A PSO approach for optimum design of dynamic inversion controller in water distribution systems

Water distribution systems have become immensely complex due to ever increasing water demand and sporadic availability of water at the sources. Generally, water management issues are handled through human intervention, which naturally leads to incompetent trial and error procedures. Moreover, the non-linear system dynamics and sequential pumping operations involved in the water distribution systems make the manual control tricky. Past studies suggest that system efficiencies can be tremendously enhanced by automatic control of such systems. However, the use of control algorithms is not yet in vogue among water engineers. Thus, to demonstrate the efficacy of auto-operated water distribution network, a water supply system dataset available in the literature has been considered as a case study in the present research. This study attempts to evaluate the performance of particle swarm optimization and Ziegler–Nichols tuned dynamic inversion control of pumps in maintaining the target water levels in a series of reservoirs, wherein the controller proficiency has been gauged for different initial conditions and non-constant demand inputs. The results indicate good performance and convergence characteristics of both the controllers and thus they can be used for real time operations.