There are number of methods (i.e. engineering, regression) and computer tools (i.e. DOE-2, BLAST, HOT2000, ENERGY-10) for the modeling and forecasting of energy. Recently, a new approach artificial neural network has been widely used for load forecasting, solar energy, heating, ventilating, refrigeration, building energy analysis and so on in the field of energy as its (i.e. NN) prediction performance is better than other approaches in non-linear modeling analysis as has been found in literatures. A Neural Network (NN) also commonly referred to as an Artificial Neural Network, is an information-processing model inspired by the way the densely interconnected, parallel structure of the brain processes information. In this paper, experiments were conducted on a refrigerator to investigate the energy performance by varying the parameters (i.e. room temperature, door opening, internal cabinet temperatures, relative humidity and so on) that influence its energy consumption. Finally, experimental data were used to investigate refrigerators' energy prediction performance using NN approach. Statistical analyses in terms of fraction of variance R 2, Coefficient of variation (COV), RMS are calculated to judge the performance of NN model.
展开▼
