作者: S. A. Pourmousavi Kani , G. H. Riahy , D. Mazhari
DOI: 10.1080/15435075.2010.548887
关键词: Wind speed 、 Markov chain 、 Term (time) 、 Algorithm 、 Hybrid algorithm 、 CPU time 、 Mathematics 、 Linear prediction 、 Statistics 、 Mean absolute percentage error 、 Nonlinear filter
摘要: A new hybrid algorithm using linear prediction and Markov chain is proposed in order to facilitate very short-term wind speed prediction. First, the transition probability matrix calculated. Then, method applied predict values. Finally, results are modified according long-term pattern by a nonlinear filter. The from compared method, persistent actual It shown that prediction-modification processes improves predictions, reducing maximum percentage error mean absolute error, while it retains simplicity low CPU time improvement uncertainty of