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

DAI Hao; HU Dong; YANG Tongliang; FU Qiang; YANG Yong; TANG Chao; TAN Weimin

doi:10.13296/j.1001-1609.hva.2026.03.007

您当前的位置：

首页 >

文章列表页 >

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

Research & Analysis | 更新时间：2026-03-11

- Prediction of Dissolved Gas in Transformer Oil Based on Optimized VMD-TCN-LSTM
- High Voltage Apparatus Vol. 62, Issue 3, Pages: 47-60(2026)
- 作者机构：
  
  西南大学工程技术学院，重庆 400715
  西南大学智能电网及装备新技术国际研发中心，重庆 400715
- 作者简介：
- 基金信息：
- DOI：10.13296/j.1001-1609.hva.2026.03.007
  CLC：
- Received：06 August 2025，
  
  Revised：2025-10-12，
  
  Published：16 March 2026
- 稿件说明：
移动端阅览
DAI Hao, HU Dong, YANG Tongliang, et al. Prediction of Dissolved Gas in Transformer Oil Based on Optimized VMD-TCN-LSTM[J]. High Voltage Apparatus, 2026, 62(3): 47-60.
DOI：

DAI Hao, HU Dong, YANG Tongliang, et al. Prediction of Dissolved Gas in Transformer Oil Based on Optimized VMD-TCN-LSTM[J]. High Voltage Apparatus, 2026, 62(3): 47-60. DOI： 10.13296/j.1001-1609.hva.2026.03.007.

摘要

针对非平稳变压器油中溶解气体序列既有长期趋势又有短期细微波动的复杂特性，文中将黄金正弦算法（GSA）优化的麻雀搜索算法（SSA）与变分模态分解（VMD）组合构成GSSA-VMD模型；对原始变压器油中溶解气体序列使用GSSA-VMD分解，最终得到一组平稳的模态分量；其次，为了精准预测变压器气体序列长期趋势和短期波动，文中将时序卷积网络（TCN）与长短期记忆网络（LSTM）组合起来，并与GSSA-VMD组合构成变压器油中溶解气体含量组合预测模型；最后，文中选用变压器油中溶解气体CO

进行实验验证，与VMD-TCN-LSTM、EMD-TCN-LSTM和GSSA-VMD-LSTM模型进行对比实验，实验结果得出文中提出的变压器油中溶解气体混合预测模型效果最佳，平均绝对百分比误差MAPE值为0.71%，均方根误差RMSE值为9.04 μL/L。

Abstract

As for the such complex characteristics as both long-term trends and short-term subtle fluctuations of nonstationary dissolved gas sequences in transformer oi

theGSSA-VMDmodel is formed by integrating the golden sine algorithm (GSA)-optimized sparrow search algorithm (SSA) with variational mode decomposition (VMD). The original dissolved gas sequences in transformer oil are decomposed byGSSA-VMDand a set of stationa

ry modal components is finally obtained. Then

for accurately predicting the long-term trends and short-term fluctuations of gas sequences of transformer

in this paper the temporal convolutional network (TCN) and long short-term memory (LSTM) are combined

which is further combined withGSSA-VMDto form a hybrid prediction model for dissolved gas content in transformer oil. Finally

in this paper the dissolved gas CO

in teranformer oil is selected for experimental verification. It is concluded by the comparative experiments with VMD-TCN-LSTM

EMD-TCN-LSTM

and GSSA-VMD-LSTMmodels that the hybrid prediction model proposed in the paper achieves the best performance

with a mean absolute percentage error (MAPE) of 0.71% and a root mean square error (RMSE) of 9.04 μL/L .

关键词

Keywords

references

尹豪杰.基于油中溶解气体的变压器故障预测研究[D]. 成都:西南交通大学,2018.

YIN Haojie. Fault prediction of transformer based on dissolved gas analysis[D]. Chengdu:Southwest Jiaotong University,2018.

变压器油中溶解气体分析和判断导则:DL/T 722—2014[S]. 2014.

Guide to the analysis and the diagnosis of gases dissolved in transformer oil:DL/T 722—2014[S].2014.

