文章摘要
付善强,王孟夏,杨明,等.架空导线载流量的多时段联合概率密度预测[J].电力系统自动化. DOI: 10.7500/AEPS20180928004.
FU Shanqiang,WANG Mengxia,YANG Ming, et al.Multi-period Joint Probability Density Forecasting for Thermal Rating of Overhead Line[J].Automation of Electric Power Systems. DOI: 10.7500/AEPS20180928004.
架空导线载流量的多时段联合概率密度预测
Multi-period Joint Probability Density Forecasting for Thermal Rating of Overhead Line
DOI:10.7500/AEPS20180928004
关键词: 架空线路  关键线挡  载流量  分位点回归  t-Copula函数  联合概率密度预测
KeyWords: overhead line  critical span  thermal rating  quantile regression  t-Copula function  joint probability density prediction
上网日期:2019-06-11
基金项目:国家自然基金项目(51407111)
作者单位E-mail
付善强 电网智能化调度与控制教育部重点实验室山东大学、国网山东省电力公司济宁供电公司 1003038430@qq.com 
王孟夏 电网智能化调度与控制教育部重点实验室山东大学 wangmx83@163.com 
杨明 电网智能化调度与控制教育部重点实验室山东大学 myang@sdu.edu.cn 
韩学山 电网智能化调度与控制教育部重点实验室山东大学 xshan@sdu.edu.cn 
陈芳 济南大学 cfunix@263.net 
李文博 国网山东省电力公司电力科学研究院 liwenbo9@qq.com 
摘要:
      受微气象环境影响,架空线路载流量波动性较强,难以被准确预测,掌握线路关键线挡载流量的分布规律对帮助运行人员把握线路未来载流量变化,充分利用架空线路载荷能力具有重要参考价值。文中基于架空线路关键线挡微气象历史数据,在分析载流量变化特性的基础上,结合分位点回归方法,首先进行载流量逐时段概率预测,而后进一步运用t-Copula函数评估多时段载流量概率分布的相关特性,建立未来多时段载流量动态相依模型,实现架空线路关键线挡载流量的多时段联合概率密度预测,得到较逐时段概率预测更为准确的载流量波动区间和分布信息。实例分析表明,所提方法可利用载流量时段间的关联性改善逐时段概率预测结果,有效缩小载流量预测结果的分布区间。
Abstract:
      Influenced by the micrometeorological conditions around the overhead line, thermal rating of the overhead line have strong volatility and are difficult to predict accurately. It is of significance for system operators to grasp the fluctuation range and distribution characteristics of the thermal ratings of the critical spans along the overhead line, thus guiding the operators to exploit the transfer capability of overhead lines. Based on the historical micrometeorological data of critical spans and the variation characteristics analysis of thermal rating, the quantile regression method is employed to predict the period-by-period probability of thermal rating. Then the t-Copula function is further used to evaluate the correlation characteristics of the probability distributions of multi-period thermal ratings. A dynamic dependence model for multi-period thermal ratings is established to realize the joint probability density prediction for multi-period thermal ratings. As well, the more accurate fluctuation interval and distribution information of thermal rating are obtained. The case studies show that the proposed method can improve the period-by-period probability prediction results by using the correlation of thermal rating between periods, and effectively reduce the distribution interval of the thermal rating prediction results.
查看全文   查看附录   查看/发表评论  下载PDF阅读器