To improve the effluent quality and trade-off such quality with energy consumption,a control strategy is proposed to optimize the set points of the control variables in the wastewater treatment process(WWTP).In this paper,the optimal problem is addressed by solving the Hamilton-Jacobi-Bellman(HJB)equation.First,a recurrent neural network is developed as a model identifier for approximating the control plant.Then an adaptive dynamic programming method based on the HJB approach is used for obtaining the policy iteration solution.This policy is used to adjust the control variables in every optimal cycle.Finally,the simulation results show that the effluent parameters are maintaining a low level and the WWTP has less energy consumption under the proposed optimal control.