Prediction of Dissolved Gas Concentration in Transformer Oil Based on Missing Data Imputation

WANG Zhenguo; NING Fei; YUAN Bao; JIA Fei; JIA Yuyan

doi:10.13296/j.1001-1609.hva.2026.10.007

您当前的位置：

首页 >

文章列表页 >

Prediction of Dissolved Gas Concentration in Transformer Oil Based on Missing Data Imputation

Research & Analysis | 更新时间：2026-10-14

- Prediction of Dissolved Gas Concentration in Transformer Oil Based on Missing Data Imputation
- High Voltage Apparatus Vol. 62, Issue 10, Pages: 59-67(2026)
- 作者机构：
  
  内蒙古电力（集团）有限责任公司乌海供电份公司，内蒙古乌海 016000
  华北电力大学电力工程系，河北保定 071003
- 作者简介：
- 基金信息：
- DOI：10.13296/j.1001-1609.hva.2026.10.007
  CLC：
- Received：26 May 2026，
  
  Revised：2026-08-05，
  
  Published：16 October 2026
- 稿件说明：
移动端阅览
WANG Zhenguo, NING Fei, YUAN Bao, et al. Prediction of Dissolved Gas Concentration in Transformer Oil Based on Missing Data Imputation[J]. High Voltage Apparatus, 2026, 62(10): 59-67.
DOI：

WANG Zhenguo, NING Fei, YUAN Bao, et al. Prediction of Dissolved Gas Concentration in Transformer Oil Based on Missing Data Imputation[J]. High Voltage Apparatus, 2026, 62(10): 59-67. DOI： 10.13296/j.1001-1609.hva.2026.10.007.

摘要

油中溶解气体体积分数预测对于早期变压器故障诊断具有重要意义。针对当前油中溶解气体体积分数预测算法对数据完整性要求高，泛化能力弱等问题，以油浸式变压器为研究对象，提出一种基于缺失数据填补和改进卡尔曼滤波的油中溶解气体体积分数预测算法。首先以贝叶斯理论和高斯先验假设为基础，构建缺失数据填补的全概率模型，将缺失数据填补转化为求解模型参数最大似然估计问题，利用期望最大（expectation maximization，EM）算法完成模型求解，最终实现缺失数据填补。然后采用卡尔曼滤波进行气体体积分数预测，针对卡尔曼滤波预测结果易发散问题，引入BP神经网络并建立反馈机制，实现对预测结果的动态修正，提升预测精度。基于实际数据的试验结果表明，所提方法在数据缺失率为30%的情况下，填补结果的相关系数达到96.3%，相对于自回归移动平均（ARIMA）和极端随机树（extremely randomized trees，ERT）分别提升11.4%和5.7% 。同时，所提方法的预测精度较高，相对于长短时记忆神经网络（long short term memory，LSTM）和长短时记忆—卷积神经网络（long short term memory-convolution neural network，LSTM-CNN）方法预测结果的均方根误差（

RMSE

）和平均绝对误差（

MAPE

）分别降低28.5%、33.6%。

Abstract

The prediction of dissolved gas concentration in oil is of great significance for earlyfault diagnosis of transformer. To address the issues of high data integrity requirements and weak generalization ability in current prediction algorithms for dissolved gas concentration in oil

in this paper the oil immersed transformers is taken as the research object anda prediction algorithm for dissolved gas concentration in oil based on missing data imputation and an improved Kalman filtering is proposed. First

a full probability model for missing data imputation is constructed based on Bayesian theory and Gaussian prior assumption.The missing data imputation problem is transformed into solving the problem maximum likelihood estimation of model parameters

and the expectation

maximization(EM)algorithm is employed to solve the modle

ultimately achieving the imputation of missing data. Then

Kalman filtering is used for gas concentration prediction. To address the issue of divergence in Kalman filtering

a BP neural network is introduced and a feedback mechanism is established so to dynamically correct the prediction results and improve prediction accuracy. The experimental results based on actual data show under a data missing rate of 30%

thecorrelation coefficient of the imputation results achieved by the proposed dmethod reaches 96.3%

which is 11.4% and 5.7% higher than those of the autoregressiveintegrated moving average(ARIMA)and extremely randomized trees(ERT)

respectively. Meanwhile

the proposed method has a higher prediction accuracy

and the root mean square error(

RMSE

)and mean absolute error(

MAPE

)are reduced by 28.5% and 33.6% respectively compared to long short term memory(LSTM)and long short term memory-convolutional neural network(LSTM-CNN)methods.

