AbstractThis research is done in Mahabad Township to estimate the amount of energy consumption for sugar beet and to compare two models, neural network with regression models. Required information is obtained randomly by completing questionnaires and face to face interviews with 95 farmers. Results show that the fertilizer, seed, and herbicide wer e the major energy consu- mers, and minor energy consumers were machi- ner y depreciation and manufacturing. Ener gy productivity, net energy gain, and energy ratio were res- pectively 1.02 kg/MJ, 189238.6 MJ and 3.9. To estimate output energy MLP, RBF, SOM, GFFN networks by changing in the number of hidden layers, training algorithm and number of neurons were used. Results showed that, MLP network have the maximum deter- mination coefficient of 98% and the minimum MSE of 0.001 with topology of 5- 5-1 and LM training. Ex- ponential model was the best model, among linear, quadratic, cubic and ex- ponential functions. Coef- ficient of determination was estimated 94.5% and it has less accuracy rather than MLP model.
Keywords: Artificial neural network, Energy efficiency, Regression model, Sugar beet .