{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,27]],"date-time":"2026-03-27T15:09:53Z","timestamp":1774624193794,"version":"3.50.1"},"reference-count":77,"publisher":"MDPI AG","issue":"4","license":[{"start":{"date-parts":[[2021,4,20]],"date-time":"2021-04-20T00:00:00Z","timestamp":1618876800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Algorithms"],"abstract":"<jats:p>This study investigates the use of machine-learning approaches to interpret Dissolved Gas Analysis (DGA) data to find incipient faults early in oil-impregnated transformers. Transformers are critical pieces of equipment in transmitting and distributing electrical energy. The failure of a single unit disturbs a huge number of consumers and suppresses economic activities in the vicinity. Because of this, it is important that power utility companies accord high priority to condition monitoring of critical assets. The analysis of dissolved gases is a technique popularly used for monitoring the condition of transformers dipped in oil. The interpretation of DGA data is however inconclusive as far as the determination of incipient faults is concerned and depends largely on the expertise of technical personnel. To have a coherent, accurate, and clear interpretation of DGA, this study proposes a novel multinomial classification model christened KosaNet that is based on decision trees. Actual DGA data with 2912 entries was used to compute the performance of KosaNet against other algorithms with multiclass classification ability namely the decision tree, k-NN, Random Forest, Na\u00efve Bayes, and Gradient Boost. Investigative results show that KosaNet demonstrated an improved DGA classification ability particularly when classifying multinomial data.<\/jats:p>","DOI":"10.3390\/a14040128","type":"journal-article","created":{"date-parts":[[2021,4,20]],"date-time":"2021-04-20T13:58:04Z","timestamp":1618927084000},"page":"128","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":36,"title":["A Multinomial DGA Classifier for Incipient Fault Detection in Oil-Impregnated Power Transformers"],"prefix":"10.3390","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-1233-7984","authenticated-orcid":false,"given":"George","family":"Odongo","sequence":"first","affiliation":[{"name":"African Center of Excellence in Internet of Things, University of Rwanda, Kigali P.O. Box 42 85, Rwanda"},{"name":"Department of Computer Science, Egerton University, Egerton P.O. Box 536-20115, Kenya"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7451-4977","authenticated-orcid":false,"given":"Richard","family":"Musabe","sequence":"additional","affiliation":[{"name":"African Center of Excellence in Internet of Things, University of Rwanda, Kigali P.O. Box 42 85, Rwanda"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6290-1256","authenticated-orcid":false,"given":"Damien","family":"Hanyurwimfura","sequence":"additional","affiliation":[{"name":"African Center of Excellence in Internet of Things, University of Rwanda, Kigali P.O. Box 42 85, Rwanda"}]}],"member":"1968","published-online":{"date-parts":[[2021,4,20]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Grigsby, L.L. (2012). Electric Power Generation, Transmission, and Distribution, CRC Press. [3rd ed.].","DOI":"10.1201\/b12056"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"14","DOI":"10.9790\/1813-0602021418","article-title":"A Study of the Root Causes of High Failure Rate of Distribution Transformer\u2014A Case Study","volume":"6","author":"Ndungu","year":"2017","journal-title":"Int. J. Eng. Sci."},{"key":"ref_3","first-page":"02017","article-title":"The Fault Causes of Overhead Lines in Distribution Network","volume":"61","author":"Wang","year":"2017","journal-title":"Int. Semin. Appl. Phys. Optoelectron. Photonics"},{"key":"ref_4","first-page":"98","article-title":"Power Outages and Production Efficiency of Firms in Africa","volume":"6","author":"Abotsi","year":"2016","journal-title":"Int. J. Energy Econ. Policy"},{"key":"ref_5","first-page":"2411","article-title":"Fault analysis of High Voltage Power","volume":"6","author":"Sarma","year":"2017","journal-title":"Int. J. Adv. Res. Electr. Electron. Instrum. Eng."