关键词

Keywords

references

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.

Zou Yang,Yu Haoyi,Jin Tao.Evaluation method of the oil-paper insulation condition of a transformer based on fuzzy K nearest neighbor and evidence theory[J].Power System Protection and Control,2023,51(14):55-63.

林智勇,苏若杭,黄国泰,等.基于等值老化容抗的变压器油纸绝缘状态评估[J].电力系统保护与控制,2025,53(3):140-147.

Lin Zhiyong,Su Ruohang,Huang Guotai,et al.Evaluation of the oilpaper insulation state of a transformer based on equivalent aging capacitance[J].Power System Protection and Control,2025,53(3):140-147.

奚广宇,徐建军,朱景鹏.基于容性功率转移原理的高压大容量直流变压器设计[J].浙江电力,2025,44(11):125-136.

Xi Guangyu,Xu Jianjun,Zhu Jingpeng.Design of high voltage and large capacity DC transformers based on capacitive power transfer principle[J].Zhejiang Electric Power,2025,44(11):125-136.

段舒帆,成立,罗豪,等.基于虚实融合的变压器油中气体输运过程数字孪生方法研究[J].智慧电力,2025,53(6):68-77.

Duan Shufan,Cheng Li,Luo Hao,et al.Research on a digital twin method for gas transport process in transformer oil based on cyberphysical integration[J].Smart Power,2025,53(6):68-77.

Wang Lin,Littler T,Liu Xueqin.Gaussian process multi-class classification for transformer fault diagnosis using dissolved gas analysis[J].IEEE Transactions on Dielectrics and Electrical Insulation, 2021,28(5):1703-1712.

Lin Jun,Sheng Gehao,Yan Yang,et al.Prediction of dissolved gas concentrations in transformer oil based on the KPCA-FFOA-LSTM model[J].Energies,2018,11(1):225.

汤建,侯慧娟,陈洪岗,等.基于BI-GRU改进的Seq2Seq网络的变压器溶解气体体积分数预测方法[J].电力自动化设备, 2022,42(3):196-202.

Tang Jian,Hou Huijuan,Chen Honggang,et al.Concentration prediction method base don Seq2Seq network improved by BIGRU for dissolved gas in transformer oil[J].Electric Power Automation Equipment,2022,42(3):196-202.

徐鹏,吴鹏,甘强,等.500 kV变压器纯瓷套管漏油事故调查及防范措施[J].电磁避雷器,2022,3(307):186-190.

Xu Peng,Wu Peng,Gan Qiang,et al.Investigation on oil leakage accident of pure porcelain bushing of 500 kV transformer and preventive measures[J].Insulators and Surge Arresters,2022,3 (307):186-190.

王杰,李永鑫,张军亮,等.基于SBOA-TVFEMD的变压器油中溶解气体浓度预测[J].变压器,2025,62(5):23-31.

Wang Jie,Li Yongxin,Zhang Junliang,et al.Prediction of dissolved gas concentration in transformer oil based onSBOA-TVFEMD[J].Transformer,2025,62(5):23-31.

黄丰,郭淳,邱冰兵,等.基于电磁—结构场的油浸式变压器振动特性及信号反演研究[J].电力科学与技术学报,2025,40 (2):276-286.

Huang Feng, Guo Chun, Qiu Bingbing, et al.Vibration characteristics and signal inversion of oil-immersed transformer based on electromagnetic-structural field[J].Journal of Electric Power Science and Technology,2025,40(2):276-286.

年泓昌.变压器绝缘油中含气量超标分析处理与防范措施[J].变压器,2024,61(8):73-75.

