A multi-objective operation optimization method based on an implicit memory strategy is proposed for irregular and nonstationary time-varying processes to solve the problems of excessive energy consumption and severe water quality in wastewater treatment process(WWTP). First, an ensemble just-in-time learning(E-JITL) algorithm is used to build an operation optimization model, which realizes an accurate description of the operation optimization objectives. Then, a multi-objective optimization algorithm based on the implicit memory(IM-NSGAII) is designed, considering the dynamic time-varying working conditions, mining the structured knowledge buried in the historical optimization information, guiding the evolutionary search process, and combining the average distance clustering index and uniform random local search, which improves the convergence, diversity and distribution of the algorithm. Finally, an intelligent decision-making method based on the E-JITL algorithm is used to obtain the optimal setpoints of dissolved oxygen concentration and nitrate concentration for tracking control. Experiments show that the proposed this method can achieve stable operation of the WWTP, improve pollutant removal performance, and reduce operating costs.