A Multinomial DGA Classifier for Incipient Fault Detection in Oil-Impregnated Power Transformers

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, Naïve Bayes, and Gradient Boost. Investigative results show that KosaNet demonstrated an improved DGA classification ability particularly when classifying multinomial data.

[1]  Iuliana Botha,et al.  Internet of Things, Challenges for Demand Side Management , 2017 .

[2]  Chao-Ming Huang,et al.  A Review of Dissolved Gas Analysis in Power Transformers , 2012 .

[3]  Sheetal Kalra,et al.  A Review on IOT Based Smart Grid , 2016 .

[4]  Zhihua Cai,et al.  Evaluation Measures of the Classification Performance of Imbalanced Data Sets , 2009 .

[6]  Michel Duval,et al.  A review of faults detectable by gas-in-oil analysis in transformers , 2002 .

[7]  David Rosenberg,et al.  Classification Trees , 2009, Encyclopedia of Database Systems.

[8]  Ahmed Abu-Siada,et al.  Improved Consistent Interpretation Approach of Fault Type within Power Transformers Using Dissolved Gas Analysis and Gene Expression Programming , 2019, Energies.

[9]  Hyontai Sug,et al.  Applying Randomness Effectively Based on Random Forests for Classification Task of Datasets of Insufficient Information , 2012, J. Appl. Math..

[10]  Deepak,et al.  Transformer Fault Diagnosis Based on DGA using Classical Methods , 2018 .

[11]  Suwarno,et al.  A Fuzzy Logic Model for Power Transformer Faults’ Severity Determination Based on Gas Level, Gas Rate, and Dissolved Gas Analysis Interpretation , 2020 .

[12]  S. Imandoust,et al.  Application of K-Nearest Neighbor (KNN) Approach for Predicting Economic Events: Theoretical Background , 2013 .

[13]  Zhongheng Zhang,et al.  Introduction to machine learning: k-nearest neighbors. , 2016, Annals of translational medicine.

[14]  Paweł Cichosz,et al.  Assessing the quality of classification models: Performance measures and evaluation procedures , 2011 .

[15]  Carlos León,et al.  Monitoring and Fault Location Sensor Network for Underground Distribution Lines , 2019, Sensors.

[16]  C. Ndungu,et al.  A Study of the Root Causes of High Failure Rate of Distribution Transformer - A Case Study , 2017 .

[17]  Mario Manana,et al.  Dissolved Gas Analysis Equipment for Online Monitoring of Transformer Oil: A Review , 2019, Sensors.

[18]  Hasmat Malik,et al.  Fuzzy reinforcement learning based intelligent classifier for power transformer faults. , 2020, ISA transactions.

[19]  Subham Dawn,et al.  Novel prediction‐reliability based graphical DGA technique using multi‐layer perceptron network & gas ratio combination algorithm , 2019, IET Science, Measurement & Technology.

[20]  Dimitris Bertsimas,et al.  From Predictive Methods to Missing Data Imputation: An Optimization Approach , 2017, J. Mach. Learn. Res..

[21]  Duoqian Miao,et al.  Three-way confusion matrix for classification: A measure driven view , 2020, Inf. Sci..

[22]  E Gockenbach,et al.  Intelligent agent-based system using dissolved gas analysis to detect incipient faults in power transformers , 2010, IEEE Electrical Insulation Magazine.

[23]  Li Wang,et al.  The Fault Causes of Overhead Lines in Distribution Network , 2016 .

[24]  M. N. Sulaiman,et al.  A Review On Evaluation Metrics For Data Classification Evaluations , 2015 .

[25]  Mohammad Golkhah,et al.  Artificial neural networks applied to DGA for fault diagnosis in oil-filled power transformers , 2011 .

[26]  Chao-Ming Huang,et al.  Data mining for oil‐insulated power transformers: an advanced literature survey , 2012, Wiley Interdiscip. Rev. Data Min. Knowl. Discov..

[27]  A. Abotsi.  Power Outages and Production Efficiency of Firms in Africa , 2016 .

[28]  Samih M. Mostafa,et al.  Imputing missing values using cumulative linear regression , 2019, CAAI Trans. Intell. Technol..

[29]  A. H. Abu Bakar,et al.  Transformer Incipient Fault Prediction Using Combined Artificial Neural Network and Various Particle Swarm Optimisation Techniques , 2015, PloS one.

[30]  Junhao Wen,et al.  An Improved Random Forest Algorithm for Predicting Employee Turnover , 2019, Mathematical Problems in Engineering.

[31]  Hazlee Azil Illias,et al.  Identification of transformer fault based on dissolved gas analysis using hybrid support vector machine-modified evolutionary particle swarm optimisation , 2018, PloS one.

[32]  Linong Wang,et al.  Power Transformer Fault Diagnosis Based on Dissolved Gas Analysis by Correlation Coefficient-DBSCAN , 2020, Applied Sciences.

[33]  Karendra Devroop.  Correlation versus Causation: Another Look at a Common Misinterpretation. , 2000 .