• myGriffith
    • Staff portal
    • Contact Us⌄
      • Future student enquiries 1800 677 728
      • Current student enquiries 1800 154 055
      • International enquiries +61 7 3735 6425
      • General enquiries 07 3735 7111
      • Online enquiries
      • Staff phonebook
    View Item 
    •   Home
    • Griffith Research Online
    • Conference outputs
    • View Item
    • Home
    • Griffith Research Online
    • Conference outputs
    • View Item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    Browse

  • All of Griffith Research Online
    • Communities & Collections
    • Authors
    • By Issue Date
    • Titles
  • This Collection
    • Authors
    • By Issue Date
    • Titles
  • Statistics

  • Most Popular Items
  • Statistics by Country
  • Most Popular Authors
  • Support

  • Contact us
  • FAQs
  • Admin login

  • Login
  • Water demand forecasting with AUTOFLOW© using State-Space approach

    Thumbnail
    View/Open
    NguyenPUB1816.pdf (698.5Kb)
    Author(s)
    Nguyen, KA
    Sahin, O
    Stewart, RA
    Zhang, H
    Griffith University Author(s)
    Stewart, Rodney A.
    Zhang, Hong
    Sahin, Oz
    Nguyen, Khoi A.
    Year published
    2016
    Metadata
    Show full item record
    Abstract
    The authors have recently developed an intelligent application, Autoflow©, which is a powerful tool to autonomously categorise residential water consumption data into a registry of single and combined events. This tool was developed using data collected in several cities in Australia, and when applied on standalone properties, the achieved accuracy ranged from 86% and 96% in terms of number of correctly classified events. Taking advantage of the analysis results from Autoflow©, the aim of this study is to propose a short-term water demand forecasting model that not only allows water utilities to predict the overall and peak ...
    View more >
    The authors have recently developed an intelligent application, Autoflow©, which is a powerful tool to autonomously categorise residential water consumption data into a registry of single and combined events. This tool was developed using data collected in several cities in Australia, and when applied on standalone properties, the achieved accuracy ranged from 86% and 96% in terms of number of correctly classified events. Taking advantage of the analysis results from Autoflow©, the aim of this study is to propose a short-term water demand forecasting model that not only allows water utilities to predict the overall and peak water demand of up to 24-hour ahead, but also obtain the disaggregated forecast volume of each end-use category. Based on the periodic pattern of the water consumption data, a state-space approach has been adopted whose components including trend, seasonal and external effect from temperature were all modelled as stochastic processes. In this model, Dynamic Harmonic Regression, Kalman Filter and Fixed Interval Smooth algorithms were employed for the estimation of all above-mentioned components. The model has been tested against datasets collected at different time of the year to estimate its efficiency as well as the impact of temperature on water consumption. The verification process has showed that the achieved R2 for all testings are above 0.9 when undertaking forecast of up to 24-hour ahead, whilst the obtained Mean Absolute Percentage Error (MAPE) of the disaggregated forecast volume of each category most of the time lie below 5%. Further model testing is planned to be applied on 1200 new apartment buildings in the Commonwealth Games Village located in Gold Coast Australia to validate its efficiency. Once finished, the overall Autoflow© application will have significant impact in practice that allows the water utility to have detailed plan in moderating water demand along the day, or assist in optimising the pumping activities at the least required energy. Most importantly, the ability to determine and approximate the peak demand will allow utilities to have flexible solutions to solve the over-capacity problem, especially at old metropolitans where the demand far exceeds the supply capacity of the old existing distribution network.
    View less >
    Conference Title
    Environmental Modelling and Software for Supporting a Sustainable Future, Proceedings - 8th International Congress on Environmental Modelling and Software, iEMSs 2016
    Volume
    4
    Publisher URI
    https://scholarsarchive.byu.edu/iemssconference/2016/Stream-D/67/
    Copyright Statement
    Copyright remains with the author(s) 2016. The attached file is reproduced here in accordance with the copyright policy of the publisher. For information about this conference please refer to the conference’s website or contact the author(s).
    Subject
    Pattern Recognition and Data Mining
    Publication URI
    http://hdl.handle.net/10072/124194
    Collection
    • Conference outputs

    Footer

    Disclaimer

    • Privacy policy
    • Copyright matters
    • CRICOS Provider - 00233E

    Tagline

    • Gold Coast
    • Logan
    • Brisbane - Queensland, Australia
    First Peoples of Australia
    • Aboriginal
    • Torres Strait Islander