A Stochastic Framework for Reliability and Sensitivity Analysis of Large Scale Water Distribution Networks

The hydraulic reliability and sensitivity analysis of large scale water distribution systems in presence of uncertainty is considered in this work. The assessment of the network reliability and sensitivity is performed by an efficient Markov chain Monte Carlo method, namely Subset simulation. Prescribed nodal heads of storage tanks, nodal demands and pipe roughness coefficients are modeled as uncertain parameters and described in a probabilistic manner. Failure is assumed to occur when the minimum nodal head in the network is lower than a minimum allowable value. The efficiency of the proposed method is demonstrated with the analysis of a real water distribution network consisting of a large number of nodes and pipes (of the order of thousands). The corresponding reliability problem represents a high dimensional problem. The approach gives an important insight into the performance, reliability and sensitivity of a class of complex utility networks.