{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,3]],"date-time":"2026-03-03T09:16:43Z","timestamp":1772529403616,"version":"3.50.1"},"reference-count":33,"publisher":"MDPI AG","issue":"7","license":[{"start":{"date-parts":[[2025,6,23]],"date-time":"2025-06-23T00:00:00Z","timestamp":1750636800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Natural Science Foundation of Jilin Province","award":["20240101104JC"],"award-info":[{"award-number":["20240101104JC"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Entropy"],"abstract":"<jats:p>Dissolved Gas Analysis (DGA) of transformer oil is a critical technique for transformer fault diagnosis that involves analyzing the concentration of gases to detect potential transformer faults in a timely manner. Given the issues of large model parameters and high computational resource demands in transformer DGA diagnostics, this study proposes a lightweight convolutional neural network (CNN) model for improving gas ratio methods, combining Knowledge Distillation (KD) and recursive plots. The approach begins by extracting features from DGA data using the ratio method and Multiscale sample entropy (MSE), then reconstructs the state space of the feature data using recursive plots to generate interpretable two-dimensional image features. A deep feature extraction process is performed using the ResNet50 model, integrated with the Convolutional Block Attention Module (CBAM). Subsequently, the Sparrow Optimization Algorithm (SSA) is applied to optimize the hyperparameters of the ResNet50 model, which is trained on DGA data as the teacher model. Finally, a dual-path distillation mechanism is introduced to transfer the efficient features and knowledge from the teacher model to the student model, MobileNetV3-Large. The experimental results show that the distilled model reduces memory usage by 83.5% and computation time by 73.2%, significantly lowering computational complexity while achieving favorable performance across various evaluation metrics. This provides a novel technical solution for the improvement of gas ratio methods.<\/jats:p>","DOI":"10.3390\/e27070669","type":"journal-article","created":{"date-parts":[[2025,6,24]],"date-time":"2025-06-24T08:50:57Z","timestamp":1750755057000},"page":"669","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":2,"title":["Transformer Fault Diagnosis Based on Knowledge Distillation and Residual Convolutional Neural Networks"],"prefix":"10.3390","volume":"27","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-1480-1314","authenticated-orcid":false,"given":"Haikun","family":"Shang","sequence":"first","affiliation":[{"name":"Key Laboratory of Modern Power System Simulation and Control & Renewable Energy Technology, Ministry of Education, Northeast Electric Power University, Jilin 132012, China"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-1669-8784","authenticated-orcid":false,"given":"Yanlei","family":"Wei","sequence":"additional","affiliation":[{"name":"Key Laboratory of Modern Power System Simulation and Control & Renewable Energy Technology, Ministry of Education, Northeast Electric Power University, Jilin 132012, China"}]},{"given":"Shen","family":"Zhang","sequence":"additional","affiliation":[{"name":"Key Laboratory of Modern Power System Simulation and Control & Renewable Energy Technology, Ministry of Education, Northeast Electric Power University, Jilin 132012, China"}]}],"member":"1968","published-online":{"date-parts":[[2025,6,23]]},"reference":[{"key":"ref_1","first-page":"4273","article-title":"Current status and development trends of transformer arc fault explosion-proof and ignition prevention technology","volume":"50","author":"Zhou","year":"2024","journal-title":"High Volt. Eng."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"97","DOI":"10.1049\/nde2.12082","article-title":"Traditional fault diagnosis methods for mineral oil\u2014Immersed power transformer based on dissolved gas analysis: Past, present and future","volume":"7","author":"Nanfak","year":"2024","journal-title":"IET Nanodielectr."},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Saad, A.M.A., Ibrahim, B.M.T., Rizk, F., and Sherif, S.M.G. (2025). Transformer fault diagnose intelligent system based on DGA methods. Sci. Rep., 15.","DOI":"10.1038\/s41598-024-78293-7"},{"key":"ref_4","first-page":"3873","article-title":"Transformer fault diagnosis using dissolved gas anomaly selection method","volume":"49","author":"Bai","year":"2023","journal-title":"High Volt. Eng."},{"key":"ref_5","first-page":"4129","article-title":"Fault diagnosis method for transformers based on improved B-spline theory with three-ratio","volume":"34","author":"Yang","year":"2024","journal-title":"Proc. CSEE"},{"key":"ref_6","unstructured":"(2022). Mineral Oil-Impregnated Electrical Equipment in Service\u2014Guide to the Interpretation of Dissolved and Free Gases Analysis. Standard No. IEC 60599: 2022."},{"key":"ref_7","first-page":"2575","article-title":"Improved intelligent methods for power transformer fault diagnosis based on tree ensemble learning and multiple feature vector analysis","volume":"105","author":"Hechifa","year":"2023","journal-title":"Electr. Eng."},{"key":"ref_8","first-page":"63","article-title":"Analysis of typical transformer fault cases based on the three-ratio method","volume":"24","author":"Beimin","year":"2023","journal-title":"Electr. Technol."},{"key":"ref_9","first-page":"135","article-title":"Transformer fault diagnosis method based on BP neural network and improved three-ratio method","volume":"37","author":"Tian","year":"2024","journal-title":"Ind. Control. Comput."},{"key":"ref_10","first-page":"115","article-title":"Basic reliability distribution calculation method for transformer fault based on the improved three-ratio method","volume":"43","author":"Zhang","year":"2015","journal-title":"Power Syst. Prot. Control."