Click on “Download PDF” for the PDF version or on the title for the HTML version.
If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.
Comparison of DRAINMOD and Artificial Neural Network for Predicting Water Table Depth and Drain Discharge in a Subsurface Drainage System
Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.orgCitation: 9th International Drainage Symposium held jointly with CIGR and CSBE/SCGAB Proceedings, 13-16 June 2010 IDS-CSBE-100087.(doi:10.13031/2013.32117)
Authors: Hamed Ebrahimian, H Ojaghlou, A Liaghat, M Parsinejad, B Nazari, H Noory
Keywords: Artificial Neural Network, DRAINMOD, Drain discharge, Subsurface Drainage, Water Table
Drainage is an effective way to control water table in the farm fields with high groundwater level in the north of Iran. This study is carried out in the Ran Behshahr field under subsurface drainage system. Artificial Neural Network and DRAINMOD model were evaluated for predicting water table depth in midpoint between two laterals designated as S3PD14 and S3PD15 and drain discharge. Depth of water table and drain discharge were measured for rainfall seasons of 2004 and 2006 years. In this study the feed-forward back propagation model of ANN was used in MATLAB Software. For evaluation of these two models, the value of absolute error (AE), standard error (SE) and R2 were calculated. For the best ANN model, these values were obtained 4.4cm, 5.8cm, and 0.57 for prediction of water table depth and 0.08 mm/day, 0.1 mm/day and 0.59 for drain discharge, respectively. For DRAINMOD model, these values were obtained 15.6 cm, 18.1 cm and 0.42 and 0.27 mm/day, 0.32mm/day and 0.71, respectively. Results indicated that the accuracy of ANN model is better than DRAINMOD model in prediction of water table depth and drain discharge in this case study.(Download PDF) (Export to EndNotes)