Evolutionary Multi-Objective Optimal Control of Combined Sewer Overflows

Abstract

This paper presents a novel multi-objective evolutionary optimization approach for the active control of intermittent unsatisfactory discharges from combined sewer systems. The procedure proposed considers the unsteady flows and water quality in the sewers together with the wastewater treatment costs. The distinction between the portion of wastewater that receives full secondary treatment and the overall capacity of the wastewater treatment works (including storm overflow tanks) is addressed. Temporal and spatial variations in the concentrations of the primary contaminants are incorporated also. The formulation is different from previous approaches in the literature in that in addition to the wastewater treatment cost we consider at once the relative polluting effects of the various primary contaminants in wastewater. This is achieved by incorporating a measure of the overall pollution called the effluent quality index. The differences between two diametrically opposed control objectives are illustrated, i.e. the minimization of the pollution of the receiving water or, alternatively, the minimization of the wastewater treatment cost. Results are included for a realistic interceptor sewer system that show that the combination of a multi-objective genetic algorithm and a stormwater management model is effective. The genetic algorithm achieved consistently the frontier optimal control settings that, in turn, revealed the trade-offs between the wastewater treatment cost and pollution of the receiving water.