A Multi-Objective Direct Heuristic Dynamic Programming Based Tracking Control for Wastewater Treatment Process

WASTEWATER treatment process (WWTP) plays an important role in human production and life. It reduces wastewater by biochemical treatment to protect the environment. Meanwhile the highly nonlinear WWTP is difficult to model mathematically with unknown disturbances and strong coupling. In the past decades, many great efforts have been made to improve the control performance, such as on-off control [1] , PID control, feed-forward control [2] , fuzzy-based control [3] , neural network (NN)-based method [4] , and model predictive control (MPC) [5] , [6] , etc. While traditional control strategies are tuned by experience offline, fuzzy-based method is difficult to implement when the expert experience is hard to obtain, and MPC relies on a precise prediction model in advance.