anticipated monthly temperatures for selected stations in isfahan province using artificial neural network multi-layer perceptron

forecasting of temperature is a very important in meteorology. air temperature prediction is of a concern in environment, industry and agriculture. temperature with precipitation are important factors in meteorology and are used in classification of climate. in this paper we want to predict average monthly temperature for chosen station of isfahan province. an artificial neural network is a powerful data modeling tool that is able to capture and represent complex input /output relationships. most of the forecasters use a multilayer perceptron network. in this paper for forecasting of monthly average temperature of selection station of isfahan province we used multilayer perceptron network. we worked on three station. isfahan, east of isfahan and kashan were our stations. we used average temperature of 7 month before for forecasting next month temperature. thus, we could forecast monthly temperature for 36 month later, in this situation we had the best correlation between data. in this paper we used matlab software 2013. in all structure of network ,there was a hidden layer with 30 neurons. all stations with hidden layer got answers. for training network used levenberg –marquardt algorithm. all network with motivation function got answer with hiperbolik tangent. in results we found this model was good for prediction.