Towards a global multi objective optimization of wastewater treatment plant based on modeling and genetic algorithms

The optimization of the Benchmark Simulation Model 1 (BSM1) through a multi objective genetic algorithm (MOGA) is studied in this paper. First, the optimization of the set points of the two Proportional Integral (PI) controllers proposed in BSM1 is performed. Then, a new controller layout composed of three PI controllers is proposed and the set points are also optimized. Among all performance indexes proposed in BSM1, only the effluent quality and the energy consumption for pumping and aeration were taken into account in both optimization problems. Since these two objectives are conflicting, the use of the MOGA allows in both cases a direct visualization of the possible trade-offs through a Pareto curve. These two case studies showed the feasibility of such optimizations even when dealing with computing intensive model like the full scale waste water treatment plant (WWTP) model.