浙江电力

2020, v.39;No.291(07) 48-54

[打印本页] [关闭]
本期目录(Current Issue) | 过刊浏览(Archive) | 高级检索(Advanced Search)

基于聚类再回归方法的光伏发电量短期预测
Short-term Forecasting of PV Power Generation Based on Clustering and Later Regression

陆爽,徐伟明,刘维亮,马青
LU Shuang,XU Weiming,LIU Weiliang,MA Qing

摘要(Abstract):

为进行光伏发电量的短期预测,提出基于天气预报的聚类再回归预测方法。首先引入Pearson相关系数分析光伏发电量与辐照强度、气温的相关性,确定辐照强度为回归模型输入以预测逐日光伏发电量。然后利用K-means聚类算法对历史气象数据进行分析并细分天气类型,提出输入为气温温差的逐时辐照强度预测模型。最后通过算例分析验证,该方法依据天气类型、气温变化量可得到逐日发电量预测值,输入量少,方法简单,预测精度较高。
This paper proposes a clustering and later regression method based on the weather forecast to shortterm forecasting of PV power generation. Firstly, the Pearson correlation coefficient is applied to analyze the correlations between PV power generation, irradiation intensity and temperature to take the irradiation intensity as the regression model input to forecast daily PV power generation; secondly, historic meteorological data are analyzed by K-means clustering algorithm, and the weather types are subdivided to propose an hourly irradiation intensity forecasting model with temperature differences as its input parameters. Finally, it is demonstrated by an example that the proposed method, characterized by small input quantity, simple procedure and high forecasting accuracy, can conclude forecasting value of daily generation according to weather types and temperature change.

关键词(KeyWords): 光伏发电;辐照强度;短期预测;回归;聚类
photovoltaic power generation;irradiation intensity;short-term forecasting;regression;clustering

Abstract:

Keywords:

基金项目(Foundation):

作者(Author): 陆爽,徐伟明,刘维亮,马青
LU Shuang,XU Weiming,LIU Weiliang,MA Qing

DOI: 10.19585/j.zjdl.202007009

参考文献(References):

扩展功能
本文信息
服务与反馈
本文关键词相关文章
本文作者相关文章
中国知网
分享