Nian Hongchang.Analysis treatment and preventive measures for excessive gas content in transformer insulation oil[J].Transformer, 2024,61(8):73-75.

陈铁,陈一夫,李咸善,等.基于SDS-SSA-LSTM的变压器油中溶解气体体积分数预测[J].电子测量技术,2022,45(12):6-11.

Chen Tie,Chen Yifu,Li Xianshan,et al.Prediction of dissolved gas concentration in transformer oil based on SDS-SSA-LSTM[J].Electronic Measurement Technology,2022,45(12):6-11.

Jiang Jun,Chen Ruyi,Zhang Chaohai,et al.Dynamic fault prediction of power transformers based on lasso regression and change point detection by dissolved gas analysis[J].IEEE Transactions on Dielectrics and Electrical Insulation,2020,27(6):2130-2137.

刘慧鑫,张江龙,连鸿松,等.基于时间序列模型的变压器油中溶解气体预测[J].高压电器,2019,55(12):193-199.

Liu Huixin,Zhang Jianglong,Lian Hongsong,et al.Prediction of the gases dissolved in transformer oil by sequential learning[J].High Voltage Apparatus,2019,55(12):193-199.

张鹏,齐波,李成榕,等.电力变压器油中溶解气体特性影响因素的量化分析[J].中国电机工程学报,2021,41(10):3620-3631.

Zhang Peng,Qi Bo,Li Chengrong,et al.Quantitative analysis of influence factors of dissolved gas characteristics in power transformer oil[J].Proceedings of the CSEE,2021,41(10):3620-3631.

安国庆,史哲文,马世峰,等.基于RF特征优选的WOA-SVM变压器故障诊断[J].高压电器,2022,58(2):171-178.

An Guoqing,Shi Zhewen,Ma Shifeng,et al.Fault diagnosis of WOASVM transformer based on RF feature optimization[J].High Voltage Apparatus,2022,58(2):171-178.

陈铁,陈卫东,李咸善,等.基于油中溶解气体分析的变压器故障预测[J].电子测量技术,2021,44(22):25-31.

Chen Tie,Chen Weidong,Li Xianshan,et al.Transformer fault prediction based on analysis of dissolved gas in oil[J].Electronic Measurement Technology,2021,44(22):25-31.

Zeng Bo,Guo Jun,Zhang Fan,et al.Prediction model for dissolved gas concentration in transformer oil based on modified grey wolf optimizer and LSSVM with grey relational analysis and empirical mode decomposition[J].Energies,2020,13(2):422.

粟磊,王廷涛,殷浩然,等.基于GWO-LSTM与NKDE的变压器油中溶解气体体积分数点一区间联合预测方法[J].高压电器,2022,58(11):88-97.

Su Lei,Wang Tingtao,Yin Haoran,et al.Point-interval joint prediction method of dissolved gas volume fraction in transformer oil based onGWO-LSTMand NKDE[J].High Voltage Apparatus, 2022,58(11):88-97.

陈铁,张治藩,李咸善,等.基于混合模态分解和LSTM-CNN的变压器油中溶解气体体积分数预测[J].中国电力,2023,56 (1):132-141.

Chen Tie,Zhang Zhifan,Li Xianshan,et al.Prediction of dissolved gas concentration in transformer oil based on hybrid mode decomposition andLSTM-CNN[J].Electric Power,2023,56(1):132-141.

周远翔,林孟龙,陈健宁,等.基于自注意力生成对抗网络的电力设备在线监测缺失数据填补[J].高电压技术,2023,49(5):1795-1809.

Zhou Yuanxiang,Lin Menglong,Chen Jianning,et al.Missing data imputation for online monitoring of power equipment based on selfattention generative adversarial networks[J].High Voltage Engineering,2023,49(5):1795-1809.

严英杰,盛戈皞,陈玉峰,等.基于时间序列分析的输变电设备状态大数据清洗方法[J].电力系统自动化,2015,39(7):138-144.

Yan Yingjie,Sheng Gehao,Chen Yufeng,et al.Cleaning method for big data of power transmission and transformation equipment state based on time sequence analysis[J].Automation of Electric Power Systems,2015,39(7):138-144.