YANG Tongliang, FANG Yun, ZHANG Chengming, et al. Prediction of dissolved gas content in transformer oil based on multi-information fusion[J]. High Voltage, 2024, 9(3): 685-699.

朱庆龙,郑智勇,杨廷方.基于LSTSVR模型的并联电抗器油中溶解气体浓度预测[J]. 机电信息,2020(30):108-109.

ZHU Qinglong,ZHENG Zhiyong,YANG Tingfang. Prediction of dissolved gas concentration in shunt reactor oil based on LSTSVR model[J]. Mechanical and Electrical Information ,2020(30):108-109.

曹正江,付文龙,文斌,等.基于TVFEMD和多模型融合的变压器油中溶解气体体积分数预测方法[J].高压电器,2024,60 (8):156-166.

CAO Zhengjiang,FU Wenlong,WEN Bin,et al. Prediction method of dissolved gas volume fraction in transformer oil based on TVFEMD and multi-model fusion[J]. High Voltage Apparatus, 2024,60(8):156-166.

范志远,杜江.基于相关变分模态分解和CNN-LSTM的变压器油中溶解气体体积分数预测[J].高电压技术,2024,50(1):263-273.

FAN Zhiyuan,DU Jiang. Prediction of dissolved gas volume fraction in transformer oil based on correlation varia-tional mode decomposition andCNN-LSTM[J]. High Voltage Engineering, 2024,50(1):263-273.

彭继慎,夏玲云,王燚增.基于CEEMD联合TGSCSO-LSTM算法的变压器油中气体浓度预测方法[J].电气工程学报,2024,19 (4):407-415.

PENG Jishen,XIA Lingyun,WANG Yizeng. Prediction method of gas concentration in transformer oil based on CEEMD and TGSCSO-LSTM[J]. Journal of Electrical Engineering,2024,19(4):407-415.

HAFIZ F, AWAL M A, QUEIROZ A R D, et al. Real-time stochastic optimization of energy storage management using deep learningbased forecasts for residential PV applications[J]. IEEE Transactions on Industry Applications, 2020, 56(3): 2216-2226.

王志勃,姜仲秋,张天舒.基于CNN-LSTM的白马湖水质预测模型研究[J]. 电脑知识与技术,2022,18(26):11-13.

WANG Zhibo,JIANG Zhongqiu,ZHANG Tianshu. Research on water quality prediction model of baima lake based onCNN-LSTM[J]. Computer Knowledge and Technology,2022,18(26):11-13.

WANG Yu, LIU Guohua. MLA-TCN: Multioutput prediction of dam displacement based on temporal convolutional network with attention mechanism[J]. Structural Control and Health Monitoring, 2023(1): 1-19.

杨文栋.基于神经网络模型的时间序列预测框架的构建与应用[D]. 大连:东北财经大学,2020.

YANG Wendong. Establishment and application of time series forecasting framework based on neural network model[D]. Dalian:Dongbei University of Finance and Economics,2020.

马星河,张登奎.基于改进经验小波变换的高压电缆局部放电噪声抑制研究[J]. 电工技术学报,2021,36(s1):353-361.

MA Xinghe,ZHANG Dengkui. Research on suppression of partial discharge noise of high voltage cable based on improved empirical wavelet transform[J]. Transactions of China Electrotechnical Society,2021,36(s1):353-361.

林琳,陈志英.基于粗糙集神经网络和振动信号的高压断路器机械故障诊断[J].电工技术学报,2020,35(s1):277-283.

LIN Lin,CHEN Zhiying. Mechanical fault diagnosis of high voltage circuit breakers based on rough set neural networks and vibration signals[J]. Transactions of China Electrotechnical Society, 2020,35(s1):277-283.

杨童亮,胡东,唐超,等.基于SMA-VMD-GRU模型的变压器油中溶解气体含量预测[J].电工技术学报,2023,38(1):117-130.

YANG Tongliang,HU Dong,TANG Chao,et al. Prediction of dissolved gas content in transformer oil based onSMA-VMD-GRU model[J]. Transactions of China Electrotechnical Society,2023,38 (1):117-130.

