Analysis of the isolation valve system in water distribution networks using the segment graph

Abstract

The mechanical reliability of Water Distribution Networks (WDNs) is a relevant technical and scientific issue. During planned maintenance or unplanned interruptions, the affected area must be isolated by valves shutdown. This operation involves the alteration of the network structure, i.e., the domain of the hydraulic system, and for this reason the isolation valve system plays a central role. Some studies started to consider the presence of the isolation valve system in WDNs reliability analysis.Accordingly, this work uses the Complex Network Theory to analyse the isolation valve system performance and to assess the importance of the segments generated by valves shutdown. Differently from the classic complex network theory approach, in this work the recently proposed WDN-relevance-based betweenness centrality is applied to the segment graph to introduce information about the relevance of the different elements into the network, considering geometric and hydraulic parameters, such as length, demand, risk of disconnection, etc. The proposed strategy also suggests an improvement in the representation of the segment graph with respect to the presence of parallel edges.The strategy is presented using a small network, while it is demonstrated and discussed using a real WDN. The results indicate that the WDN-relevance-based betweenness centrality allows to effectively assess the importance of the segments generated by valves shutdowns, also providing indications to improve the isolation valve system design.

Criticality of isolation valves in water distribution networks with hydraulics and topology

“…Besides the design of new networks, analysing the operation of existing systems is also beneficial in terms of maintenance. The importance of isolation valves is investigated with topological quantities: betweenness centrality [11] and relevance-based betweenness [10]. Topological metrics have the advantage in computational time; however, they might not be able to catch the essence of some crucial scenarios due to neglecting the hydraulics [21].…”

“…In other words, it transforms the nodes into edges and the edges into nodes. Every metric is calculated for the segment graph of the networks, where segments are the nodes and the connection edges are the isolation valves [8,10]. Table 1 summarises the data of the quantities: name, notation, reference, and whether link graph transformation is needed.…”

“…Moreover, building accurate and calibrated hydraulic models are expensive, or the models are too large for comprehensive analysis. Recently complex network theory tools have been applied to discover the exposed elements in extensive networks [10,11]. The advantages of such quantities are the computational efficiency and the lack of need for a hydraulic model.…”

Potentialities of Complex Network Theory Tools for Urban Drainage Networks Analysis

“…The original aspect of the present study lays in considering the different role of nodes (e.g., inlet nodes, connection nodes, outfall nodes, etc.) embedding the information about their intrinsic relevance (Simone et al., 2022). The concept of intrinsic relevance was introduced in WDNs analysis by Giustolisi et al.…”

“…The original aspect of the present study lays in considering the different role of nodes (e.g., inlet nodes, connection nodes, outfall nodes, etc.) embedding the information about their intrinsic relevance (Simone et al, 2022). The concept of intrinsic relevance was introduced in WDNs analysis by Giustolisi et al (2020), that proposed a novel tailoring of the centrality metrics considering the information about the intrinsic relevance of each node and developed a strategy based on Relevance-embedding centralities (degree, harmonic, betweenness and edge betweenness).…”