A multi-objective approach to solve an optimal control problem applied to a simplified activated sludge system model

Abstract

In this work a model representing the 24 hours operation of an activated sludge system from a wastewater treatment plant is presented. An optimal control problem is formulated in order to minimize the dissolved oxygen (control variable) and the quality index (amount of pollution). This problem is treated using direct methods, leading to a multi-objective nonlinear problem, solved using a weighted sum multi-objective approach. It was coded in the AMPL programming language and numerical experiments were carried out using the NLP solver from the NEOS Server platform.