environment of WWTP. Second, an adaptive multiobjective particle swarm optimization (AMOPSO), based on distributed knowledge, is presented to determine the suitable optimal set-points of WWTP. Third, a fuzzy neural network (FNN) control method is designed to track the obtained optimal set-points for keeping effluent equality and reducing energy consumption. Finally, this DMOC-KD is compared with other optimal control strategies on benchmark simulation model 1 (BSM1). The results show that this DMOC-KD is superior than most compared strategies.