Integrating entropy theory and cospanning tree technique for redundancy analysis of water distribution networks

• Published in: Reliability engineering & system safety Vol. 176; pp. 102 - 112
• Main Authors: Zarghami, Seyed Ashkan, Gunawan, Indra, Schultmann, Frank
• Format: Journal Article
• Language: English
• Published: Barking Elsevier Ltd 01.08.2018; Elsevier BV
• Subjects: Case studies; Cospanning edge betweenness; Cospanning tree; Distribution management; Entropy; Hydraulics; Informational entropy; Networks; Redundancy; Reliability engineering; Theoretical mathematics; Topology; Water analysis; Water distribution; Water distribution networks; Water engineering; Water supply; Water distribution networks; Cospanning tree; Informational entropy; Redundancy; Cospanning edge betweenness
• ISSN: 0951-8320; 1879-0836
• DOI: 10.1016/j.ress.2018.04.003

•The proposed method embraces the topological characteristics of WDNs.•CEB gives more meaningful indication of the redundancy concept.•Informational entropy can be used as a measure of the overall redundancy.•Superiority of the proposed index over the conventional redundancy metrics. A large number of recent studies have addressed the redundancy evaluation of Water Distribution Networks (WDNs) from a hydraulic perspective. There already exist a few topological redundancy metrics, which address very basic structural characterizations of networks and therefore fail to realistically capture the inherent topological redundancy. To remedy this weakness, we introduce, for the first time, a two-tiered approach to evaluate the redundancy of WDNs. Tier one is supported by the cospanning tree technique which offers a novel method to measure the local redundancy of pipes. Tier two uses the results of the level one and posits the informational entropy theory as a tool to measure the global redundancy of networks. An attempt has been made to generate a new robustness index as a measure to quantify the redundancy. The proposed redundancy index can be interpreted as a measure of distance from the maximum possible redundancy. In order to demonstrate the proposed method, the paper presents two case studies, a hypothetical network and a real world WDN of an Australian town. Comparison of the presented method with conventional redundancy measures reveals the superiority of the proposed redundancy method.