Intelligent Optimal Control for Wastewater Treatment Process with Flexible Objective Functions

In wastewater treatment process (WWTP), it is very challenging to realize the optimal control of the operation variables, since the process contains a changing number of objectives. In this paper, an intelligent optimal control with flexible objective functions (FOF-IOC)is proposed to deal with the optimal control problem for improving the operation performance. Compared with the previous optimal control methods, this proposed FOF-IOC has three main advantages. Firstly, an optimal control scheme is developed to describe the dynamics of the operation process. In this scheme, the different time-scale operation objectives are formulated to facilitate the optimal control. Secondly, a flexible objective functions strategy is designed for determining the appropriate objective functions, where the number of the objectives can be different under the distinct operation time. Thirdly, a dynamic multiobjective particle swarm optimization (DMOPSO)algorithm is investigated for optimizing the flexible objective functions, where a maximum dimension criterion is designed to cope with the nouniform optimization dimension. Finally, the proposed FOF-IOC is tested in benchmark simulation model No.1 (BSM1)to evaluate its effectiveness. The results demonstrate that the proposed FOF-IOC can achieve satisfied performance.