引用本文: | 李芬,李春阳,闫全全,赵晋斌,段善旭.基于变分贝叶斯学习的光伏功率波动特性研究[J].电力自动化设备,2017,37(8): |
| LI Fen,LI Chunyang,YAN Quanquan,ZHAO Jinbin,DUAN Shanxu.Photovoltaic output fluctuation characteristics research based on variational Bayesian learning[J].Electric Power Automation Equipment,2017,37(8): |
|
摘要: |
光伏出力波动严重影响电力系统稳定运行。对光伏出力爬坡率进行分析,建立光伏出力爬坡率的高斯混合模型,并用变分贝叶斯学习算法估计模型参数。某光伏电站大量实测数据检验表明,在进行光伏功率波动特性研究方面,在不同时间尺度和天气类型下,变分贝叶斯学习算法比单一分布及基于最大期望算法的方法具有更好的拟合效果。 |
关键词: 光伏功率波动 变分贝叶斯学习 高斯混合模型 爬坡率 |
DOI:10.16081/j.issn.1006-6047.2017.08.013 |
分类号:TM615 |
基金项目:国家自然科学基金青年项目(51307105);上海绿色能源并网工程技术研究中心(13DZ2251900);上海市经济和信息委员会专项资金资助项目(沪CXY-2016-012) |
|
Photovoltaic output fluctuation characteristics research based on variational Bayesian learning |
LI Fen1, LI Chunyang1, YAN Quanquan2, ZHAO Jinbin1, DUAN Shanxu3
|
1.College of Electrical Engineering, Shanghai University of Electric Power, Shanghai 200090, China;2.Maintenance Company of SMEPC, Shanghai 200063, China;3.State Key Laboratory of Advanced Electromagnetic Engineering and Technology, Huazhong University of Science and Technology, Wuhan 430074, China
|
Abstract: |
The fluctuation of PV(PhotoVoltaic) output influences the stable operation of power system significantly. The ramp rate of PV output is analyzed, its Gaussian mixture model is built, and the variational Bayesian learning method is applied to estimate the model parameters. The test based on the massive measured data of an PV station shows that, the variational Bayesian learning method has better fitting effect than the single distribution method and EM algorithm method in researching the characteristics of PV output fluctuation for different time scales and weather types. |
Key words: PV output fluctuation variational Bayesian learning Gaussian mixture model ramp rate |