Version 1
: Received: 29 May 2019 / Approved: 30 May 2019 / Online: 30 May 2019 (05:33:06 CEST)
How to cite:
Samadianfard, S.; Jarhan, S.; Salwana, E.; Mosavi, A.; Shamshirband, S. Support Vector Regression Integrated with Fruit Fly Optimization Algorithm for River Flow Forecasting in Lake Urmia Basin. Preprints2019, 2019050361 (doi: 10.20944/preprints201905.0361.v1).
Samadianfard, S.; Jarhan, S.; Salwana, E.; Mosavi, A.; Shamshirband, S. Support Vector Regression Integrated with Fruit Fly Optimization Algorithm for River Flow Forecasting in Lake Urmia Basin. Preprints 2019, 2019050361 (doi: 10.20944/preprints201905.0361.v1).
Cite as:
Samadianfard, S.; Jarhan, S.; Salwana, E.; Mosavi, A.; Shamshirband, S. Support Vector Regression Integrated with Fruit Fly Optimization Algorithm for River Flow Forecasting in Lake Urmia Basin. Preprints2019, 2019050361 (doi: 10.20944/preprints201905.0361.v1).
Samadianfard, S.; Jarhan, S.; Salwana, E.; Mosavi, A.; Shamshirband, S. Support Vector Regression Integrated with Fruit Fly Optimization Algorithm for River Flow Forecasting in Lake Urmia Basin. Preprints 2019, 2019050361 (doi: 10.20944/preprints201905.0361.v1).
Abstract
Adequate knowledge about the development and operation of the components of water systems is of high importance in order to optimize them. For this reason, forecasting of future events becomes greatly significant due to making the appropriate decision. Moreover, operational river management severely depends on accurate and reliable flow forecasts. In this regard, current study inspects the accuracy of support vector regression (SVR), and SVR regulated with fruit fly optimization algorithm (FOASVR) and M5 model tree (M5), in river flow forecasting. Monthly data of river flow in two stations of the Lake Urmia Basin (Vaniar and Babarud stations on the Aji Chay and the Barandouz Rivers) were utilized in the current research. Additionally, the influence of periodicity (π) on the forecasting enactment was examined. To assess the performance of mentioned models, different statistical meters were implemented, including root mean squared error (RMSE), mean absolute error (MAE), correlation coefficient (R), and Bayesian information criterion (BIC). Results showed that the FOASVR with RMSE (4.36 and 6.33 m3/s), MAE (2.40 and 3.71 m3/s) and R (0.82 and 0.81) values had the best performances in forecasting river flows in Babarud and Vaniar stations, respectively. Also, regarding BIC parameters, Qt-1 and π were selected as parsimonious inputs for predicting river flow one month ahead. Overall findings indicated that, although both FOASVR and M5 predicted the river flows in suitable accordance with observed river flows, the performance of FOASVR was moderately better than the M5 and periodicity noticeably increased the performances of the models; consequently, FOASVR can be suggested as the accurate method for forecasting river flows.
Subject Areas
river flow forecasting; optimization; M5 model tree; support vector regression; fruit fly optimization algorithm
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
