计及气象特征的线路覆冰荷载预测OA
Prediction of ice load on power lines incorporating meteorological characteristics
针对架空线路覆冰灾害频发及传统覆冰预测方法的局限性,提出一种计及气象特征的线路覆冰荷载预测方法.首先,依据线路荷载变化量,将气象环境划分为载荷增长、平衡和衰减3个模式集,并通过灰色关联度分析,将各气象参量的关联权重融入马氏距离算法.其次,为解决气象模式集之间模糊隶属关系的问题,基于历史经验,定义了IGP(覆冰增长势能)因子和IMP(覆冰融化势能)因子,定量描述气象特征变化.最后,将隶属度与监测数据结合,形成含气象特征信息的扩充历史样本,使用SVM(支持向量机)进行覆冰荷载预测.算例对仅考虑气象参量信息、气象过程模糊分类及所提方法3种训练模式的预测结果进行了对比分析,结果表明所提预测模型预测精度更高.
To address the frequent occurrence of ice-induced disasters on overhead transmission lines and the limita-tions of conventional icing prediction methods,this paper proposes a novel ice load prediction approach that system-atically incorporates meteorological characteristics.First,based on the load patterns of transmission lines,meteoro-logical conditions are classified into three distinct regime sets:load growth,equilibrium,and decay.The grey rela-tional analysis is employed to integrate the weighted contributions of meteorological parameters into the Mahalanobis distance.Second,to resolve the fuzzy membership relationships between meteorological regimes,two physically in-terpretable indices are empirically defined:the ice growth potential(IGP)factor and ice melting potential(IMP)factor,which quantitatively characterize meteorological characteristics.Finally,membership degrees are combined with monitoring data to construct augmented historical samples embedding meteorological characteristics,and a sup-port vector machine(SVM)is applied for ice load prediction.Case studies compare three training paradigms:me-teorological parameters only,fuzzy classification of meteorological processes,and the proposed method.Results demonstrate that the proposed model achieves higher prediction accuracy.
李江;王尚玉
上海电力大学 电气工程学院,上海 200090上海电力大学 电气工程学院,上海 200090
输电线路气象特征马氏距离支持向量机
transmission linemeteorological characteristicMahalanobis distanceSVM
《浙江电力》 2026 (4)
3-11,9
国家自然科学基金(51977030)
评论