Journal of Hydraulic Engineering · 1987 · 317 citations · 16 references
EngineeringWater Resources ManagementDesign OptimizationOptimizationWater Resources EngineeringDistribution Network PlanningReliability ModelingReliability-based DesignWater Distribution SystemsWater Supply EngineeringWater Supply System AnalysisMinimal Cut SetsMinimum CutReliability OptimizationWater Distribution ModelingWater Distribution Network AnalysisWater Distribution Network ModelingWater Distribution System DesignWater Distribution System Optimization
Reliability is defined as the probability of meeting nodal demands and pressure heads under various pipe‑failure scenarios in a water distribution system. This paper presents a framework for determining the optimal (least‑cost) design of a water distribution system subject to continuity, energy conservation, nodal head bounds, and reliability constraints. The model links a steady‑state simulation that solves continuity and energy constraints and defines minimum cut sets, a minimum‑cut‑set reliability model that computes system and nodal reliability, and a generalized reduced‑gradient optimization that finds the least‑cost design under the stated constraints. Illustrative examples demonstrate the model’s application.
This paper presents the basic framework for a model that can be used to determine the optimal (least‐cost) design of a water distribution system subject to continuity, conservation of energy, nodal head bounds, and reliability constraints. Reliability is defined as the probability of satisfying nodal demands and pressure heads for various possible pipe failures (breaks) in the water distribution system. The overall model includes three that are linked: a steady‐state simulation model, a reliability model, and an optimization model. The simulation model is used to implicitly solve the continuity and energy constraints and is used in the reliability model to define minimum cut sets. The reliability model, which is based on a minimum cut‐set method, determines the values of system and nodal reliability. The optimization model is based on a generalized reduced‐gradient method. Examples are used to illustrate the model.
16