基于关联信息熵和轻量级梯度提升机的油纸绝缘特征优选策略

Abstract
Figure/Table
References
Related Citation (1)

Download: PDF (921 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract In order to fully explore the aging features, which are beneficial for comprehensive diagnosis results of transformer oil-paper insulation, a feature optimization strategy based on correlation information entropy and light gradient boosting machine (LightGBM) is proposed. Firstly, the initial high-dimensional feature space is formed with various time-domain features, which are extracted from the measured data of dielectric response of transformers in different aging states. Secondly, the correlation and redundancy of feature subsets is measured by correlation information entropy. Then the importance of features is evaluated according to LightGBM, so as to obtain the optimal feature space. Finally, the diagnostic performance of the optimal feature space is compared and analyzed against different control groups, and the superiority of the optimal feature space determined through the proposed optimization strategy is effectively verified.

Key words： oil-paper insulation aging comprehensive diagnosis correlation information entropy gradient boosting algorithm feature selection

Received: 23 October 2023

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	LAI Wenhong
	LIU Qingzhen
	YAN Renwu

Cite this article:

LAI Wenhong,LIU Qingzhen,YAN Renwu. Optimization strategy for oil-paper insulation features based on correlation information entropy and light gradient boosting machine[J]. Electrical Engineering, 2024, 25(1): 34-41.

URL:

https://dqjs.cesmedia.cn/EN/Y2024/V25/I1/34

[1] MHARAKURWA E T.In-service power transformer life time prospects: review and prospects[J]. Journal of Electrical and Computer Engineering, 2022, 2022: 1-20.
[2] 徐晴川, 王圣康, 林福昌, 等. 基于时频域介电响应的绝缘油弛豫过程分析[J]. 电工技术学报, 2022, 37(9): 2355-2365.
[3] 蔡金锭, 陈汉城. 基于陷阱密度谱特征量的油纸绝缘变压器老化诊断[J]. 高电压技术, 2017, 43(8): 2574-2581.
[4] 张晓燕, 刘庆珍, 蔡金锭. 基于末端双点解析法的变压器油纸绝缘新特征量提取及老化诊断[J]. 高电压技术, 2019, 45(10): 3317-3326.
[5] 蔡超, 刘庆珍, 范映, 等. 基于油纸绝缘变压器回复电压法的半峰周期极化谱分析[J]. 电气技术, 2020, 21(3): 79-83.
[6] 林晓宁, 蔡金锭. 基于粗糙集理论的变压器油纸绝缘状态评估[J]. 电力系统保护与控制, 2019, 47(7): 22-29.
[7] 邹阳, 俞豪奕, 金涛. 融合模糊K近邻及证据理论的变压器油纸绝缘状态评估方法[J]. 电力系统保护与控制, 2023, 51(14): 55-63.
[8] 苏凯强, 刘庆珍. 基于TLS-ESPRIT算法的变压器油纸绝缘等效电路参数辨识及新特征量提取[J]. 电气技术, 2022, 23(7): 89-96.
[9] 刘庆珍, 张晓燕, 蔡金锭. 油纸绝缘弛豫法谱线特征量提取及老化诊断[J]. 电机与控制学报, 2020, 24(5): 124-134.
[10] 刘庆珍, 陈俊鸿. 基于等级云模型的油纸绝缘老化状态评估[J]. 高压电器, 2023, 59(1): 176-184.
[11] 刘庆珍, 黄昌硕. 基于FCBF特征选择和XGBoost原则的油纸绝缘介电响应特征量优选研究[J]. 电力系统保护与控制, 2022, 50(15): 50-59.
[12] 杨峰, 唐超, 周渠, 等. 基于等效电路的油纸绝缘系统受潮状态分析[J]. 电工技术学报, 2020, 35(21): 4586-4596.
[13] 蔡金锭, 叶荣, 陈汉城. 回复电压多元参数回归分析的油纸绝缘老化诊断方法[J]. 电工技术学报, 2018, 33(21): 5080-5089.
[14] 高浩, 刘庆珍, 蔡金锭. 基于去极化电流Prony拟合的油纸绝缘德拜参数辨识方法[J]. 高压电器, 2020, 56(11): 210-218.
[15] 杨银松, 郭英, 李红光, 等. 基于CMFS-MIC特征选择的跳频电台个体识别方法[J]. 计算机应用研究, 2019, 36(12): 3811-3814.
[16] 周璇, 熊智翔, 黄晓斐, 等. 基于两步特征选择和贝叶斯优化LightGBM的冷水机组故障诊断策略研究[J]. 建筑科学, 2022, 38(12): 51-60.
[17] 傅泽坤, 刘庆珍. 基于物元可拓模型的电力变压器绝缘老化研究[J]. 电气技术, 2021, 22(5):32-37.
[18] 蔡金锭, 陈汉城. 基于样本集的变压器油纸绝缘状态区间灰靶分类及老化诊断[J]. 高电压技术, 2018, 44(8): 2486-2492.