International Journal of Environmental Science and Development

Abstract—Ambient air temperature prediction is of a concern in environment, industry and agriculture. The increase of average temperature results in natural disasters, higher energy consumption, damage to plants and animals and global warming. Ambient air temperature predictions are notoriously complex and stochastic models are not able to learn the non-linear relationships among the considered variables. Artificial Neural Network (ANN) has potential to capture the complex relationships among many factors which contribute to prediction. The aim of this study is to develop ANN for daily and monthly ambient air temperature prediction in Kerman city located in the south east of Iran. The mean, minimum and maximum ambient air temperature during the years 1961-2004 was used as the input parameter in Feed Forward Network and Elman Network. The values of R, MSE and MAE variables in both networks showed that ANN approach is a desirable model in ambient air temperature prediction, while the results of one day ahead mean temperature and one month ahead maximum temperature are more precise using Elman network.

Index Terms—Artificial neural network, feed forward network, elman network, temperature.

Mahboubeh Afzali is with Faculty of Computer Science and Information Systems, Universiti Teknologi Malaysia, Malaysia and Islamic Azad University, Bardsir Branch, Iran (e-mail: afzali_mahboobeh@yahoo.com).
Afsaneh Afzali is with Faculty of Chemical Engineering, Universiti Teknologi Malaysia, Malaysia and Young Researchers Club, Islamic Azad University, Kerman Branch, Iran (e-mail: afzali_afsaneh@yahoo.com).
G. Zahedi is with the Chemical Engineering Department, Universiti Teknologi Malaysia, Malaysia (email: grzahedi@cheme.utm..my)

[PDF]

Cite: Mahboubeh Afzali, Afsaneh Afzali, and Gholamreza Zahedi, "The Potential of Artificial Neural Network Technique in Daily and Monthly Ambient Air Temperature Prediction," International Journal of Environmental Science and Development vol. 3, no. 1, pp. 33-38, 2012.