赵洪山,王奎,王震,等.基于VMD-MGRU的箱式变压器高压套管温度预测[J].电机与控制学报,2021,25(8):18-28.

ZHAO Hongshan,WANG Kui,WANG Zhen,et al. Temperature prediction of box-type transformer high-voltage bushing based on VMD-MGRU[J]. Electric Machines and Control,2021,25(8):18-28.

徐冬梅,王亚琴,王文川.基于VMD-TCN的月降水量预测模型[J]. 水文,2022,42(2):13-18.

XU Dongmei,WANG Yaqin,WANG Wenchuan. Monthly precipitation prediction model based onVMD-TCN[J]. Journal of China Hydrology,2022,42(2):13-18.

李青,张新燕,马天娇,等.基于ECBO-VMD-WKELM的风电功率超短期多步预测[J].电网技术,2021,45(8):3070-3078.

LI Qing,ZHANG Xinyan,MA Tianjiao,et al. Multi-step ahead ultra-short term forecasting of wind power based onECBO-VMD-WKELM[J]. Power System Technology,2021,45(8):3070-3078.

曹奇,岳东杰,高永攀,等.基于非平稳时间序列的不同趋势项提取方法对比研究[J].大地测量与地球动力学,2013,33(6):150-154.

CAO Qi,YUE Dongjie,GAO Yongpan,et al. Contrast study on various methods extracting trend extraction based on non stationary time series[J]. Journal of Geodesy and Geodynamics,2013,33(6):150-154.

JI Qiyan, HAN Lei, JIANG Lifang, et al. Short-term prediction of the significant wave height and average wave period based on the variational mode decomposition-temporal convolutional networklong short-term memory(VMD-TCN-LSTM)algorithm[J]. Ocean Science, 2023, 19(6): 1561-1578.

ALI M, KHAN A, REHMAN N U. Hybrid multiscale wind speed forecasting based on variational mode decomposition[J]. European Transactions on Electrical Power, 2018, 28(1): 2466.

叶鑫,王海瑞,李远博,等.基于优化VMD和集成模型的锂电池寿命预测[J].化工自动化及仪表,2023,50(4):500-506.

YE Xin,WANG Hairui,LI Yuanbo,et al. Remaining life prediction method for lithium batteries based on optimized vmd and integrated model[J]. Control and Instruments in Chemical Industry,2023,50(4):500-506.

XUE Jiankai, SHEN Bo. A novel swarm intelligence optimization approach: Sparrow search algorithm[J]. Systems Science & Control Engineering, 2020, 8(1): 22-34.

聂方鑫,王宇嘉.基于自适应t分布与随机游走的麻雀搜索算法[J]. 电子科技,2023,36(7):75-80.

NIE Fangxin,WANG Yujia. Sparrow search algorithm based on adaptive t-distribution and random walk[J]. Electronic Science and Technology,2023,36(7):75-80.

TANYILDIZI E, DEMIR G. Golden sine algorithm:a novel mathinspired algorithm[J]. Advances in Electrical and Computer Engineering, 2017, 17(2): 71-78.

BAI S, KOLTER J Z, KOLTUN V.An empirical evaluation of generic convolutional and recurrent networks for sequence modeling[EB/OL].[2025-08-06].https://arxiv.org/abs/1803.01271.

LUO Diansheng, FANG Jie, HE Hongying, et al. Prediction for dissolved gas in power transformer oil based on TCN and GCN[J]. IEEE Transactions on Industry Applications, 2022, 58(6): 7818-7826.

HOCHREITER S, SCHMIDHUBER J. Long short-term memory[J]. Neural Computation, 1997, 9(8): 1735-1780.

CHICCO D, WARRENS M J, JURMAN G. The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSEand RMSE in regression analysis evaluation[J]. Peerj Computer Science, 2021(7): 1-24.

刘海东,李星诚,张文豪.改进Adam优化器在瓦斯涌出量预测中的应用研究[J]. 工矿自动化,2023,49(12):25-32.

