On the usage of artificial neural networks in chlorine control applications for water distribution networks with high quality water

Soyupak S., Kilic H., KARADİREK İ. E., MUHAMMETOĞLU H.

JOURNAL OF WATER SUPPLY RESEARCH AND TECHNOLOGY-AQUA, cilt.60, sa.1, ss.51-60, 2011 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 60 Sayı: 1
  • Basım Tarihi: 2011
  • Doi Numarası: 10.2166/aqua.2011.086
  • Dergi Adı: JOURNAL OF WATER SUPPLY RESEARCH AND TECHNOLOGY-AQUA
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.51-60
  • Anahtar Kelimeler: artificial neural networks, control, forecasting, free residual chlorine, water distribution, PARTIAL MUTUAL INFORMATION, VARIABLE SELECTION, SYSTEMS
  • Akdeniz Üniversitesi Adresli: Evet

Özet

Artificial neural network (ANN) methodology has found some recent applications as efficient control tools for satisfying free residual chlorine (FRC) levels at critical locations of water distribution systems. This particular research was started to critically investigate the potential and applicability of the ANN approach as a tool for controlling FRC levels for complex water distribution systems supplied by high quality waters with low chlorine demands. Konyaalti Water Distribution System, operated by Antalya Water and Wastewater Administration, Turkey, has been selected as a pilot. The selected system is complex in structure and supplied with raw water which has high quality and low decay rate of chlorine. The study has shown that ANN models with high predictive power and precision can be developed for such water distribution systems, and that these models can be utilized for forecasting purposes. The data for model building should be collected properly if the developed ANN models are to be utilized as control instruments for FRC levels within water distribution systems.