Data-Driven Optimal Tracking Control for the Wastewater Treatment Plant via Generalized GDHP

In order to implement the tracking control design of dissolved oxygen and nitrate nitrogen concentrations in a nonaffine wastewater treatment system, a globalized dual heuristic dynamic programming (GDHP) algorithm is established. First, the algorithm is initialized with a positive definite function to ensure that the sequence of cost functions is monotonically nonincreasing. Then, the updating rules of the cost function, the costate function, and the tracking control law are provided in the framework of generalized value iteration. Moreover, the monotonicity and the admissibility analysis of the generalized iteration algorithm are given. The model network, the critic network, and the action network are introduced to facilitate the implementation of the algorithm. Finally, the evolving policies in the iterative process are adopted to adjust the concentrations of dissolved oxygen and nitrate nitrogen in the experimental simulation, which verifies that the proposed generalized GDHP-based optimal tracking control method is applicable online.