},{"key":"ref_11","first-page":"2753","article-title":"Stratified Multivariate Multiscale Dispersion Entropy for Physiological Signal Analysis","volume":"69","author":"Kafantaris","year":"2022","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"106975","DOI":"10.1016\/j.ymssp.2020.107073","article-title":"Bearing fault detection and recognition methodology based on weighted multiscale entropy approach","volume":"147","author":"Minhas","year":"2021","journal-title":"Mech. Syst. Signal Process."},{"key":"ref_13","first-page":"102424","article-title":"Hydrological complexity analysis using multiscale entropy: Methodological explorations and insights","volume":"85","author":"Liu","year":"2023","journal-title":"Ecol. Inform."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Orlando, G., and Lampart, M. (2023). Expecting the Unexpected: Entropy and Multifractal Systems in Finance. Entropy, 25.","DOI":"10.3390\/e25111527"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"80079","DOI":"10.1109\/ACCESS.2019.2918560","article-title":"Multiscale fluctuation-based dispersion entropy and its applications to neurological diseases","volume":"7","author":"Azami","year":"2019","journal-title":"IEEE Access"},{"key":"ref_16","first-page":"18","article-title":"Two-dimensional multiscale entropy analysis: Applications to image texture evaluation","volume":"146","author":"Silva","year":"2018","journal-title":"Signal Process."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"021906","DOI":"10.1103\/PhysRevE.71.021906","article-title":"Multiscale entropy analysis of biological signals","volume":"71","author":"Costa","year":"2002","journal-title":"Phys. Rev. E"},{"key":"ref_18","first-page":"14","article-title":"Transformer fault diagnosis method based on adaptive deep learning model","volume":"12","author":"Mou","year":"2018","journal-title":"S. China Grid Technol."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"111142","DOI":"10.1016\/j.ress.2025.111142","article-title":"Intelligent diagnosis method for early faults of electric-hydraulic control system based on residual analysis","volume":"261","author":"Kong","year":"2025","journal-title":"Reliab. Eng. Syst. Saf."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"116818","DOI":"10.1016\/j.oceaneng.2024.116818","article-title":"Concurrent fault diagnosis method for electric-hydraulic system: Subsea blowout preventer system as a case study","volume":"294","author":"Kong","year":"2024","journal-title":"Ocean. Eng."},{"key":"ref_21","first-page":"658","article-title":"Transformer fault diagnosis based on oil gas analysis using ReLU-DBN method","volume":"42","author":"Dai","year":"2018","journal-title":"Power Syst. Technol."},{"key":"ref_22","first-page":"7","article-title":"Transformer load prediction research based on deep learning","volume":"15","author":"Wang","year":"2024","journal-title":"Electr. Eng."},{"key":"ref_23","unstructured":"Hinton, E.G., Vinyals, O., and Dean, J. (2015). Distilling the knowledge in a neural network. arXiv."},{"key":"ref_24","first-page":"1390","article-title":"Incremental partial discharge recognition method combining knowledge distillation with graph neural network","volume":"38","author":"Zhang","year":"2023","journal-title":"Trans. China Electrotech. Soc."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"e2485","DOI":"10.7717\/peerj-cs.2485","article-title":"Foreground separation knowledge distillation for object detection","volume":"10","author":"Li","year":"2024","journal-title":"PeerJ. Comput. Sci."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"126519","DOI":"10.1016\/j.neucom.2023.126519","article-title":"Incremental event detection via an improved knowledge distillation based model","volume":"551","author":"Lin","year":"2023","journal-title":"Neurocomputing"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"236018","DOI":"10.1016\/j.jpowsour.2024.236018","article-title":"Deep learning-based fault diagnosis of high-power PEMFCs with ammonia-based hydrogen sources","volume":"629","author":"Chen","year":"2025","journal-title":"J. Power Sources"},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Rybczak, M., and Kozakiewicz, K. (2024). Deep machine learning of MobileNet, Efficient, and Inception models. Algorithms, 17.","DOI":"10.3390\/a17030096"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"4702","DOI":"10.1109\/TIP.2024.3445737","article-title":"Fast and high-performance learned image compression with improved checkerboard context model, deformable residual module, and knowledge distillation","volume":"33","author":"Fu","year":"2024","journal-title":"IEEE Trans. Image Process."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"1539","DOI":"10.1016\/j.procs.2024.04.145","article-title":"EMG physical action detection using recurrence plot approach","volume":"235","author":"Ajayan","year":"2024","journal-title":"Procedia Comput. Sci."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"1145","DOI":"10.1007\/s11633-023-1445-5","article-title":"Hybrid CBAM-EfficientNetV2 fire image recognition method with label smoothing in detecting tiny targets","volume":"21","author":"Wang","year":"2024","journal-title":"Mach. Intell. Res."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Xu, K., Chen, Y., Zhang, X., Ge, Y., Zhang, X., Li, L., and Guo, C. (2024). Improved sparrow search algorithm based on multistrategy collaborative optimization performance and path planning applications. Processes, 12.","DOI":"10.3390\/pr12122775"},{"key":"ref_33","first-page":"3845","article-title":"MSSA-SVM transformer fault diagnosis method based on TLR ADASYN balanced data set","volume":"47","author":"Yu","year":"2021","journal-title":"High Volt. Eng."}],"container-title":["Entropy"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1099-4300\/27\/7\/669\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,9]],"date-time":"2025-10-09T17:56:55Z","timestamp":1760032615000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1099-4300\/27\/7\/669"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,23]]},"references-count":33,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2025,7]]}},"alternative-id":["e27070669"],"URL":"https:\/\/doi.org\/10.3390\/e27070669","relation":{},"ISSN":["1099-4300"],"issn-type":[{"value":"1099-4300","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,6,23]]}}}