摘要
文章制备环氧树脂样片,并开展加速热老化试验,结合频域介电谱(FDS)测试技术获取样片频域介电特性数据,采用Havriliak-Negami模型提取老化特征参数,并基于极限学习机(ELM)建立老化状态评估模型。结果表明:不同老化温度和周期下样片的介电损耗参数随老化程度增加而显著变化,方法的集中准确率高达97.5%,较传统介电损耗角正切(tanδ)单参数诊断方法提升约22%。研究结果为环氧树脂老化状态评估提供了高效精准的方法,对干式套管绝缘性能的预测和寿命评估具有重要意义。
In the paper,epoxy resin samples were prepared,accelerated thermal aging tests were carried out,and the dielectric characteristics data of the sample in the frequency domain were obtained by combining frequency domain dielectric spectroscopy(FDS)testing technology,the aging characteristic parameters were extracted by Havriliak-Negami model,and the aging state evaluation model was established based on the extreme learning machine(ELM).The results show that the dielectric loss parameters of the samples change significantly with the increase of aging degree under different aging temperatures and periods,and the centralized accuracy of the method is as high as 97.5%,which is about 22%higher than that of the traditional single-parameter diagnostic method of dielectric loss tangent(tanδ).The study results provide an efficient and accurate method for the assessment of epoxy resin aging status,which is of great significance for the prediction and life evaluation of dry sleeve insulation performance.
作者
乔亚军
潘欢
刘芸
周福升
QIAO Yajun;PAN Huan;LIU Yun;ZHOU Fusheng(Substation Management DivisionⅢ,Guangzhou Power Supply Bureau,Guangzhou 511449,China;Electric Power Research Institute,China Southern Power Grid Co.,Ltd.,Guangzhou 510530,China;School of Electrical Engineering,Xi'an Jiaotong University,Xi'an 710049,China)
出处
《塑料科技》
北大核心
2025年第9期186-191,共6页
Plastics Science and Technology
基金
国家自然科学基金(52107028)
中国南方电网有限责任公司创新项目[No.030100KK52222048(GDKJXM20222370)]。
