Ultra-short-term Wind Power Forecasting Based on CEEMD and Chaos Theory

Authors

Lijie Wang, 1. School of Automation, Beijing Information Science and Technology University, Beijing 100192 China; ;
Zhang Li, 1. School of Automation, Beijing Information Science and Technology University, Beijing 100192 China; ;
Zhang Yan, 2. State Grid Beijing Electric Power Research Institute, Beijing 100045, China;

Abstract

Abstract: This paper studies the ultra-short-term prediction of wind power generating capacity by means of CEEMD and chaos theory. Wind power time series is decomposed by CEEMD to decrease the non-stationary of time series. CEEMD can overcome the modal aliasing problem of EMD. The phase space reconstruction method is used to extract characteristics of each sequence, which provides the basis for the selection of input dimension when building a model. The least squares support vector machine models are built for each sequence and the prediction are made separately. The predicted results are added to get the final prediction. Simulation is performed to the real data from a wind farm of Inner Mongolia. The results show that the proposed method is effective and the mean absolute error decreases by 3.8% compared with the conventional EMD and neural network model.

Recommended Citation

Wang, Lijie; Li, Zhang; and Yan, Zhang (2019) "Ultra-short-term Wind Power Forecasting Based on CEEMD and Chaos Theory," Journal of System Simulation: Vol. 30: Iss. 4, Article 43.
DOI: 10.16182/j.issn1004731x.joss.201804043
Available at: https://dc-china-simulation.researchcommons.org/journal/vol30/iss4/43

First Page

1560

Revised Date

2016-09-09

DOI Link

https://doi.org/10.16182/j.issn1004731x.joss.201804043

Last Page

1565

CLC

TP391.9

Recommended Citation

Wang Lijie, Zhang Li, Zhang Yan. Ultra-short-term Wind Power Forecasting Based on CEEMD and Chaos Theory[J]. Journal of System Simulation, 2018, 30(4): 1560-1565.

DOI

10.16182/j.issn1004731x.joss.201804043