融合微气象参数预测的输电线动态增容模型  

Dynamic Line Rating Model of Transmission Line Combined with Prediction of Micrometeorological Parameters

作　　者：刘志成 董向明 严昊 李群山 易本顺[2] LIU Zhicheng;DONG Xiangming;YAN Hao;LI Qunshan;YI Benshun(Central China Branch of State Grid Corporation of China,Wuhan 430077,China;Electronic Information School,Wuhan University,Wuhan 430072,China)

机构地区：[1]国家电网公司华中分部,武汉430077 [2]武汉大学电子信息学院,武汉430072

出　　处：《电力系统及其自动化学报》2022年第1期56-64,共9页Proceedings of the CSU-EPSA

基　　金：国网华中分部科技项目(SGHZ0000DKJS2000269)。

摘　　要：为了挖掘输电线路的输送潜力和保障输电线路的稳定运行,提出了经验模态分解-双向长短期记忆网络-贝叶斯优化预测模型,从而实现微气象参数递推多步预测。基于预测误差建立高斯分布模型,从预测结果中选择相对保守的气象参数代入热平衡方程进行输电线动态容量计算;实现对输电线路所跨越区间中环境最恶劣部分的微气象数据日前预测,以及输电线路动态容量的保守计算。实验仿真结果表明,本文提出的动态增容技术能够在保证输电线路安全可靠的前提下,大幅挖掘现有输电线路的传输潜力,对输电网络的运行有一定的指导作用。To tap the transmission potential of transmission lines while ensuring their stable operation,a method of empirical mode decomposition-bidirectional long short-term memory-Bayesian optimization(EMD-BiLSTM-BO)prediction model is proposed to realize the recursive multi-step prediction of micro-meteorological parameters.Furthermore,a Gaussian distribution model is established based on the prediction error,and the relatively conservative meteorological parameters are selected from the prediction results,which are further utilized to calculate the dynamic capacity of trans⁃mission lines according to the heat balance equation.This method realizes the day-ahead prediction of micro-meteorolog⁃ical data under the worst part of the regional environment covered by transmission lines and the conservative capacity calculation of transmission lines.Finally,the simulation results of a case study show that the proposed dynamic line rating technology can greatly tap the transmission potential of existing transmission lines on the premise of ensuring their safety and reliability,providing guidance to the operation of transmission network.

关 键 词：经验模态分解 双向长短期记忆网络 动态增容 微气象 递推多步预测 

分 类 号：TM751[电气工程—电力系统及自动化]