By Topic

Application of an adaptive neural-fuzzy system to establish a relationship among nonlinear phenomena in meteorology to obtain monthly rainfall

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

3 Author(s)
Heidari, M. ; Young Res. Club, Islamic Azad Univ., Semnan, Iran ; Nabavi, S.H. ; Shamshirband, S.

In this article we have used an adaptive neural fuzzy system to construct a smart model for obtaining monthly rainfall in four of the main cities of the province of Semnan (Semnan, Shahroud, Damghan, and Garmsar) through the use of climatic parameters of the areas studied as input. In fact, fuzzy logic has been used to establish a relationship among nonlinear meteorological phenomena for which a mathematical and formulated relationship has not been offered. To construct this model and to test it, we first studied the relationship among the observed and measured meteorological phenomena in the province of Semnan with rainfall and finally chose six meteorological parameters as input. Then, after extracting and sorting input-output data, we divided it into three groups, the first of which was used for designing the model and the other two groups were used for testing the performance of the system in the interval of the training data and also outside of the interval of training data. The results obtained show that the adaptive neural fuzzy system can be used to derive the amount of rainfall with acceptable accuracy and with a 6.5 percent error for untrained data which are in the range of trained data and with a 13 percent error for test data outside of the interval of trained data.

Published in:

Software Technology and Engineering (ICSTE), 2010 2nd International Conference on  (Volume:2 )

Date of Conference:

3-5 Oct. 2010