摘要
针对风速时间序列的非线性和非平稳性的特点以及传统迭代法累计误差较大的不足,提出了基于并行相关向量机的多步预测方法。利用相空间重构进行样本重构,通过建立并行相关向量机(RVM)的短期风速预测模型对风速进行预测。仿真结果表明,与传统的多步预测模型相比,该方法预测精度更高。
According to the characteristics of the nonlinear and non-stationary of wind speed time series and the large accumulation error of the traditional iterative method, this paper puts forward a multi-step prediction method based on a parallel relevance vector machine. By using phase space reconstruction to reconstruct samples, short-term wind speed prediction is conducted by using parallel relevance vector machine （RVM）. Compared with the traditional multi-step prediction method, the simulation results show that the proposed method is more accurate than the traditional multi-step prediction model.
出处
《电网与清洁能源》
北大核心
2017年第2期112-116,共5页
Power System and Clean Energy
基金
中央高校基本科研业务费(2015MS102)
国家自然科学基金项目(51306059)~~
关键词
相关向量机
多步预测
相空间重构
relevance vector machine
multi-step prediction
phase space reconstruction
