Title: Application of artificial neural network to model the energy output of dairy farms in Iran

Authors: Paria Sefeedpari; Shahin Rafiee; Asadollah Akram

Addresses: Department of Agricultural Machinery Engineering, Faculty of Agricultural Engineering and Technology, University of Tehran, P.O. Box 4111, Karaj 31587-77871, Iran ' Department of Agricultural Machinery Engineering, Faculty of Agricultural Engineering and Technology, University of Tehran, P.O. Box 4111, Karaj 31587-77871, Iran ' Department of Agricultural Machinery Engineering, Faculty of Agricultural Engineering and Technology, University of Tehran, P.O. Box 4111, Karaj 31587-77871, Iran

Abstract: An artificial neural network (ANN) model was developed to assess the energy input-output prediction in dairy farms of Iran. Data used were culled from 50 randomly selected farms using face to face questionnaire approach. The energy input-output analysis was carried out for the parameters of ANN model. Based on performance measures, single hidden layers with 16 neurons in the hidden layer were finally selected as the best configuration for predicting energy output. In this study, we calculated total energy input and output to be 53,102 and 58,315 MJ cow⁻¹, respectively. The predicted values of the best and optimal structure of ANN model were correlated well with actual values with coefficient of determination (R²) of 0.88 and root mean square error (RMSE) of 0.015. Therefore, since the ANN model can accurately predict the derived energy output of milk production system, it could be alternated by other predicting approaches such as regression.

Keywords: dairy farms; energy output; artificial neural networks; ANNs; energy input; milk production; Iran; farming.

DOI: 10.1504/IJETP.2013.055819

International Journal of Energy Technology and Policy, 2013 Vol.9 No.1, pp.82 - 91

Received: 08 May 2012
Accepted: 13 May 2013
Published online: 21 Jun 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article