Research article Special Issues

Deep reinforcement learning based valve scheduling for pollution isolation in water distribution network

  • Received: 25 June 2019 Accepted: 18 September 2019 Published: 26 September 2019
  • Public water supply facilities are vulnerable to intentional intrusion. In particular, Water Distribution Network (WDN) has become one of the most important public facilities that are prone to be attacked because of its wide coverage and constant open operation. In recent years, water contamination incidents happen frequently, causing serious losses and impacts to the society. Various measures have been taken to tackle this issue. Pollution or contamination isolation by localizing the contamination via sensors and scheduling certain valves have been regarded as one of the most promising solutions. The main challenge is how to schedule water valves to effectively isolate contamination and reduce the residual concentration of contaminants in WDN. In this paper, we are motivated to propose a reinforcement learning based method for valve real time scheduling by treating the sensing data from the sensors as state, and the valve scheduling as action, thus we can learn scheduling policy from uncertain contamination events without precise characterization of contamination source. Simulation results show that our proposed algorithm can effectively isolate the contamination and reduce the risk exclosure to the customers.

    Citation: Chengyu Hu, Junyi Cai, Deze Zeng, Xuesong Yan, Wenyin Gong, Ling Wang. Deep reinforcement learning based valve scheduling for pollution isolation in water distribution network[J]. Mathematical Biosciences and Engineering, 2020, 17(1): 105-121. doi: 10.3934/mbe.2020006

    Related Papers:

  • Public water supply facilities are vulnerable to intentional intrusion. In particular, Water Distribution Network (WDN) has become one of the most important public facilities that are prone to be attacked because of its wide coverage and constant open operation. In recent years, water contamination incidents happen frequently, causing serious losses and impacts to the society. Various measures have been taken to tackle this issue. Pollution or contamination isolation by localizing the contamination via sensors and scheduling certain valves have been regarded as one of the most promising solutions. The main challenge is how to schedule water valves to effectively isolate contamination and reduce the residual concentration of contaminants in WDN. In this paper, we are motivated to propose a reinforcement learning based method for valve real time scheduling by treating the sensing data from the sensors as state, and the valve scheduling as action, thus we can learn scheduling policy from uncertain contamination events without precise characterization of contamination source. Simulation results show that our proposed algorithm can effectively isolate the contamination and reduce the risk exclosure to the customers.