廖才波,杨金鑫,邱志斌,等.基于缺失数据填补的油浸式变压器故障诊断[J].高电压技术,2024,50(9):4091-4100.

Liao Caibo,Yang Jinxin,Qiu Zhibin,et al.Fault diagnosis for oilimmersed transformer based on missing data imputation[J].High Voltage Engineering,2024,50(9):4091-4100.

程逍,李平,郭凌旭,等.基于量测数据贝叶斯概率矩阵分解的变压器运行状态监测方法[J].电力系统及其自动化学报, 2022,34(1):100-107.

Cheng Xiao,Li Ping,Guo Lingxu,et al.Transformer operation status monitoring method based on Bayesian probabilistic matrix factorization of measured data[J].Proceedings of the CSU-EPSA,2022,34(1):100-107.

Jolliffe I T.Principal component analysis[M].New York:Springer, 1986.

肖怀硕,李清泉,施亚林,等.灰色理论—变分模态分解和NSGA-Ⅱ优化的支持向量机在变压器油中气体预测中的应用[J].中国电机工程学报,2017,37(12):3643-3653.

Xiao Huaishuo,Li Qingquan,Shi Yalin,et al.Prediction of dissolved gases in oil for transformer based on grey theory variational mode decomposition and support vector machine improved byNSGA-Ⅱ[J].Proceedings of the CSEE,2017,37(12):3643-3653.

Nyamundanda G,Brennan L,Gormley I C.Probabilistic principal component analysis for metabolomic data[J].BMC Bioinformatics, 2010(11):571-581.

黄靖,陈斌艺,黄晓生,等.基于卡尔曼滤波便捷式接地线接地电阻检测装置[J].电磁避雷器,2023(1):65-73.

Huang Jing,Chen Binyi,Huang Xiaosheng,et al.Portable grounding resistance detection device of grounding wire based on kalman filer[J].Insulators and Surge Arresters,2023(1):65-73.

张佳伟,张江南,吴坡,等.基于融合容积卡尔曼滤波的直流微电网虚假数据检测[J].智慧电力,2024,52(3):87-93.

Zhang Jiawei,Zhang Jiangnan,Wu Po,et al.False data detection in DC microgrids based on fusion of cubature kalman filters[J].Smart Power,2024,52(3):87-93.

Wang Zheng,Wang Bo,Liu Chun,et al.Improved BP neural network algorithm to wind power forecast[J].The Journal of Engineering,2017(13):940-943.

左宏基,陈宏军,徐峰亮,等.基于BP神经网络建模的并联型有源电力滤波器谐波综合治理控制算法[J].电力电容器与无功补偿,2022,43(1):50-56.

Zuo Hongji,Chen Hongjun,Xu Fengliang,et al.Integrated control algorithm on harmonics of shunt active power filter based on BP neural network modeling[J].Power Capacitor & Reactive Power Compensation,2022,43(1):50-56.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Prediction of Dissolved Gas in Transformer Oil Based on Optimized VMD-TCN-LSTM

Prediction Method of Dissolved Gas in Transformer Oil Based on Multi Decomposition Strategy and BO-LSTM

Application of Inverse Algorithm of Second Balance Degassing to Chromatography Test

Partial Discharge Characteristics of Typical Defects in 110 kV Transformer and Live Detection Technology

Calculation of Winding Hot Spot Temperature of Oil Immersed Transformer by the Discontinuous Galerkin Method

Related Author

TAN Weimin

TANG Chao

YANG Yong

FU Qiang

YANG Tongliang

HU Dong

DAI Hao

DU Jiang

Related Institution

School of Engineering and Technology, Southwest University

International R&D Center for Smart Grid and New Equipment Technology, Southwest University

Key Laboratory of Electromagnetic Field and Electrical Apparatus Reliability of Hebei Province, Hebei University of Technology

State Key Laboratory of Reliability and Intelligence of Electrical Equipment, Hebei University of Technology

Quzhou Electric Power Bureau

⁰