A New Wind Power Forecasting Algorithm Based on Spark

DOI

10.2991/icmmita-15.2015.101How to use a DOI?

Keywords

Wind Power Prediction; Spark; Parallelization Improvement; Artificial Intelligence

Abstract

With the arrival of the era of big data, wind power forecasting data develops to the massive and multi-dimensional direction. Single computer lack of resources is difficult to solve the problem of wind power forecasting. In this paper, Spark cloud computing platform is introduced to solve the massive data problem that the traditional single machine can not deal with. The combination of BP neural network with the bacterial colony optimization algorithm can avoid the local optimum and improve the accuracy of forecasting results. In the experiment?the real data of a wind farm in Nei Mongol is used as the input data set and Spark cloud computing programming framework improve the parallel algorithm. The results show that the improved algorithm can increase the efficiency of prediction results about 10%.

Copyright

© 2015, 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  - Shaomin Zhang
AU  - Peng Chen
AU  - Baoyi Wang
PY  - 2015/11
DA  - 2015/11
TI  - A New Wind Power Forecasting Algorithm Based on Spark
BT  - Proceedings of the 2015 3rd International Conference on Machinery, Materials and Information Technology Applications
PB  - Atlantis Press
SP  - 520
EP  - 525
SN  - 2352-538X
UR  - https://doi.org/10.2991/icmmita-15.2015.101
DO  - 10.2991/icmmita-15.2015.101
ID  - Zhang2015/11
ER  -