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"1220","DOI":"10.1016\/j.egypro.2011.12.1079","article-title":"A review of dissolved gas analysis in power transformers","volume":"14","author":"Sun","year":"2012","journal-title":"Energy Procedia"},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Chakravorti, S., Dey, D., and Chatterjee, B. (2013). Recent Trends in the Condition Monitoring of Transformers, Springer.","DOI":"10.1007\/978-1-4471-5550-8"},{"key":"ref_8","first-page":"325","article-title":"Dissolved Gas Analysis based Incipient Fault Diagnosis of Transformer: A Review","volume":"1","author":"Ranjan","year":"2015","journal-title":"Impending Power Demand Innov. Energy Paths"},{"key":"ref_9","unstructured":"Theraja, B.L., and Theraja, A.K. (1999). A Textbook of Electrical Technology, S Chand & Co Ltd."},{"key":"ref_10","first-page":"679","article-title":"Design and fabrication of a Single-phase 1KVA Transformer with automatic cooling system","volume":"5","author":"Turkar","year":"2018","journal-title":"Int. Res. J. Eng. Technol."},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Nickelson, L. (2019). Electromagnetic Theory and Plasmonics for Engineers, Springer.","DOI":"10.1007\/978-981-13-2352-2"},{"key":"ref_12","unstructured":"IEEE-C57.104 (2019). IEEE Guide for the Interpretation of Gases Generated in Mineral Oil-Immersed Transformers, IEEE."},{"key":"ref_13","unstructured":"Hartnack, M. (2018). Rising Power Outage Cost and Frequency Is Driving Grid Modernization Investment. Navig. Res., Available online: https:\/\/www.navigantresearch.com\/news-and-views\/rising-power-outage-cost-and-frequency-is-driving-grid-modernization-investment."},{"key":"ref_14","unstructured":"(2020). Africa Energy Series: Kenya Special Report, Invest in the Energy Sector of Kenya."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Apte, S., Somalwar, R., and Wajirabadkar, A. (2018, January 18\u201321). Incipient Fault Diagnosis of Transformer by DGA Using Fuzzy Logic. Proceedings of the 2018 IEEE International Conference on Power Electronics, Drives and Energy Systems (PEDES), Chennai, India.","DOI":"10.1109\/PEDES.2018.8707928"},{"key":"ref_16","first-page":"2348","article-title":"Condition Monitoring of Power Transformer Using Dissolved Gas Analysis of Mineral Oil: A Review","volume":"3","author":"Soni","year":"2015","journal-title":"Int. J. Adv. Eng. Res. Dev."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"785","DOI":"10.3103\/S1068371214120207","article-title":"Diagnosing power transformers faults","volume":"85","author":"Faiz","year":"2014","journal-title":"Russ. Electr. Eng."},{"key":"ref_18","first-page":"1","article-title":"Transformer Fault Diagnosis Based on DGA using Classical Methods","volume":"4","author":"Bage","year":"2016","journal-title":"Int. J. Eng. Res. Technol."},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Abu-Siada, A. (2019). Improved consistent interpretation approach of fault type within power transformers using dissolved gas analysis and gene expression programming. Energies, 12.","DOI":"10.3390\/en12040730"},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Prasojo, R.A., Gumilang, H., Maulidevi, N.U., and Soedjarno, B.A. (2020). A fuzzy logic model for power transformer faults\u2019 severity determination based on gas level, gas rate, and dissolved gas analysis interpretation. Energies, 13.","DOI":"10.3390\/en13041009"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"390","DOI":"10.1016\/j.isatra.2020.01.016","article-title":"Fuzzy reinforcement learning based intelligent classifier for power transformer faults","volume":"101","author":"Malik","year":"2020","journal-title":"ISA Trans."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"836","DOI":"10.1049\/iet-smt.2018.5397","article-title":"Novel prediction-reliability based graphical DGA technique using multi-layer perceptron network & gas ratio combination algorithm","volume":"13","author":"Chatterjee","year":"2019","journal-title":"IET Sci. Meas. Technol."