Research on Short-Term Wind Power Forecasting Based on RMADE-SEN-IRFR

DOI

10.2991/iceeecs-16.2016.158How to use a DOI?

Keywords

Wind Power Forecasting; Random Forest Regression; M5P Model Tree; Differential Evolution; Selective Ensemble

Abstract

To solve the problem of slowly convergence rate,weakly generalization ability, complicated model structure and parameter determination method of traditional machine learning model, the RFR model is introduced to forcast the short-term wind power. The wind power prediction model of RFR has the advantages of generalized error control, fast convergence rate, few adjustment parameters, strong interpretability, but its prediction accuracy is low and the computational complexity is high. So this paper constructs an RFR model using the M5P model tree and improves the accuracy of wind power prediction and reduces the computational complexity by combining the selective ensemble method (SEN) based on improved differential evolution algorithm (RMADE). Experimental results show that the proposed method has higher prediction accuracy and better generalization performance than SVM improved by GA (GA-SVM), OS-ELM and traditional RFR.

Copyright

© 2016, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Baoyi Wang
AU  - Hui Wang
AU  - Shaomin Zhang
PY  - 2016/12
DA  - 2016/12
TI  - Research on Short-Term Wind Power Forecasting Based on RMADE-SEN-IRFR
BT  - Proceedings of the 2016 4th International Conference on Electrical & Electronics Engineering and Computer Science (ICEEECS 2016)
PB  - Atlantis Press
SP  - 806
EP  - 811
SN  - 2352-538X
UR  - https://doi.org/10.2991/iceeecs-16.2016.158
DO  - 10.2991/iceeecs-16.2016.158
ID  - Wang2016/12
ER  -