微气象分析与载流量预测结合的动态线路增容方法

Dynamic Line Rating Method Combining Micrometeorological Analysis and Ampacity Forecasting

为提升架空输电线路在用电高峰期的输电能力保证用电质量,提出一种结合Transformer模块、卷积神经网络与自回归模块的网络结构,用以进行多步负荷预测及计算线路在极端条件下允许的最大载流量限值。基于网络训练历史数据获取各气象因素与载流量之间的相互联系,再将气象数据输入网络,输出对应的气象条件下线路允许的最大载流量限值,以此来达到动态增容的目的。实验结果表明,该网络在进行载流量多步预测时具有更高的精度,对提高现有线路的输电能力具有指导性作用。

To improve the transmission capacity of overhead transmission lines in the peak power consumption period to ensure power quality,a network structure combining Transformer,convolutional neural network(CNN)and auto-regressive(AR)modules is proposed,which be used for multi-step load prediction and the calculation of maximum allowable ampacity limit of lines under extreme conditions.The complex correlation between influencing factors and the ampacity data is obtained by training the historical data based on the network.Then,the meteorological data is input into the network to output the maximum allowable ampacity limit of lines under the corresponding meteorological conditions,thus realizing the objective of dynamic line rating.Experimental results show that the proposed network has a higher accuracy in multi-step ampacity prediction,which plays a guiding role in improving the transmission capacity of the existing lines.