Model Predictive Control of Stochastic Wastewater Treatment Process for Smart Power, Cost-Effective Aeration

Abstract

Wastewater treatment is an essential process to ensure the good chemical and environmental status of natural water bodies. The energy consumption for wastewater treatment represents an important cost for water utilities. Meanwhile has the increasing fraction of renewable energy sources in the electricity market created the possibility of exploiting cheaper (and greener) electricity. This paper proposes model predictive control driven by stochastic differential equations and genetic optimization to prioritize aeration in periods with low electricity prices thereby reducing costs and empowering smart use of green electricity. This is without violation of legislation and equipment constraints. The method is tested with real plant data and electricity market prices to demonstrate efficiency and feasibility.