LIU Haidong,LI Xingcheng,ZHANG Wenhao. Research on the application of improved Adam training optimizer in gas emission prediction[J]. Industry and Mine Automation,2023,49(12):25-32.

李晨,梁书锋,刘传鹏,等.基于改进的集合经验模态分解的爆破振动信号趋势项消除方法[J].北京理工大学学报,2021,41 (6):636-641.

LI Chen,LIANG Shufeng,LIU Chuanpeng,et al. Trend removing method of blasting vibration signals based onMEEMD[J]. Transactions of Beijing Institute of Technology,2021,41(6):636-641.

任刚,贾继德,梅检民,等.基于变分模态分解和去趋势波动分析的柴油机振动信号去噪方法[J].内燃机工程,2019,40(2):76-81.

REN Gang,JIA Jide,MEI Jianmin,et al. Vibration signal denoising method of diesel engines based on VMD and DFA[J]. Chinese Internal Combustion Engine Engineering ,2019,40(2):76-81.

宗琳,周晓华,罗文广,等.蜣螂算法优化概率神经网络的变压器故障诊断[J].智慧电力,2024,52(5):98-104.

ZONG Lin,ZHOU Xiaohua,LUO Wenguang,et al.Transformer fault diagnosis based on probabilistic neural network optimized by dung beetle optimizer[J]. Smart Power,2024,52(5):98-104.

李永康,张丙旭,杜振斌,等.一起低端换流变压器油中缓慢产生乙炔故障分析[J].变压器,2024,61(2):74-76.

LI Yongkang,ZHANG Bingxu,DU Zhenbin,et al.Analysis of acetylene fault in oil of a low - end converter transformer[J]. Transformer,2024,61(2):74-76.

刘道生,华叙云,王仕会,等.变压器油流带电影响因素及其抑制方法[J].变压器,2024,61(7):1-12.

LIU Daosheng,HUA Xuyun,WANG Shihui,et al.Factors influencing electrification of oil flow in transformer and its suppression method[J]. Transformer,2024,61(7):1-12.

连鸿松,刘慧鑫,黄锦,等.变压器油多参量一体化分析平台[J].变压器,2025,62(4):53-61.

LIAN Hongsong,LIU Huixin,HUANG Jin,et al.Transformer oil multi-parameter integrated analysis platform[J]. Transformer,2025, 62(4):53-61.

刘建锋,李志远,周亚茹.基于漏磁场和ICOA-ResNet的变压器绕组早期故障诊断[J]. 电力系统保护与控制,2024,52(9):99-110.

LIU Jianfeng,LI Zhiyuan,ZHOU Yaru.Transformer windings based on leakage field and ICOA-ResNet early fault diagnosis[J]. Power System Protection and Control,2024,52(9):99-110.

杨金鑫,廖才波,胡雄,等.基于DGA与TPE-LightGBM的变压器故障诊断[J].电力科学与技术学报,2024,39(4):70-77.

YANG Jinxin,LIAO Caibo,HU Xiong,et al.Transformer fault diagnosis based on DGA and TPE-LightGBM[J]. Journal of Electric Power Science and Technology,2024,39(4):70-77.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Dynamic Early-warning Method Based on Dissolved Gas in Oil of 220 kV Transformer

Operation Condition Analysis of Transformer Based on Deep Belief Network

Analysis and Diagnosis on Looseness of Transformer Winding Based on VMD

Vibration Signal Diagnosis Method of Transformer Winding Looseness Defect Based on VMD and Optimized SVM

Typical Defect Identification Method of Transformer Based on Robust Local Mean Decomposition and Voiceprint Features

Related Author

CHEN Tai

ZHENG Zuolin

LOU Jianxin

KANG Zheng

WU Zhiwei

ZHENG Yuesheng

SHU Shengwen

SU Lei

Related Institution

Fujian Hoshing Hi-Tech Industrial Co.,Ltd.

College of Electrical Engineering and Automation, Fuzhou University

State Grid Shanghai Electric Power Company Electric Power Research Institute

Department of Electrical Engineering, Shanghai Jiao Tong University

College of Energy and Electrical Engineering,Hohai University

⁰