The Effect Of Adding Time Factor On The Artificial Neural Network Performance In Estimating Daily Evaporation In Mountainous Region From Syrian Coast


Abstract in English

This study is aiming at building a mathematical model to estimate evaporation from Mountainous region in Syrian Coast, using an artificial neural network, based on four metrological parameters (i.e. temperature, relative humidity, wind speed and sun hours), then studying the effect of adding time variable on evaporation estimation. The mathematical model was built by the (NN-tool box), which is one of the MATLAB tools, using the daily value of the above mentioned parameters in addition to time, as the network inputs and the evaporation measured from the American pan class A as the network output . The results show that ANN4+T model which have 5 inputs (temperature, relative humidity, wind speed, sun hours, time) is the best in estimation evaporation with correlation factor of 0.8919 and Mean square error of 0.02166 for the validation set where the correlation factor in ANN4 (without time) was 0.8324 and MSE of 0.0327for the validation set.

References used

Jadeja, V, Artificial neural network estimation of Reference Evapotranspiration from pan evaporation in a semi-arid environment. National Conference on Recent Trends in Engineering & Technology, 13-14 May 2011
Kumar,P. et al, Evaporation Estimation Using Artificial Neural Networks and Adaptive Neuro-Fuzzy Inference System Techniques , Pakistan Journal of Meteorology, Vol. 8, Issue 16: Jan 2012, 81-88
SAMMEN, S. Forecasting of evaporation from Hemeren reservoir by using artificial neural networks. College of Engineering, Diyala University, Iraq. 2012

Download