A short-term electric load forecasting method using an Adaptive Multiresolution-based BiLinear Recurrent NeuralNetwork (AMBLRNN) is proposed in this paper. The AMBLRNN is based on the BLRNN that has been proven to have robust abilities in modeling and predicting time series. The learning process is further improved by using a multiresolution-based learning algorithm which employs the wavelet transform for multiresolution analysis of signal. Experiments are conducted on load data from the North-American Electric Utility (NAEU). Results show that the AMBLRNN out performs other conventional models 10%-25% in terms of MAPE (Mean Absolute Percentage Error) on forecasting accuracies.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Electric Load Forecasting Using Adaptive Multiresolution-Based Bilinear Recurrent Neural Network


    Contributors:


    Publication date :

    2008-05-01


    Size :

    249959 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English





    Electric Short-Term Load Forecasting Based on Adaptive Artificial Neural Networks

    Gan, W. / Wang, Z. / Liu, Y. et al | British Library Online Contents | 1997