Improved Convolutional Neural Network Quantification of Mixed Fault Characterization Gases in Transformers Based on Raman Spectroscopy
CHEN Xin-gang1, 2, ZHANG Wen-xuan1, MA Zhi-peng1*, ZHANG Zhi-xian1, WAN Fu3, AO Yi1, ZENG Hui-min1
1. School of Electrical and Electronic Engineering, Chongqing University of Technology, Chongqing 400054, China
2. Chongqing Energy Internet Engineering Technology Research Center, Chongqing 400054, China
3. State Key Laboratory of Power Transmission Equipment & System Security and New Technology, (Chongqing University), Chongqing 400044, China
Abstract：Laser Raman spectroscopy has obvious advantages in detecting transformer fault characteristic gases. With the development of intelligent transformer condition monitoring, it is of great significance to study the fast and accurate quantitative analysis method of mixed fault characteristic gases.Conventional Raman spectral analysis requires a preprocessing process that greatly relies on human experience and spectral feature extraction. Although it can reduce the signal dimensions, it can also result in partially missing or altered spectral features. Aiming at the above problems, a method for quantitative analysis of Raman spectra based on the fusion of improved 1DCNN and LSSVR is proposed; that is, the introduction of global mean pooling and least squares support vector regression improves traditional CNN, and the use of the Dropout method to improve model generalization performance and prevent over-fitting. Design and build the transformer fault characteristic gas Raman spectroscopy detection platform, collect the Raman signal of 7 kinds of fault characteristic gases and N2, O2 mixed gases, in the spectrogram near 2 900 cm-1 frequency shift, CH4, C2H6 gases show the overlap of the spectral peaks, and when transformer overheating or partial discharge fault occurs, it will produce the main fault characteristic gas CH4, choose different content ratio of CH4, C2H6 mixed gas as a representative research object, 146 groups of mixed gas samples with different contents of CH4 and C2H6 are prepared according to different ratios. Nitrogen is chosen as the standard gas for detection, the Raman spectral data of the mixed gas samples with different content ratios are collected, and the spectral data enhancement method is utilized to construct the gas sample dataset suitable for deep neural networks. Through continuous experiments, we optimize the network structure parameters and network weights, complete the model training and test its prediction effect, compare and analyze with multiple quantitative models, study the effect of spectral preprocessing on different quantitative models, and then evaluate the model performance. The results show that when using the original data set for modeling, the improved CNN model has the best prediction accuracy and regression fitting goodness, the R2 can reach 0.999 8, and the RMSE is only 0.000 5 MPa; using preprocessed data. When modeling the set, the RMSE of the improved convolutional neural network model is 0.002 3 MPa, which is an increase of 0.001 8 compared to the modeling error using the original data set. In contrast, the errors of traditional methods have declined.The results of this study show that the proposed method integrates the spectral pre-processing, feature extraction, and quantitative analysis processes compared with the traditional Raman spectroscopy quantitative method, which simplifies the spectral analysis process based on ensuring the prediction accuracy and provides new ideas and references for the fast and accurate analysis of transformer mixed fault characteristic gases.
[1] Yang D K, Chen W G, Wan F, et al. IEEE Transactions on Dielectrics and Electrical Insulation, 2020, 27(6): 1770.
[2] WANG Jian-xin, CHEN Wei-gen, WANG Pin-yi, et al(王建新, 陈伟根, 王品一，等). Proceedings of the CSEE(中国电机工程学报), 2022, 42(16): 6136.
[3] DL/T 722-2000, Guide to the Analysis and the Diagnosis of Gases Dissolved in Transformer Oil(变压器油中溶解气体分析和判断导则). Industry Standards-Electricity(行业标准-电力), 2000.
[4] CHEN Wei-gen, ZHANG Zhi-xian, LI Jian, et al(陈伟根, 张知先, 李 剑，等). Proceedings of the CSEE(中国电机工程学报), 2020, 40(S1): 323.
[5] LI Guang-mao, QIAO Sheng-ya, ZHU Chen, et al(李光茂, 乔胜亚, 朱 晨，等). High Voltage Engineering(高电压技术), 2021, 47(6): 2007.
[6] Wan Fu, Wang Rui, Ge Hu, et al. Optics Express, 2024, 32(7): 12428.
[7] Molavi H, Yousefpour A, Mirmostafa A, et al. Chromatographia, 2017, 80(7): 1129.
[8] Tang S, Chen W, Jin L, et al. Sensors & Actuators: B. Chemical, 2020, 312: 127998.
[9] Zhou Q, Xu L, Umar A, et al. Sensors & Actuators: B. Chemical, 2018, 256: 656.
[10] CHEN Wei-gen, WAN Fu, GAO Tuo-yu, et al(陈伟根, 万 福, 高拓宇，等). Proceedings of the CSEE(中国电机工程学报), 2015, 35(18): 4833.
[11] YANG Tian-he, CHEN Wei-gen, LI Zhi-jun(杨天荷, 陈伟根, 李志军). High Voltage Engineering(高电压技术), 2020, 46(6): 1922.
[12] CHEN Xin-gang, YANG Ding-kun, TAN Hao, et al(陈新岗, 杨定坤, 谭 昊，等). High Voltage Engineering(高电压技术), 2017, 43(7): 2256.
[13] CHEN Xin-gang, FAN Yi-jie, MA Zhi-peng, et al(陈新岗, 范益杰, 马志鹏，等). Laser & Optoelectronics Progress(激光与光电子学进展), 2023, 60(21): 2130002.
[14] Yang D K, Chen W G, Zhou W R, et al. IEEE Access, 2019, 7: 127862.
[15] WANG Pin-yi, WAN Fu, WANG Jian-xin, et al(王品一, 万 福, 王建新，等). Optics and Precision Engineering(光学精密工程), 2018, 26(8): 1917.
[16] WAN Fu, WANG Rui, KONG Wei-ping, et al(万 福, 王 锐, 孔维平，等). Acta Photonica Sinica(光子学报), 2023, 52(9): 0930001.
[17] Ge Hu, Kong Weipin, Wang Rui, et al. Optics Letters, 2023, 48(8): 2186.
[18] YANG Yong, DONG Hao, WANG Shu, et al(杨 勇, 董 浩, 王 澍，等). Chinese Journal of Lasers(中国激光), 2022, 49(15): 1507405.
[19] ZHAO Yong, RONG Kang, TAN Ai-ling(赵 勇, 荣 康, 谈爱玲). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2019, 39(12): 3755.
[20] Li Dongdong, Zhao Yang, Zhao Yao. Protection and Control of Modern Power Systems, 2022, 7: 22.
[21] WU Hao, CHEN Yun-tao, ZHU Zhao-hui, et al(吴 浩, 陈运涛, 朱赵辉，等). Journal of Basic Science and Engineering(应用基础与工程科学学报), 2024, 32(1): 145.
[22] ZHOU Can-ru, WANG Zhe-tao, YANG Si-wei, et al(周粲入, 王哲涛, 杨思危，等). Chinese Journal of Analytical Chemistry(分析化学), 2023, 51(8): 1232.
