Output details
13 - Electrical and Electronic Engineering, Metallurgy and Materials
University of Birmingham : A - Electronic, Electrical and computer engineering
Genetic Hybrid Predictive Controller for Optimized Dissolved-Oxygen Tracking at Lower Control Level
A new hierarchical two-level controller for optimal input of oxygen in activated sludge processes is developed and validated on a real wastewater treatment plant at Tczew, northern Poland, achieving about 30% reduction of energy cost and decreased blower switching frequency. The lower level controller (LLC) controls the aeration system, which has complicated hybrid and heavily constrained nonlinear dynamics. This paper derives a superior nonlinear hybrid predictive control algorithm for the LLC. A novel genetic optimiser is derived with dedicated operators for the problem, allowing for efficient handling the air blower switching constraints. Work was within EU project SMAC.