},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Bustamante, S., Manana, M., Arroyo, A., Castro, P., Laso, A., and Martinez, R. (2019). Dissolved Gas Analysis Equipment for online monitoring of transformer oil: A review. Sensors, 19.","DOI":"10.3390\/s19194057"},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Benmahamed, Y., Kemari, Y., Teguar, M., and Boubakeur, A. (2018, January 1\u20135). Diagnosis of Power Transformer Oil Using KNN and Na\u00efve Bayes Classifiers. Proceedings of the 2018 IEEE 2nd International Conference on Dielectrics ICD, Budapest, Hungary.","DOI":"10.1109\/ICD.2018.8514789"},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"012016","DOI":"10.1088\/1757-899X\/643\/1\/012016","article-title":"K-nearest neighbor method for power transformers condition assessment","volume":"643","author":"Tanfilyeva","year":"2019","journal-title":"IOP Conf. Ser. Mater. Sci. Eng."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Parejo, A., Personal, E., Larios, D.F., Guerrero, J.I., Garc\u00eda, A., and Le\u00f3n, C. (2019). Monitoring and Fault Location Sensor Network for Underground Distribution Lines. Sensors, 19.","DOI":"10.3390\/s19030576"},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Illias, H.A., Chai, X.R., Bakar, A.H.A., and Mokhlis, H. (2015). Transformer incipient fault prediction using combined artificial neural network and various particle swarm optimisation techniques. PLoS ONE, 10.","DOI":"10.1371\/journal.pone.0129363"},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Illias, H.A., and Liang, W.Z. (2018). Identification of transformer fault based on dissolved gas analysis using hybrid support vector machine-modified evolutionary particle swarm optimisation. PLoS ONE, 13.","DOI":"10.1371\/journal.pone.0191366"},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Liu, Y., Song, B., Wang, L., Gao, J., and Xu, R. (2020). Power transformer fault diagnosis based on dissolved gas analysis by correlation coefficient-DBSCAN. Appl. Sci., 10.","DOI":"10.3390\/app10134440"},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Pattanadech, N., Sasomponsawatline, K., Siriworachanyadee, J., and Angsusatra, W. (2019, January 23\u201327). The conformity of DGA interpretation techniques: Experience from transformer 132 units. Proceedings of the 2019 IEEE 20th International Conference on Dielectric Liquids (ICDL), Roma, Italy.","DOI":"10.1109\/ICDL.2019.8796588"},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Borcard, D., Gillet, F., and Legendre, P. (2018). Numerical Ecology with R, Springer International Publishing AG.","DOI":"10.1007\/978-3-319-71404-2"},{"key":"ref_32","unstructured":"Bin, N., and Bakar, A. (2016). A New Technique to Detect Loss of Insulation Life in Power Transformers. [Ph.D. Thesis, Curtin University]."},{"key":"ref_33","first-page":"39","article-title":"A Review of Dissolved Gas","volume":"30","author":"Bakar","year":"2014","journal-title":"Deis Featur. Artic."},{"key":"ref_34","first-page":"36","article-title":"Chromatographic analysis of gases from the transformer","volume":"2","author":"Sisic","year":"2015","journal-title":"Transform. Mag."},{"key":"ref_35","first-page":"296","article-title":"Investigations on power transformer faults based on dissolved gas analysis","volume":"8","author":"Ravichandran","year":"2019","journal-title":"Int. J. Innov. Technol. Explor. Eng."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"8","DOI":"10.1109\/MEI.2002.1014963","article-title":"A review of faults detectable by gas-in-oil analysis in transformers","volume":"Volume 18","author":"Duval","year":"2002","journal-title":"IEEE Electrical Insulation Magazine"},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"27","DOI":"10.1109\/MEI.2010.5599977","article-title":"Intelligent agent-based system using dissolved gas analysis to detect incipient faults in power transformers","volume":"Volume 26","author":"Akbari","year":"2010","journal-title":"IEEE Electrical Insulation Magazine"},{"key":"ref_38","first-page":"1","article-title":"Artificial neural networks applied to DGA for fault diagnosis in oil-filled power transformers","volume":"3","author":"Golkhah","year":"2011","journal-title":"J. Electr. Electron. Eng. Res."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"138","DOI":"10.1002\/widm.1043","article-title":"Data mining for oil-insulated power transformers: An advanced literature survey","volume":"2","author":"Huang","year":"2012","journal-title":"Wiley Interdiscip. Rev. Data Min. Knowl. Discov."},{"key":"ref_40","unstructured":"CIGRE-SC.15 (2006). Recent Developments on the Interpretation of Dissolved Gas Analysis in Transformers, Commission Electrotechnique Internationale."