变压器短路耐受能力的理论评估及绕组累积变形识别方法

As power system capacity increases, power transformers frequently endure multiple short-circuit impulses of varying intensities. Existing evaluation methods cannot accurately predict their cumulative effects, which hinders effective operation and maintenance. In this study, a double-leg, three-winding mock-up transformer has been built up. Theoretical short-circuit withstand capability is evaluated using WELDINST software and methods recommended by Chinese and Japanese transformers technical committee. Short-circuit impulse experiments with varying current ratios are performed to investigate cumulative effects under radial and axial instability. The results show that the prediction accuracy varies across different methods and instability modes. The Chinese transformer technical committee recommended method performs better under axial instability, while the WELDINST method excels excellent under radial instability. The experimental results showing average deviations of 19.75% and 8.92%, respectively. The experiment show that significant cumulative effects appear when the current ratio exceeds 70%. Furthermore, an exponential function was established between impedance cumulative effects and current ratios. Finally, the steady-state vibration main fundamental ratio and vibration entropy, enable real-time identification of the winding state, showing a relationship with changes in winding impedance. Larger impedance variations are associated with lower steady-state vibration main fundamental ratio and higher vibration entropies.