High Impact Factor : 4.396 icon | Submit Manuscript Online icon | UGC Approved icon

Maximizing Accuracy of Electricity Load Forecasting With Deep Learning


Kavyashree KR , Sapthagiri college of engineering, Bangalore, India; Kamalakshi Naganna, Sapthagiri college of engineering, Bangalore, India; Ashapurna Marndi, Council of scientific and industrial research-4th paradigm institute, Bangalore, India


Deep learning, Electricity load forecasting, Neural networks, RMSE


Electricity is one such kind of energy that cannot be stored for a longer duration. The excess production than what is required can cause wastage where as the limited production can lead to scarcity. Thus, it is important to have a balanced production and consumption of electricity. Predicting the consumption that could occur in advance can help in this regard. The work involves generating more accurate predictions with the aid of deep learning. Initially, the neural network is made to learn from the historical data based on which it is expected to produce predictions for a new data set. The number of hidden layers and the hidden neurons is adjusted so as to get the minimum error. The accuracy of prediction is measured in terms of root mean square error (RMSE) and correlation coefficient. The number of hidden layers is increased gradually and accuracy of prediction is measured and compared with different network configurations.

Other Details

Paper ID: IJSRDV4I30091
Published in: Volume : 4, Issue : 3
Publication Date: 01/06/2016
Page(s): 53-55

Article Preview

Download Article