Optimal sensor placement for leak location in water distribution networks using genetic algorithms
Document typeConference report
Rights accessRestricted access - publisher's policy
European Commisision's projectISENSE - Making Sense of Nonsense (EC-FP7-270428)
EFFINET - Efficient Integrated Real-time Monitoring and Control of Drinking Water Networks (EC-FP7-318556)
In this paper, a new approach for sensor placement in water distribution networks (WDN) is proposed. The sensor placement problem is formulated as an integer optimization problem. The optimization criterion consists in minimizing the number of non-isolable leaks according to the isolability criteria introduced. Because of the non-linear integer and largescale nature of the resulting optimization problem, genetic algorithms (GA) are used as solution approach. To validate the results obtained, they are compared with exhaustive search methods with higher computational cost proving that GA allow to find near-optimal solutions with less computational load. The proposed sensor placement algorithm is combined with a projection-based isolation scheme. However, the proposed methodology does not depend on the isolation method chosen by the user and it could be easily adapted to any other isolation scheme. Experiments on a real network allow to evaluate the performance of such approach.
CitationCasillas, M. [et al.]. Optimal sensor placement for leak location in water distribution networks using genetic algorithms. A: Conference on Control and Fault-Tolerant Systems. "Proceedings SYSTOL'13-2nd International Conference on Control and Fault-Tolerant Systems". Niça: 2013, p. 61-66.
DL978-1-4799-2854-5/13/$31.00 ©2013 IEE
|0082 Optimal se ... in water distribution.pdf||166,2Kb||Restricted access|