GENERAL INFORMATION
American Journal of Innovative Research & Applied Sciences
Share us on Social Media Links:
| ISSN: 2429-5396 (e) | www.american-jiras.com | |
| Web Site Form: v 0.1.05 | JF 22 Cours, Wellington le Clairval, Lillebonne | France |
| Web Site Form: v 0.1.05 | JF 22 Cours, Wellington le Clairval, Lillebonne | France |
ResearchBib, Google Scholar, SIS database, i.f.s.i.j, Scribd, IISJ, Eurasian Scientific Journal Index (ESJI), Indianscience.in, arastirmax, Directory of Research Journals Indexing, Pak Academic Sesearch, AcademicKeays, CiteSeerX, UDL Library, CAS Abstracts, J-Gate, WorldCat, Scirus, IET Inspec Direct, and getCited
Indexed by:
| HOME || ABOUT US || ARCHIVES || AIMS AND SCOP || AUTHORS || REVIEW|| SUBMIT MANUSCRIPT || EDITORIAL BOARD || CONTACT US |
| HOME | ABOUT US | ARCHIVE | AIMS AND SCOP | AUTHORS | REVIEW | SUMIBMIT MANUSCRIPT | EDITORIAL BOARED | PUBLICATION FEE |
| | ARTICLES | Am. J. innov. res. appl. sci. Volume 4, Issue 6, Pages 247-256 (June 2017) |
| Research Article |
American Journal of innovative
Research & Applied Sciences
Research & Applied Sciences
ISSN 2429-5396 (Online)
OCLC Number: 920041286
OCLC Number: 920041286
|
| JUNE | VOLUME 4 | N° 6 | 2017 |
ABSTRACT
Background: Rainfall_Runoff modeling is necessary for hydrologic and hydraulic engineering design, integrated management of water resources, and forecasting flood. Objectives: The Objectives of this research was to build amodelof the relation between the runoff and the climatic parameters (rainfall, evaporation, temperature, relative humidity) in Alkabeer Aljanobee river in Syria by using artificial neural networks (ANN). Methods: Two type of dynamic neural networkswere used, the first type is Focused Time _Delay Neural network (FTDNN), and the second type is NARX network (HYPERLINK "https://en.wikipedia.org/wiki/Nonlinear_autoregressive_exogenous_model" \\o "Nonlinear autoregressive exogenous model" nonlinear autoregressive exogenous model). Also, Twenty four models were tested, each model had different combinations of inputs with time steps delaystake place in space [0 -3] days, in addition to historical values of runoff with time steps delaystake place in space [-1 -3]days.Results:This study reached to that the model which has input layer consists of rainfall, temperature, evaporation, relative humidity at time t=0 through t=-3, in addition to previous runoff at time t=-1 through t=-3, gives the best performance of formed neural networks. The architecture (19-25-1) ( 19 neurons in input layer and 25 neurons in hidden layer and one neuron in output layer) gives the smallest value of mean squared error (MSE)6.23*10^6, while the correlation of coefficient (R)equals 97.42 % for using data set. Conclusions: Thus, this researchhasshown the capability of using ANN in forecasting runoff depending on climatic variables and their effect on the runoff, the results have also shown that using historical runoff data improve the performance of the network very well.
Keywords : Rainfall_Runoff, DynamicNeural Networks, NARX Network, FTDNN Network.
Background: Rainfall_Runoff modeling is necessary for hydrologic and hydraulic engineering design, integrated management of water resources, and forecasting flood. Objectives: The Objectives of this research was to build amodelof the relation between the runoff and the climatic parameters (rainfall, evaporation, temperature, relative humidity) in Alkabeer Aljanobee river in Syria by using artificial neural networks (ANN). Methods: Two type of dynamic neural networkswere used, the first type is Focused Time _Delay Neural network (FTDNN), and the second type is NARX network (HYPERLINK "https://en.wikipedia.org/wiki/Nonlinear_autoregressive_exogenous_model" \\o "Nonlinear autoregressive exogenous model" nonlinear autoregressive exogenous model). Also, Twenty four models were tested, each model had different combinations of inputs with time steps delaystake place in space [0 -3] days, in addition to historical values of runoff with time steps delaystake place in space [-1 -3]days.Results:This study reached to that the model which has input layer consists of rainfall, temperature, evaporation, relative humidity at time t=0 through t=-3, in addition to previous runoff at time t=-1 through t=-3, gives the best performance of formed neural networks. The architecture (19-25-1) ( 19 neurons in input layer and 25 neurons in hidden layer and one neuron in output layer) gives the smallest value of mean squared error (MSE)6.23*10^6, while the correlation of coefficient (R)equals 97.42 % for using data set. Conclusions: Thus, this researchhasshown the capability of using ANN in forecasting runoff depending on climatic variables and their effect on the runoff, the results have also shown that using historical runoff data improve the performance of the network very well.
Keywords : Rainfall_Runoff, DynamicNeural Networks, NARX Network, FTDNN Network.
Authors Contact
*Correspondant author and authors Copyright © 2017:
| Ghatfan, Abd Alkareem Ammar 1 | Badia, Yousef Haidar 2* | and | Mais Mohammad Alean 3 |
| Ghatfan, Abd Alkareem Ammar 1 | Badia, Yousef Haidar 2* | and | Mais Mohammad Alean 3 |
Affiliation.
1.Tishreen University| Department of Water Engineering and Irrigation | Lattakia | Syria |
2.Tishreen University | Department of Structural Engineering|Lattakia| Syria |
3.Faculty of Civil Engineering | Department of Water Engineering and Irrigation | Lattakia | Syria|
1.Tishreen University| Department of Water Engineering and Irrigation | Lattakia | Syria |
2.Tishreen University | Department of Structural Engineering|Lattakia| Syria |
3.Faculty of Civil Engineering | Department of Water Engineering and Irrigation | Lattakia | Syria|
This article is made freely available as part of this journal's Open Access: ID | Ghatfan-ManuscriptRef.2-ajira170517 |
MODELING OF THE RELATION BETWEEN RUNOFF AND CLIMATIC PARAMETERS IN ALKABEER ALJANOBEE CATCHMENT IN SYRIA BY USING ARTIFICIAL NEURAL NETWORK
| Ghatfan, Abd Alkareem Ammar *1 | Badia, Yousef Haidar 2 | and | Mais Mohammad Alean 3 |. Am. J. innov. res. appl. sci. 2017; 4(6):247-256.
| PDF FULL TEXT | |Received | 17 May 2017| |Accepted | 23 May 2017| |Published 28 May 2017 |
| Ghatfan, Abd Alkareem Ammar *1 | Badia, Yousef Haidar 2 | and | Mais Mohammad Alean 3 |. Am. J. innov. res. appl. sci. 2017; 4(6):247-256.
| PDF FULL TEXT | |Received | 17 May 2017| |Accepted | 23 May 2017| |Published 28 May 2017 |