期刊文献+

基于改进支持向量机的轨道交通光伏发电预测 被引量:2

Photovoltaic power generation prediction of rail transit based on improved support vector machine
下载PDF
导出
摘要 轨道交通系统作为用电大户,将光伏发电系统接入轨道交通牵引供电系统不但可以降低交通系统的运营成本,而且可以很好地实现节能环保.但因光伏发电具有随机性、不确定性,将光伏发电直接接入轨道交通牵引供电系统,将会对轨道交通牵引供电系统带来一定的冲击,精确的光伏发电预测是减少光电并网冲击的有效解决方法.首先,采用自适应粒子群算法提高了训练光伏发电历史数据时粒子的寻优能力,然后采用优化后的参数代入最小二乘支持向量机对光伏发电负荷进行了预测,有效地保证了预测的精度,进而提高光伏发电接入轨道交通供电系统时的稳定性. Rail transit system is a big power consumer. Connecting photovoltaic power generation system to rail transit system can not only reduce the operation cost of transportation system, but also realize energy saving and environmental protection. However, due to the randomness and uncertainty of photovoltaic power generation, the direct access of photovoltaic power generation to rail transit power supply system will bring a certain impact on rail transit power supply system. Accurate photovoltaic power generation prediction is an efective way to reduce the impact of photovoltaic grid connection. Firstly, the adaptive particle swarm optimization algorithm is used to improve the optimization ability of the particles in the training of the historical data of photovoltaic power generation, and then the optimized parameters are substituted into the least squares support vector machine to predict the photovoltaic power generation load, which effectively ensures the accuracy of the prediction, and thus improves the stability of the photovoltaic power generation when connected to the rail transit power supply system.
作者 黄元生 田立霞 孙仕泽 邓佳佳 赵恒凤 HUANG Yuansheng;TIAN Lixia;SUN Shize;DENG Jiajia;ZHAO Hengfeng(College of Economy and Management,North China Electric Power University,Baoding 071003,China;State Grid Communication Canpany,Beijing 100052,China;PetroChina North China Petrochemical Company,Renqiu 062552,China)
出处 《河北大学学报(自然科学版)》 CAS 北大核心 2021年第3期238-244,共7页 Journal of Hebei University(Natural Science Edition)
基金 国家自然科学基金面上项目(71471061)。
关键词 轨道交通 光伏发电 支持向量机 预测 rail transit PV generation support vector machine(SVM) forecasting
  • 相关文献

参考文献23

二级参考文献256

共引文献394

同被引文献23

引证文献2

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部