Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
6692981 | Applied Energy | 2013 | 18 Pages |
Abstract
⺠Three new methods are proposed to predict wind speed for the wind power system. ⺠The Wavelet Packet-BFGS is better than the Wavelet Packet-ARIMA-BFGS. ⺠The Wavelet Packet-BFGS is better than the Wavelet-BFGS. ⺠They are compared to the Neuro-Fuzzy, ANFIS, RBF neural network and PM.
Keywords
WPDMSENWPANNFnnSBMMAPEBFGSBPNNDWTRBFNNRNNVector autoregression modelRpropSCMsResilient propagationautoregressive integrated moving average modelWind speed predictionsRBFCWTARIMAEMDMLPANFISDRAback propagationEvolutionary programmingMAEPSOParticle swarm optimizationDiscrete wavelet packet transformContinuous wavelet transformDiscrete wavelet transformWavelet packet decompositionEmpirical mode decompositionsignal decompositionWavelet decompositionadaptive neuro-fuzzy inference systemsFeed-forward neural networkBackpropagation neural networkRadial basis function neural networkRecurrent neural networkartificial neural networksRadial basis functionKalman filterSupport vector machinesSVMHybrid modelAutoregressive modelMean Absolute Errormean absolute percentage errorMean Square ErrorBayesian theoryMultilayer perceptronNumerical weather predictionWind speed forecastingVARGARCH
Related Topics
Physical Sciences and Engineering
Energy
Energy Engineering and Power Technology
Authors
Hui Liu, Hong-qi Tian, Di-fu Pan, Yan-fei Li,