Transient-based Optimal Decomposition of Urban Water Distribution Networks for Effective Pipeline Condition Assessment

Currently, there have been many transient-based methods (TBMs) for pipe condition assessment in urban water distribution networks (WDNs), such as the transient reflection method (TRM), the transient damping method (TDM), the system response function method (SRFM) and the inverse transient method (ITM). However, these TBMs are largely limited to be used in relatively simple pipeline systems with several pipes and simple topological configurations in view of difficulty in understanding transient signals in practical complex WDNs with thousands of pipes with numerous branches and/or loops. From this perspective, it’s necessary to conduct effective decomposition for the complex networks. This paper aims to develop a transient-based optimal decomposition method for the urban water distribution networks (WDNs), so as to improve the efficiency and accuracy of the applications of TBMs for pipe condition assessment. During this partition optimization process, several transient indicators (the number and location of transient exciters, transient influence range, and transient response amplitude) are considered as decision variables of the bi-objective optimization for the WDN. Specifically, the first objective is to minimize the difference of transient response of the WDN before and after decomposition for all considered transient exciters in the transient simulation of the system. The other objective is to maximize the transient reflection intensities of all the obtained decomposed partitions of the WDN, in order to improve the applicability and accuracy of the TBMs. To this end, the Non-dominated Sorting Genetic Algorithm (NSGA II) is adopted to achieve the globally optimal solution of the WDN decomposition. Finally, the effectiveness of the proposed method is validated and verified through the examination and application of the TBMs for pipe condition assessment such as leakage detection in WDNs.

Author Keywords: water distribution networks (WDNs), transient-based decomposition, transient-based method, bi-objective optimization, defect detection.