},{"key":"ref_41","unstructured":"IEC-60599 (2015). Mineral Oil-Filled Electrical Equipment in Service\u2013Guidance on the Interpretation of Dissolved and Free Gases Analysis, IEC."},{"key":"ref_42","unstructured":"ASTM-D923-15 (2015). Standard Practices for Sampling Electrical Insulating Liquids, ASTM International."},{"key":"ref_43","unstructured":"ASTM-D3612-02 (2017). Standard Test Method for Analysis of Gases Dissolved in Electrical Insulating Oil by Gas Chromatography, Conseil International des Grands R\u00e9seaux \u00c9lectriques."},{"key":"ref_44","unstructured":"Kumar, A. (2020). Master Data Science and Data Analysis With Pandas, Packt Publishing Ltd."},{"key":"ref_45","unstructured":"Garg, H. (2018). Mastering Exploratory Analysis with Pandas, Packt Publishing Ltd."},{"key":"ref_46","doi-asserted-by":"crossref","unstructured":"Larose, D.T., and Larose, C.D. (2014). Discovering Knowledge in Data: An Introduction to Data Mining, John Wiley & Sons, Inc.","DOI":"10.1002\/9781118874059"},{"key":"ref_47","doi-asserted-by":"crossref","first-page":"182","DOI":"10.1049\/trit.2019.0032","article-title":"Imputing missing values using cumulative linear regression","volume":"4","author":"Mostafa","year":"2019","journal-title":"CAAI Trans. Intell. Technol."},{"key":"ref_48","unstructured":"Niederhut, D. (September, January 28). Safe handling instructions for missing data. Proceedings of the Python in Science Conferences, Trento, Italy."},{"key":"ref_49","first-page":"1","article-title":"From predictive methods to missing data imputation: An optimization approach","volume":"18","author":"Bertsimas","year":"2018","journal-title":"J. Mach. Learn. Res."},{"key":"ref_50","first-page":"271","article-title":"Correlation versus Causation: Another Look at a Common Misinterpretation","volume":"41","author":"Devroop","year":"2000","journal-title":"Alberta J. Educ. Res."},{"key":"ref_51","doi-asserted-by":"crossref","unstructured":"Basuki, A. (2018, January 29\u201331). Online dissolved gas analysis of power transformers based on decision tree model. Proceedings of the 2018 Conference on Power Engineering and Renewable Energy (ICPERE), Solo, Indonesia.","DOI":"10.1109\/ICPERE.2018.8739761"},{"key":"ref_52","first-page":"84","article-title":"Classification trees","volume":"3","author":"Galitskaya","year":"2013","journal-title":"Sotsiologicheskie Issled."},{"key":"ref_53","doi-asserted-by":"crossref","unstructured":"Chiu, S., and Tavella, D. (2008). Introduction to Data Mining, Routledge.","DOI":"10.1016\/B978-0-7506-8234-3.00007-1"},{"key":"ref_54","doi-asserted-by":"crossref","unstructured":"Luo, X., Yu, J.X., and Li, Z. (2014, January 19\u201321). Advanced Data Mining and Applications. Proceedings of the 10th International Conference, ADMA, Guilin, China.","DOI":"10.1007\/978-3-319-14717-8"},{"key":"ref_55","doi-asserted-by":"crossref","first-page":"218","DOI":"10.21037\/atm.2016.03.37","article-title":"Introduction to machine learning: K-nearest neighbors","volume":"4","author":"Zhang","year":"2016","journal-title":"Ann. Transl. Med."},{"key":"ref_56","first-page":"605","article-title":"Application of K-Nearest Neighbor (KNN) Approach for Predicting Economic Events: Theoretical Background","volume":"3","author":"Imandoust","year":"2013","journal-title":"Int. J. Eng. Res. Appl."},{"key":"ref_57","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1155\/2019\/4140707","article-title":"An Improved Random Forest Algorithm for Predicting Employee Turnover","volume":"2019","author":"Gao","year":"2019","journal-title":"Math. Probl. Eng."},{"key":"ref_58","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random Forests","volume":"45","author":"Breiman","year":"2001","journal-title":"Mach. Learn."},{"key":"ref_59","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1155\/2012\/258054","article-title":"Applying randomness effectively based on random forests for classification task of datasets of insufficient information","volume":"2012","author":"Sug","year":"2012","journal-title":"J. Appl. Math."},{"key":"ref_60","first-page":"11","article-title":"A Review on IOT Based Smart Grid","volume":"7","author":"Kaur","year":"2016","journal-title":"Int. J. Energy Inf. Commun."