You are here

Home » Content

Compare the accuracy of artificial neural network and regression models to estimate flood (Case Study: Urban area of Joghatay)

Journal Name:

  • Cumhuriyet Üniversitesi Fen Fakültesi Fen Bilimleri Dergisi

Publication Year:

  • 2015

Key Words:

Author NameUniversity of Author
Abstract (2. Language): 
In this study, regression and neural network techniques is used to estimate the flood peak discharge moment in Joghatay urban basin with the use of physiographic and climatic stations surrounding the place. In order to assess the stations, Climatology and around the basin of the hydrometer, 8 stations that had at least 37 years of daily data selection were chosen and the following data were used for the input of the model: data area, the slope of the basin, average height, channel length, Gravylyus index, annual rainfall and also flood peak discharge was used as the output of the model. First the peak instantaneous flow rate was estimated by regression and then 63% of inputs were used for training neural network models, and 37% of the remaining data were used for testing the models. Finally, in order to compare the results and assess the efficiency of the method mentioned in the discharge peak moment, we used the correlation and root mean square error. The results showed that the technique of artificial neural network is superior to the regression method to estimate the flood peak discharge point. Based on these results, the problem of short period of peak discharge data in stations related to maximum flow rate can be solved by using artificial neural network, and predicting the watershed flood of the area.
Bookmark/Search this post with
FULL TEXT (PDF): 
PDF icon 5000138709-5000239126-1-pb.pdf
  • 6
305
312
  • English

REFERENCES

References: 

[1] Department of Natural Resources of Khorasan Razavi
[2] Department of water Resources of Khorasan Razavi
[3] Meteorological department in Khorasan Razavi
[4] eraghinajad, shahab and mohammad karamouz, 1384, long-term prediction of ravantab
using artificial neural network and fuzzy conclusion, researches of water resourses
magazine, no.2, pages 29-41.
[5] karamouz, mohammad and shahab ereghinejad, 1384, advanced hydrology, publication
amirkabir university, pages 293-295.
[6] rezaei, ali, 1383, regional modeing of peak flowrates of watershed places of sefidrood
dam using artificial neural network, doctoral thesis in watershed engneering and
science, natural resourses facility, university of tehran.
[7] menhaj, mohammad bagher, 1384, fundumentals of calculating neural networks, vol.1,
professor hesabi publication.
[8] dehghani, amir ahmad and reza ahmadi, 1387, estimation of dicharge of watershed
areas with statistics using artificial neural network. First international conference,
esfand 1387, university of zabol.
[9] bayazidi, 1390, statistical precedures in Humanities science, khordad 1390, sepahan
publication of tehran.
[10] kochakzadeh and bahmani, 1384, evaluation of operation of artificial neural network in
reduction of parameters needed for estimation of evaporation and Transpiration,
agricalutural science magazine, year 11, no.4.
[11] kia, mostafa, 1387, neural networks in MATLAB, Tehran, kianrayaneh publication.
TARAHOMI, ALINIA, SHAKERI
312
[12] misaghi, farhad, koroush mohammadi, 1381, simulation of rainfall in ravanab and
procedure-finding in river using artificial neural network,university of shahid chamran,
ahvaz.
[13] negaresh and others, 1391, use of artificial neural network in simulation and predicting
flood in watershed area,geoghraphic and development magazine 31, summer
1392,pages 15-28.
[14] khoshhal dastjerdi and seyyed mohammad hosseini, 1389, use of artificial neural
network in simulation of climate elements and predicting drought cycle, magazine of
geoghraphy and Environmental planning, year 21, no.3, fall 1389, pages 107-120.
[15] alborzi, 1381, introduction with artificial neural network, publication of amirkabir
university, page 137.

Thank you for copying data from http://www.arastirmax.com