Prediction of Heating Energy Consumption in Houses via Deep Learning Neural Network
محورهای موضوعی : Analytical and Numerical Methods in Mechanical Design
Newsha Valadbeygi
1
(Faculty of Mechanical Engineering, Karaj Branch, Islamic Azad University, Karaj, Iran)
Ali Shahrjerdi
2
(Faculty of Mechanical Engineering, Karaj Branch, Islamic Azad University, Karaj, Iran)
کلید واژه: Neural network, heat transfer, energy consumption, deep learning,
چکیده مقاله :
This paper presents a novel model for prediction of energy consumption and heat transfer in houses on the basis of neural network by the use of experimental dataset of some cities of Iran for the learning process. To this end, a deep learning neural network (DNN) is designed by means of real set of data as input. In order to evaluate the proposed network, the predicted results are compared with the results obtained from the practical schemes. The comparison approved the effectiveness and feasibility of the suggested network in prediction of energy consumption and heat transfer with a low error for regression.