    加载中


    [1] S. E. Hrudey, Safe drinking water: lessons from recent outbreaks in affluent nations, 2005.
    [2] D. J. Kroll, Securing our water supply: protecting a vulnerable resource. PennWell Books, 2006.
    [3] S. Rathi and R. Gupta, A simple sensor placement approach for regular monitoring and contamination detection in water distribution networks, KSCE J. Civ. Eng., 20 (2016), 597-608.
    [4] C. Hu, G. Ren, C. Liu, et al., A spark-based genetic algorithm for sensor placement in large scale drinking water distribution systems, Cluster Comput., 20 (2017), 1089-1099.
    [5] X. Yan, K. Yang, C. Hu, et al., Pollution source positioning in a water supply network based on expensive optimization, Desalin Water Treat, 110 (2018), 308-318.
    [6] H. Wang and K. W. Harrison, Improving efficiency of the bayesian approach to water distribution contaminant source characterization with support vector regression, J. Water Res. Pl., 140 (2012), 3-11.
    [7] X. Yan, J. Zhao, C. Hu, et al., Multimodal optimization problem in contamination source determination of water supply networks, Swarm Evol. Comput., 2017.
    [8] W. Gong, Y. Wang, Z. Cai, et al., Finding multiple roots of nonlinear equation systems via a repulsion-based adaptive differential evolution, IEEE T. Syst. Man Cy., (2018), 1-15.
    [9] A. Afshar and M. A. Marino, Multiobjective consequent management of a contaminated network under pressure-deficient conditions, J. Am. Water Works. Ass., 106 2014.
    [10] T. Ren, S. Li, X. Zhang, et al., Maximum and minimum solutions for a nonlocal p-laplacian fractional differential system from eco-economical processes, Bound Value Probl., 2017 (2017), 118.
    [11] T. Ren, X. H. Lu, Z. Z. Bao, et al., The iterative scheme and the convergence analysis of unique solution for a singular fractional differential equation from the eco-economic complex system's co-evolution process, Complexity, 2019.
    [12] A. Preis and A. Ostfeld, Multiobjective contaminant response modeling for water distribution systems security, J. Hydroinform., 10 (2008), 267-274.
    [13] A. Afshar and E. Najafi, Consequence management of chemical intrusion in water distribution networks under inexact scenarios, J. Hydroinform., 16 (2014), 178-188.
    [14] D. Silver, A. Huang, C. J. Maddison, et al., Mastering the game of go with deep neural networks and tree search, Nature, 529 (2016), 484.
    [15] B. G. Kim, Y. Zhang, M. van der Schaar, et al., Dynamic pricing and energy consumption scheduling with reinforcement learning, IEEE T. Smart Grid, 7 (2016), 2187-2198.
    [16] M. H. Moghadam and S. M. Babamir, Makespan reduction for dynamic workloads in cluster-based data grids using reinforcement-learning based scheduling, J. Comput. Sci-neth., 2017.
    [17] S. Rathi and R. Gupta, A critical review of sensor location methods for contamination detection in water distribution networks, Water Qual. Res. J. Can., 50 (2015), 95-108.
    [18] T. P. Lambrou, C. C. Anastasiou, C. G. Panayiotou, et al., A low-cost sensor network for real-time monitoring and contamination detection in drinking water distribution systems, IEEE Sens. J., 14 (2014), 2765-2772.
    [19] D. Zeng, L. Gu, L. Lian, et al., On cost-efficient sensor placement for contaminant detection in water distribution systems, IEEE T. Ind. Inform., (2016), 1-1.
    [20] L. Perelman and A. Ostfeld, Operation of remote mobile sensors for security of drinking water distribution systems, Water Res., 47 (2013), 4217-4226.
    [21] M. R. Bazargan-Lari, An evidential reasoning approach to optimal monitoring of drinking water distribution systems for detecting deliberate contamination events, J. Clean Prod., 78 (2014), 1-14.
    [22] A. Poulin, A. Mailhot, P. Grondin, et al., Optimization of operational response to contamination in water networks, in WDSA 2006, (2008), 1-15.
    [23] A. Poulin, A. Mailhot, N. Periche, et al., "Planning unidirectional flushing operations as a response to drinking water distribution system contamination," J. Water Res Pl., 136 (2010), 647-657.
    [24] M. Gavanelli, M. Nonato, A. Peano, et al., Genetic algorithms for scheduling devices operation in a water distribution system in response to contamination events, 7245 (2012), 124-135.
    [25] A. Rasekh and K. Brumbelow, Water as warning medium: Food-grade dye injection for drinking water contamination emergency response, J. Water Res. Pl., 140 (2014), 12-21.
    [26] A. Rasekh, M. E. Shafiee, E. Zechman, et al., Sociotechnical risk assessment for water distribution system contamination threats, J. Hydroinform, 16 (2014), 531-549.
    [27] A. Rasekh and K. Brumbelow, Drinking water distribution systems contamination management to reduce public health impacts and system service interruptions, Environ. Model Softw., 51 (2014), 12-25.
    [28] M. Knowles, D. Baglee and S. Wermter, Reinforcement learning for scheduling of maintenance, in Research and Development in Intelligent Systems XXVII. Springer, (2011), 409-422.
    [29] K. L. A. Yau, K. H. Kwong and C. Shen, Reinforcement learning models for scheduling in wireless networks, Front Comput. Sci-Chi., 7 (2013), 754-766.
    [30] L. A. Rossman et al., Epanet 2: users manual, 2000.
    [31] R. S. Sutton and A. G. Barto, Reinforcement learning: An introduction, in Neural Information Processing Systems, 1999.
    [32] V. Mnih, K. Kavukcuoglu, D. Silver, et al., Playing atari with deep reinforcement learning, ArXiv, abs/1312.5602, 2013.
    [33] Y. Xuesong, S. Jie, and H. Chengyu, Research on contaminant sources identification of uncertainty water demand using genetic algorithm, Cluster Comput., 20, (2017), 1007-1016.
  • Reader Comments
  • © 2020 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(3571) PDF downloads(577) Cited by(7)

Article outline

Figures and Tables

Figures(8)  /  Tables(3)

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog