Recently electricity consumption has significantly increased due to increasing urbanization. To minimize the gap for future demands there is need for efficient load forecasting. This will economically assist greatly in terms of generating and distributing the power respectively. Building sector consumes about 40% of the electricity. To curtail the gap between demand and supply, new paradigms have to be employed that will use automated methods to dynamically forecast the buildings energy consumption. NNs are recently applied for computing electric load prediction due to their wide application. Deep recurrent neural network models provide a suitable approach and method to prediction of energy consumption. This paper review deep neural network for building energy consumption prediction that utilize machine learning algorithm including Artificial neural network, deep belief network, recurrent neural network, Elman neural network, deep recurrent neural network, convolutional neural network and Nonlinear autoregressive network. the Review explore existing research gaps and research directions for future work are highlighted. Finally, we suggest a framework for future work to enhance the prediction performance and reliability of electricity use forecast using a NARX and LSTM deep recurrent neural network model.
Key words: Deep Learning, Artificial intelligence Neural network, Recurrent neural network, Nonlinear autoregressive network, Deep recurrent neural network
|
