Quantitative evaluation method for severity of transformer discharge faults based on weighted carbon–hydrogen ratio of characteristic gases

Abstract Dissolved gas analysis (DGA) has been widely used for the condition diagnosis of oil-filled equipment. However, it mainly focuses on analyzing gas generation characteristics and diagnosing fault types, making it challenging to quantitatively assess the severity of faults. In this study, a discharge test platform was established to simulate discharge faults under different intensities. The content and relative percentage of dissolved gases in oil at varying discharge intensities were compared and studied. Experimental results showed that as the discharge intensity increased, the proportion of hydrocarbon gases followed the order: CH 4 → C 2 H 6 → C 2 H 4 → C 2 H 2 , with the carbon-to-hydrogen ratio of the corresponding characteristic gases also increasing. Based on this, a quantitative evaluation method for the severity of transformer discharge faults was proposed, using the weighted carbon-hydrogen ratio (R CH ) of characteristic gases. Case studies demonstrate that, compared with traditional DGA methods, this new method can quantitatively assess fault severity and overcome the limitations of the three-ratio method, such as discontinuity and missing codes, offering higher accuracy and sensitivity.