电网技术2017，Vol.41Issue(2)：536-542,7.DOI:10.13335/j.1000-3673.pst.2016.0831
基于改进AdaBoost.RT和KELM的风功率预测方法研究
Research on Wind Power Forecasting Method Based on Improved AdaBoost.RT and KELM Algorithm
胡梦月 1胡志坚 1仉梦林 1傅晨宇1
作者信息
- 1. 武汉大学 电气工程学院，湖北省 武汉市 430072
- 折叠
摘要
Abstract
ABSTRACT:In order to improve forecastingaccuracyand generalization ability, a wind power forecastingapproachbased on improved AdaBoost.RT algorithm isputforward in this article,toenhance performance of weak learning algorithm effectively. At first, a kernel extreme learning machine (KELM) model optimizedwithimprovedbat-inspired algorithmwith better search ability and performance of jumping out of local optima by introducing local search and Lévy flight was established. On thisbasis, several KELM models (called base learners) were built according to AdaBoost.RT algorithm.Base learners’ weights and weight of every sample in training subset were modified to achieve optimal model. Finally, forecasting results were obtained by integrating predictionresults of base learners.Simulation experimentfor different time scalesusing data of different months was conducted. Compared with BP neuralnetwork, support vector machine,and extreme learning machine prediction models, the prediction results prove excellent accuracy and generalization performanceof the proposed approach.关键词
风功率预测/基学习器/AdaBoost.RT/核极限学习机/蝙蝠算法Key words
wind power prediction/base learner/AdaBoost.RT/kernel extreme learning machine/bat-inspired algorithm分类
信息技术与安全科学引用本文复制引用
胡梦月,胡志坚,仉梦林,傅晨宇..基于改进AdaBoost.RT和KELM的风功率预测方法研究[J].电网技术,2017,41(2):536-542,7.
