Optimal Control of a Sewer Network

The paper deals with the optimal control of a sewer network that serves a city with a population of 250,000 inhabitants. The sewer network was modeled using BSMSewer simulation environment and contains 7 tanks, corresponding to 5 wastewater collecting areas. The first 4 tanks collect domestic polluted water and rainwater from urban areas, while the tank no. 5 collects wastewater from an industrial area. Two performance indicators were considered for optimization: the volume of overflow and the discharge quality index. Four optimization methods were approached: 1. genetic algorithm-based optimization, 2. Particle Swarm Optimization (PSO) - based algorithm, 3. optimization by random generating of the solution, 4. optimization using the FminSearch method. The results were obtained by numerical simulation and were analyzed by comparison with the situation in which no control algorithm was used (all valves of the sewer network were considered fully open to the maximum value − 100% and the pumps operate to the maximum capacity).