Wind power prediction based on PSO-Kalman

Because of its clean and green, wind power is broadly used all over the world. Wind power is random and unstable, so wind power integration will inevitably bring great impact to power system. Accurate wind power prediction can effectively alleviate the impact caused by wind power uncertainty. In ord...

Full description

Saved in:

Bibliographic Details
Published in:	Energy reports Vol. 8; pp. 958 - 968
Main Authors:	Li, Daoqing, Yu, Xiaodong, Liu, Shulin, Dong, Xia, Zang, Hongzhi, Xu, Rui
Format:	Journal Article
Language:	English
Published:	Elsevier Ltd 01.07.2022 Elsevier
Subjects:	Kalman filter algorithm Non-parametric kernel density estimation Particle swarm optimization algorithm (PSO) Wind power prediction Non-parametric kernel density estimation Particle swarm optimization algorithm (PSO) Wind power prediction Kalman filter algorithm
ISSN:	2352-4847, 2352-4847
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Because of its clean and green, wind power is broadly used all over the world. Wind power is random and unstable, so wind power integration will inevitably bring great impact to power system. Accurate wind power prediction can effectively alleviate the impact caused by wind power uncertainty. In order to increase the accuracy of wind power prediction, this article uses paper swarm optimization algorithm (PSO) to improve the traditional Kalman filter, and PSO-Kalman wind power point prediction model is established. The proposed model solves the problem of low prediction accuracy of traditional Kalman filter caused by observation noise and process noise. Finally, based on point prediction error, non-parametric kernel density estimation is used for interval prediction. By experimental simulation, by comparing the error evaluation indexes of point prediction and interval prediction, it can be found that the point prediction error of PSO-Kalman is the smallest, indicating that PSO can effectively improve the prediction accuracy of Kalman. On this basis, the interval prediction performance is also better than before. Moreover, the model proposed in this article converges fast and has better general applicability.
ISSN:	2352-4847 2352-4847
DOI:	10.1016/j.egyr.2022.02.077