Recognition of dissolved gas in transformer oil by ant colony optimization support vector machine

Dissolved gas analysis (DGA) has been approved as an accurate method to assess the insulation condition of power transformer. The ant colony optimization algorithm is competent to obtain the global optimum values of the support vector machine (SVM). This paper presents an approach of ant colony optimization SVM to recognize histograms of characteristic dissolved gas in transformer oil. DGA data of transformer oil with normal operation and four types of typical power transformer faults were selected, and characteristic dissolved gas were used to establish histograms. Statistical characteristic parameters involving mean, skewness and kurtosis, etc. were calculated as recognition features. Three approaches including IEC three ratio method, SVM and ant colony optimization SVM were used to diagnosis DGA of power transformer. Results show that the proposed approach can effectively improve the accuracy of the recognition for characteristic dissolved gas histograms compared with the other two methods.