The estimation of daily temperature by artificial and fuzzy neural network methods


BİLGE A.

2007 International Conference on Artificial Intelligence, ICAI 2007, Las Vegas, NV, United States Of America, 25 - 28 June 2007, vol.1, pp.217-220 identifier

  • Publication Type: Conference Paper / Full Text
  • Volume: 1
  • City: Las Vegas, NV
  • Country: United States Of America
  • Page Numbers: pp.217-220
  • Keywords: Fuzzy logic, Neural networks, Weather forecasting
  • Anadolu University Affiliated: Yes

Abstract

In this study, the estimation of daily temperature of a region by fuzzy logic is discussed and some results are presented. Temperature is an atmospheric parameter which can be estimated by means of some other atmospheric parameters such as pressure, water vapor pressure, relative humidity, and wind speed. Since it is a dependent variable to some other parameters, it also can be estimated using neural network methods. With this purpose, two different methods are used in this study. First simulation is done with an artificial neural network using a feed-forward back propagation algorithm and the second one is done with a Sugeno type fuzzy neural network This application is put into practice to show that even a complex meteorological parameter, temperature, can be estimated by neural network methods and the advantage of integrated fuzzy logic to neural networks over artificial neural networks.