考虑不同天气影响OPGW电能损耗预测研究

魏勇; 杨会峰; 王永强; 张铁峰

魏勇,杨会峰,王永强,张铁峰.考虑不同天气影响OPGW电能损耗预测研究[J].电测与仪表,2025,62(11):26-32.
WEI Yong,YANG Huifeng,WANG Yongqiang,ZHANG Tiefeng.Research on power loss prediction of OPGW considering different weather effects[J].Electrical Measurement & Instrumentation,2025,62(11):26-32.

考虑不同天气影响OPGW电能损耗预测研究

Research on power loss prediction of OPGW considering different weather effects

DOI：10.19753/j.issn1001-1390.2025.11.003

中文关键词: 天气光纤复合架空地线架空地线电能损耗预测算法

英文关键词:weather, OPGW, overhead ground wire, power loss, prediction algorithm

基金项目:国家电网公司科技资助项目（项目编号:SGHEXT00GCJS2310463）

作者中文名	作者英文名	单位
魏勇	WEI Yong	国网河北省电力有限公司信息通信分公司，石家庄 050000
杨会峰	YANG Huifeng	国网河北省电力有限公司信息通信分公司，石家庄 050000
王永强	WANG Yongqiang	华北电力大学电力工程系，河北保定 071003
张铁峰	ZHANG Tiefeng	华北电力大学电力工程系，河北保定 071003

摘要点击次数: 18

中文摘要:

目前在我国电力系统中，光纤复合架空地线（optical fiber composite overhead ground wire， OPGW）在高压输电线路中得到广泛应用，对OPGW电能损耗进行精准预测能够提高我国电力系统的可靠性和经济性，更好地优化电力系统的设计和运行，减少因天气变化导致的电能损耗，从而提高整个电力系统的效率和稳定性。提出了一种考虑不同天气影响的OPGW电能损耗预测方法，将天气预报数据与负荷预报数据纳入预测模型，考虑了不同天气对电能损耗的综合影响，采用多种智能算法与长短期记忆（long short-term memory， LSTM）神经网络结合进行预测比较，发现蜣螂优化-长短期记忆（dung beetle optimizer and long short-term memory，DBO-LSTM）算法的预测精度更高、收敛速度更快。最后采用ATP-EMTP(alternative transient program-electromagnetic transients program)软件对预测线路进行建模仿真，将仿真结果与预测结果进行对比，两者的相对误差为5.65%，验证了预测模型能够准确预测案例线路的电能损耗，能够真实有效反映不同天气条件下OPGW电能损耗情况。

英文摘要:

At present, optical fiber composite overhead ground wire (OPGW) is widely used in high-voltage transmission lines in power system of China. Accurately predicting the power loss of OPGW can improve the reliability and economy of power system in China, better optimize the design and operation of the power system, reduce power loss caused by weather changes, and thus improving the efficiency and stability of the entire power system. This paper proposes a prediction method for OPGW power loss considering different weather effects. The weather forecast data and load forecast data are included in the prediction model, and the comprehensive impact of different weather conditions on power loss is considered. Multiple intelligent algorithms are combined with long short-term memory (LSTM) neural network for prediction comparison, and it is found that the dung beetle optimizer-long short-term memory (DBO-LSTM) algorithm has higher prediction accuracy and faster convergence speed. Finally, ATP-EMTP software is used to model and simulate the predicted line, and the simulation results are compared with the predicted results. The relative error between the two is 5.65%, verifying that the prediction model can accurately predict the power loss of the case line and can truly and effectively reflect the OPGW power loss situation under different weather conditions.

查看全文查看/发表评论下载PDF阅读器

关闭