Water quality modelling and optimisation of wastewater treatment network using mixed integer programming

Abstract

Instream water quality management encompasses field monitoring and utilisation of mathematical models. These models can be coupled with optimisation techniques to determine more efficient water quality management alternatives. Among these activities, wastewater treatment plays a crucial role. In this work, a Streeter-Phelps dissolved oxygen model (DO) is implemented in a semi-hypothetical Upper Olifants River system to forecast instream dissolved oxygen profiles in response to different wastewater discharge scenarios. A mixed integer programming (MIP) numerical approach was used in the simulation and determination of the best treatment regimen to meet the instream DO standard at the minimum cost for the chosen river catchment. The Olifants River catchment modelled in this study features 9 wastewater treatment plants. Three treatment levels were evaluated for biochemical oxygen demand (BOD) and the impact was evaluated at specific measuring points (checkpoints) within the river system. Using this model, it was demonstrated that water quality standards can be met at all monitoring points at a minimum cost by simultaneously optimising treatment levels at each treatment plant.