Modeling of discharge capacity of H-weir using experiments, bio-inspired optimization and data preprocess based on SVM

Kartal V., Emin Emiroglu M., KATİPOĞLU O. M.

International Journal of Environmental Science and Technology, vol.21, no.11, pp.7647-7666, 2024 (SCI-Expanded) identifier

  • Publication Type: Article / Article
  • Volume: 21 Issue: 11
  • Publication Date: 2024
  • Doi Number: 10.1007/s13762-024-05494-y
  • Journal Name: International Journal of Environmental Science and Technology
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Agricultural & Environmental Science Database, Aqualine, Aquatic Science & Fisheries Abstracts (ASFA), Biotechnology Research Abstracts, CAB Abstracts, Compendex, Environment Index, Geobase, INSPEC, Pollution Abstracts, Veterinary Science Database
  • Page Numbers: pp.7647-7666
  • Keywords: Data preprocessing, Discharge capacity, H-weir, Open channel flow, Optimization
  • Erzincan Binali Yildirim University Affiliated: Yes

Abstract

H-weir is a combination of a rectangular gate and weir. Side weirs are applied to discharge flow and regulate the flow height. The present study presents the modeling of the H-side weirs using experiments and optimization based on SVM in a straight channel under free and subcritical flow conditions. The geometrical and flow parameters affecting the discharge capacity were determined based on dimensional analysis. The findings show that the weir flow is not in contact with the flow below the gate with increased vertical distance between the weir crest and gate top (d) and the upstream Froude number (F1). It was found that the discharge capacity of the H-side weir is 1.3–4.0 times higher than conventional side weirs and 1.6–9.0 times higher than side gates under the same conditions. Moreover, a discharge equation is derived from a/y1, h/b, b/B, F1 and d/a with an absolute percentage error of about 5.84%. The most successful model for predicting the discharge of the H-side weir was VMD-LSSVM. In addition, it was concluded that the established LSSVM, PSO-LSSVM and PSO-VMD-LSSVM models produced very satisfactory predictions.