Şeref SAĞIROĞLU, Kenan DANIŞMAN, Mete ÇELİK

Neural predictor for temperature estimation

Günlük insan aktiviteleri tamamen meteorolojik tahminlerle ilişkilidir. Eğer tahminler iyi yapılmazsa, umulmayan yer ve zamanlarda kötü hava şartları yüzünden insanoğlu zorluklarla karşılaşabilir. Meteorolojik parametrelerin tahmini silahlı kuvvetler için de önemlidir. Bu çalışmada, sıcaklık tahmininin önemli olduğu kritik işlerde güvenli bir ortam oluşturmada yapay sinir ağları kullanımı ve karşılaştırılması yapılmıştır. Çok katlı perseptronlar (MLP) kullanılarak günlük sıcaklık tahmini için yeni bir yaklaşım sunulmuştur. Geriyayılım, hızlı-yayılım ve geliştirilmiş delta-bar-delta öğrenme algoritmaları MLP'leri eğitmek için kullanılmıştır. Bu yaklaşımlar kullanılarak bulunan tahmini sıcaklık değerleri, Kayseri Meteoroloji İstasyonu'nda kaydedilen sonuçlarla kaşılaştmldığında uyumluluk içerisindedir.

Sıcaklık tahmininde yapay sinir ağı kullanımı

Daily human activities are closely related to.the meteorological forecasts. If the forecasts are not predicted precisely human face difficulty because of the bad weather conditions at unexpected locations and times. The forecast of meteorological parameters has also important for the armed forces and military operation. This study presents a temperature forecast based on artificial neural networks to support critical tasks. Multilayered perceptrons (MLPs) for predicting the daily temperature is presented. The three learning algorithms, the backpropagation, the quickpropagation, and the extended delta-bar-delta, are used to train the MLPs. The predicted temperatures obtained by using these approaches are in very good agreement with the measured results recorded at State Meteorological Station in Kayseri, Turkey.

___

1.Krishnamurti, T.N. and Bounoua, L, An Introduction to Numerical Weather Prediction techniques, Lewis Publishers, Inc. ISBN: 08493 89100, (1996)
2.Buffa, F. and Porceddu, I. "Temperature forecast and dome seeing minimization", Astron. Astrophys.'Suppl. Ser. 126, 547-553, (1997).
3.Haltiner, G. J. and Williams, R. T., Numerical Prediction and Dynamic Meteorology, 2nd ed., John Wiley & Sons; ISBN: 0471059714, (1980)
4.Haykin, S., Neural Networks: A Comprehensive Foundation, Macmillan College Publishing Company, New York, USA, ISBN 0-02-352761-7, (1994)
5.Maren, A., Harston, C. and Pap, R., Handbook of Neural Computing Applications. London, Academic Press, ISBN 0-12-471260-6, (1990).
6.Knotanzad, A., Davis, M.H., Abaye, A. and Maratukulam,D.J., "An artificial neural network hourly temperature forecaster, with applications in load forecasting", IEEE Transactions on Power Systems, Vol 11, No.2, (May 1996)
7.Kalogirou, S.A., Panteliou, S. and Dentsoras, A., "Modelling of solar domestic water heating systems using artificial neural network", Solar Energy Vol 65, No. 6 pp. 335-342,(1999)
8.Miyano, T. and Girosi, F., "Forecasting global temperature variations by neural networks", Massachusetts Institute of Technology Artificial Intelligence Laboratory and Denter for Biological and Computational Learning Department of Brain and Cognitive Sciences, A. I. Memo No. 1447, C. B. C. L. Memo No. 101
9.Sagıroglu, Ş., Güney, K., Erler, M., and Babayigit, B., "Neural prediction of the daily temperature" 2nd. Int. Symp. on Mathematical & Computational Applications, Baku, Azerbaijan, PP.256-262, (1-3 September 1999).
10.Sönmez, İ. and Şen, Z., "Artificial neural networks approach for natural atmospheric event dynamics and application in meteorology", Proc. of 2nd Int. Symp. Intelligent Manufacturing Systems, Sakarya University, Sakarya, Turkey, 325-332, 6-7 (August 1998)
11.Yu, Z., "A temperature match based optimization method for daily load prediction considering DLC effect", IEEE Transactions on Power Systems, Vol 11, No.2, (May 1996)
12.Hippert, H.S., Pedreira, C.E. and Souza, R.C., "Combining neural networks and ARIMA models for hourly temperature forecast", IJCNN 2000 Proceedings of the IEEE-INNS-ENNS, Vol.4, 414-419, (2000)
13.Bankert, R.L., "Cloud classification of AVHRR imagery in maritime regions using a probalistic neural networks", J. App. Meteor., 33, 909-918, (1994)
14.Bankert, R.L. and Aha, D.W., "Improvement to a neural networks cloud classifier", J. App. Meteor., 35, 2036-2039, (1996).
15.Marzban, C. and Stumpf, G.J., "A neural network for tornado prediction based on doppler radar-derived attributes", J. App. Meteor., 35, 617-626, (1996)
16.Atluri, V., Hung, C. and Coleman, T.L., "An artificial jipural network for classifying predicting moisture and temperature using Levenberg-Marquardt algorithm", IEEE Southeast Con'99, Proceedings of IEEE, 10-13, (1999)
17.El Desouky, A.A. and ElKateb, M.M., "Hybrid adaptive techniques for electric-load forecast using ANN and ARIMA", IEE Proc.-Gener. Transn. Distrib., Vol. 147, No 4, (July 2000)
18.Park, D.C., El-Sharkawi, M.A., II Marks, R.J., Atlas, L.E. and Damborg, M.J., "Electric load forecasting using an artificial neural network", IEEE Transactions on Power Systems, Vol. 6, No. 2, (May 1991)
19.Rumelhart, D.E. and McClelland, J.L., Parallel Distributed Processing: Explorations in the Microstructure of Cognition, Vol. 1, MIT Press, Cambridge, Massachusetts, (1986)
20.Minai, A.A. and Williams, R.D., "Acceleration of backpropagation through learning rate and momentum adaptation", Int. Joint Conf. on Neural Networks, 1, 676-679, (Jan. 1990)
21.Neural Computing, A Technology Handbook for Professional II/PLUS and Neural Works Explorer, Pittsburgh, USA, NeuralWare, Inc. Technical Publications Group, (1996)
22.Jacobs, R.A., "Increased rate of convergence through learning rate adaptation", Neural Networks, 1, 295-307, (1988)
23.Fahlman, S.E., "An emprical study of learning speed in backpropagation networks", Technical Report CMU-CS-88-162, Carnegie Mellon University, (June 1988)