Water Distribution System Optimization Accounting for Worst-Case Transient Loadings

This paper presents optimal transient design as a two-step optimization problem - identification and mitigation of the worst-case in water distribution system. In first step, particle swarm was used to identify set critical nodes that result loading condition. second dual-objective determine pipe sizes simultaneously minimize cost likelihood damaging events, measured by parameter named surge damage potential factor. Nondominated sorting genetic algorithms were combined with analysis produce Paretooptimal solutions search space The New York tunnel system tested case results show not always obvious cannot be assumed priori. Therefore, comprehensive systematic is required network. It also confirmed consideration phase, conjunction conventional least-cost size optimization, will help utilities yield tangible savings along improvement performance.