Artificial intelligence for the modeling of water pipes deterioration mechanisms

被引:41
作者
Dawood, Thikra [1 ]
Elwakil, Emad [1 ]
Mayol Novoa, Hector [2 ]
Garate Delgado, Jose Fernando [2 ]
机构
[1] Purdue Univ, Sch Construct Management, 401 N Grant St, W Lafayette, IN 47907 USA
[2] Natl Univ St Augustin Arequipa, Sch Civil Engn, Arequipa, Peru
关键词
Artificial intelligence; Machine learning; Pipe failure; Condition assessment; Water Main deterioration; Infrastructure; State-of-the-art review; NEURAL-NETWORK; LEAK DETECTION; FAILURE RATE; PREDICTION; LOCALIZATION; REPLACEMENT; RELIABILITY; PERFORMANCE; INFERENCE; SYSTEM;
D O I
10.1016/j.autcon.2020.103398
中图分类号
TU [建筑科学];
学科分类号
0813 ;
摘要
Water pipes deterioration modeling has been a prevalent research topic in the last two decades due to high water break incidents and contamination rates. Failure processes are de facto very intricate to be diagnosed since there is a time lag between the failure incidence and consequences. Artificial intelligence (A.I.) techniques have gained much momentum during the last two decades, specifically for the deterioration modeling and assessment of water distribution networks. However, a comprehensive critical review on water infrastructure modeling via artificial intelligence and machine learning techniques is missing in the literature. This paper aims to bridge the gap in the body of knowledge and address the aforementioned limitations. The intellectual contributions of this paper are twofold. First, a comprehensive literature review method is presented through sequential steps that systematize and synthesize the literature in a scientific way. The state-ofthe-art of AI-based deterioration modeling for urban water systems is revealed along with models' methodologies, contributions, drawbacks, comparisons, and critiques. Second, future research directions and challenges are recommended to assist the construction automation research community in setting a vibrant agenda for the upcoming years.
引用
收藏
页数:12
相关论文
共 90 条
  • [1] Prediction of Water Pipe Asset Life Using Neural Networks
    Achim, D.
    Ghotb, F.
    McManus, K. J.
    [J]. JOURNAL OF INFRASTRUCTURE SYSTEMS, 2007, 13 (01) : 26 - 30
  • [2] Condition rating model for underground infrastructure sustainable water mains
    Al-Barqawi, H
    Zayed, T
    [J]. JOURNAL OF PERFORMANCE OF CONSTRUCTED FACILITIES, 2006, 20 (02) : 126 - 135
  • [3] Infrastructure Management: Integrated AHP/ANN Model to Evaluate Municipal Water Mains' Performance
    Al-Barqawi, Hassan
    Zayed, Tarek
    [J]. JOURNAL OF INFRASTRUCTURE SYSTEMS, 2008, 14 (04) : 305 - 318
  • [4] Amaitik N.M., 2008, Pipelines 2008: Pipeline asset management: Maximizing performance of our pipeline infrastructure, P1
  • [5] [Anonymous], 2001, Urban water
  • [6] [Anonymous], 2003, DETERIORATION AND INSPECTION OF WATER DISTRIBUTION SYSTEM
  • [7] ASCE (American Society of Civil Engineers), 2017, Infrastructure report card: A comprehensive assessment of America's infrastructure
  • [8] Forecasting watermain failure using artificial neural network modelling
    Asnaashari, Ahmad
    McBean, Edward A.
    Gharabaghi, Bahram
    Tutt, Donald
    [J]. CANADIAN WATER RESOURCES JOURNAL, 2013, 38 (01) : 24 - 33
  • [9] Estimation of Failure Rate in Water Distribution Network Using Fuzzy Clustering and LS-SVM Methods
    Aydogdu, Mahmut
    Firat, Mahmut
    [J]. WATER RESOURCES MANAGEMENT, 2015, 29 (05) : 1575 - 1590
  • [10] Bubtiena A. M., 2011, 2011 Proceedings of IEEE 7th International Colloquium on Signal Processing & its Applications (CSPA 2011), P50, DOI 10.1109/CSPA.2011.5759841