},{"key":"ref_61","doi-asserted-by":"crossref","unstructured":"Bakhtouchi, A. (2018, January 20\u201324). A Tree Decision Based Approach for Selecting Software Development Methodology. Proceedings of the 2018 International Conference on Smart Communications in Network Technologies (SaCoNeT), Nice, France.","DOI":"10.1109\/SaCoNeT.2018.8585699"},{"key":"ref_62","unstructured":"Williams, E. (2019). Python for Data Science, O\u2019Reilly Media, Inc."},{"key":"ref_63","unstructured":"Shaw, Z.A. (2017). Learn Python 3 the Hard Way, Addison-Wesley."},{"key":"ref_64","unstructured":"Morgan, P. (2019). Data Analysis From Scratch with Python, AI Sciences LLC."},{"key":"ref_65","unstructured":"G\u00e9ron, A. (2019). Machine Learning with Scikit-Learn, Keras, and TensorFlow, O\u2019Reilly Media, Inc.. [2nd ed.]."},{"key":"ref_66","doi-asserted-by":"crossref","first-page":"1153","DOI":"10.1109\/COMST.2015.2494502","article-title":"A Survey of Data Mining and Machine Learning Methods for Cyber Security Intrusion Detection","volume":"18","author":"Buczak","year":"2016","journal-title":"IEEE Commun. Surv. Tutor."},{"key":"ref_67","doi-asserted-by":"crossref","first-page":"772","DOI":"10.1016\/j.ins.2019.06.064","article-title":"Three-way confusion matrix for classification: A measure driven view","volume":"507","author":"Xu","year":"2020","journal-title":"Inf. Sci."},{"key":"ref_68","unstructured":"Beauxis-Aussalet, E., and Hardman, L. (2014, January 6\u20137). Simplifying the visualization of confusion matrix. Proceedings of the Belgian\/Netherlands Artificial Intelligence Conference, Belgian, The Netherlands."},{"key":"ref_69","unstructured":"Visa, S., Ramsay, B., Ralescu, A., and van der Knaap, E. (2011, January 16\u201317). Confusion Matrix-based Feature Selection. Proceedings of the 22nd Midwest Artificial Intelligence and Cognitive Science, Cincinnati, OH, USA."},{"key":"ref_70","first-page":"2586","article-title":"High Impedance Fault Detection and Location in a Power Transmission Line Using ZIGBEE","volume":"5","author":"Gopinath","year":"2016","journal-title":"Int. J. Innov. Res. Sci. Eng. Technol."},{"key":"ref_71","first-page":"59","article-title":"Internet of Things, Challenges for Demand Side Management","volume":"21","author":"Oprea","year":"2017","journal-title":"Inform. Econ."},{"key":"ref_72","doi-asserted-by":"crossref","unstructured":"Bikmetov, R., Raja, M.Y.A., and Sane, T.U. (2017, January 17\u201319). Infrastructure and applications of Internet of Things in smart grids: A survey. Proceedings of the 2017 North American Power Symposium (NAPS), Morgantown, West Virginia.","DOI":"10.1109\/NAPS.2017.8107283"},{"key":"ref_73","first-page":"461","article-title":"Evaluation Measures of the Classification Performance of Imbalanced Data Sets","volume":"51","author":"Gu","year":"2009","journal-title":"Commun. Comput. Inf. Sci."},{"key":"ref_74","doi-asserted-by":"crossref","first-page":"1","DOI":"10.5121\/ijdkp.2015.5201","article-title":"Review on Evaluation Metrics for Data Classification Evaluations","volume":"5","author":"Hossin","year":"2015","journal-title":"Int. J. Data Min. Knowl. Manag. Process."},{"key":"ref_75","first-page":"132","article-title":"Assessing the quality of classification models: Performance measures and evaluation procedures","volume":"1","author":"Cichosz","year":"2011","journal-title":"Cent. Eur. J. Eng."},{"key":"ref_76","unstructured":"Staeheli, L.A., and Mitchell, D. (2006, January 25\u201329). The Relationship between Precision-Recall and ROC Curves. Proceedings of the 23rd International Conference on Machine Learning, Pittsburgh, PA, USA."},{"key":"ref_77","unstructured":"Flach, P. (February, January 27). Performance Evaluation in Machine Learning: The Good, the Bad, the Ugly, and the Way Forward. Proceedings of the AAAI Conference on Artificial Intelligence, Honolulu, HI, USA."}],"container-title":["Algorithms"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1999-4893\/14\/4\/128\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T05:50:23Z","timestamp":1760161823000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1999-4893\/14\/4\/128"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,4,20]]},"references-count":77,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2021,4]]}},"alternative-id":["a14040128"],"URL":"https:\/\/doi.org\/10.3390\/a14040128","relation":{},"ISSN":["1999-4893"],"issn-type":[{"value":"1999-4893","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,4,20]]}}}