Author(s): Adell Moradi Sabzkouhi, Ali Haghighi, Amin Minaei
Linked Author(s): Ali Haghighi
Keywords: Uncertainty analysis, Water Distribution Network (WDN), fuzzy sets theory, multi-objective optimization, reliability
Abstract: Water distribution networks (WDNs) are of crucial infrastructures that are required to satisfy urban water needs with desirable quality and quantity. The hydraulic performance of each WDN is highly affected by uncertainty in the system's parameters, such as nodal demands and pipe friction factors. The input uncertainties are spread out over the system and introduce uncertainty to the network responses, nodal pressures, and pipe velocities. Those uncertainties may easily violate the design criteria and significantly influence the system reliability. Hence, an important issue in the design of a WDN is to know the extent to which the imprecision of the independent input parameters affects the system output variables and hydraulic performance. Answering this question can be very complicated when the pipe network is large and complex in its configuration. Accordingly, analyzing the uncertainty effects in pipe systems has become a challenging topic for researchers in recent years. In designing a water distribution network, the system's hydraulic performance is taken into account by some indices like the reliability, resilience, and vulnerability. This study introduces a new fuzzy index represents the fuzzy behavior of system responses and its application as an alternative objective function in the multi-objective design of WDNs. The new index needs uncertainty analysis of each design scenario. Using a self-adaptive Non-Dominated Sorting Genetic Algorithm (NSGA-II), the fuzzy design is solved and the optimum trade-off between the objective functions is derived, and the solutions are discussed. Alongside the cost-fuzzy index, some other features of the solutions are analysed and the Pareto optimal front is investigated
Year: 2017