INCREASING THE ROBUSTNESS OF CLASSIFICATION ALGORITHMS TO QUANTIFY LEAKS THROUGH OPTIMIZATION
Leaks in water pipeline networks have cost billions of dollars each year. Robust leak quantification (to detect and to localize) methods are needed to minimize the lost. We quantify leaks by classifying their locations using machine learning algorithms, namely Support Vector Machine and C4.5. The algorithms are chosen due to their high performance in classification. We simulate leaks at different positions at different sizes and use the data to train the algorithms. We tune the algorithm by optimizing the algorithms' parameters in the training process. Then, we tested the algorithms' models against real observation data. We also experimented with noisy data, due to sensor inaccuracies, that often happen in real situations. Lastly, we compared the two algorithms to investigate how accurate and robust they localize leaks with noisy data. We found that C4.5 is more robust against noisy data than SVM.
7000 / 8000 Series Flow Meter With Optional Outputs. Technical report, Clark Solutions.
Typical surface roughness. Technical report, Engineering page, 2015.
Ignacio Barradas, Luis E Garza, Ruben Morales-Menendez, and Adriana Vargas-Martınez. Leaks detection in a pipeline using artificial neural networks. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 5856 LNCS:637–644, 2009.
Rachel Cardell-Oliver, Verity Scott, Thomas Chapman, John Morgan, and Angus Simpson. Designing sensor networks for leak detection in water pipeline systems. In 2015 IEEE Eighth International Conference on Intelligent Sensors, Sensor Networks and Information Processing (IEEE ISSNIP 2015), Singapore, April 2015.
Thewodros G Mamo. Virtual DMA Municipal Water Supply Pipeline Leak Detection and Classification Using Advance Pattern Recognizer Multi-Class SVM. Journal of Pattern Recognition Research, 1:25–42, 2014.
J Mashford, D De Silva, D Marney, and S Burn. An approach to leak detection in pipe networks using analysis of monitored pressure values by support vector machine. In Network and System Security, 2009. NSS ’09. Third International Conference on, pages 534–539, Oct 2009.
Water Services Association of Australia. National performance report 2012–13: Urban water utilities. Technical report, Technical report, Nation-al Water Commission and Water Services Association of Australia, 2013.
Ranko S Pudar and James A Liggett. Leaks in pipe networks. Journal of Hydraulic Engineering, 118(7):1031–1046, 1992.
Lewis A Rossman et al. EPANET 2: users manual. U.S. Environmental Protection Agency, Cincinnati, 2000.
Jose A Saez, Mikel Galar, Julian Luengo, and Francisco Herrera. Tackling the problem of classification with noisy data using multiple classifier systems: Analysis of the performance and robustness. Information Sciences, 247:1–20, 2013.
Seshan Srirangarajan, Muddaser Iqbal, Hock Beng Lim, Michael Allen, Ami Preis, and Andrew J. Whittle. Water Main Burst Event Detection and Localization. Water Distribution Systems Analysis 2010, pages 1324–1335, 2011.
Andrew J Whittle, Michael Allen, Ami Preis, and Mudasser Iqbal. Sensor networks for monitoring and control of water distribution systems. International
- There are currently no refbacks.
JUTI (Jurnal Ilmiah Teknologi Informasi) by Department of Informatics, ITS is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. JUTI is accordance with CC BY-SA.