利用小波包能量区分变压器内外部故障

摘要

TA saturation is one of the key issues to be solved in transformer differential protection.Deep analysis and simulation study of transformer TA saturation showed that when TA saturates, external fault currents contain a great amount of point mutations, while the internal fault current is continuous.According to the conclusion, we propose a new method for distinguishing transformer internal and external fault based on wavelet packet transform.Firstly, wave-let packet transform respectively decomposes internal and external fault current with TA saturation, then calculates en-ergy of decomposed signals.The difference among signals is quantified and can be observed visually, and finally puts forward the criterion.Simulation results show that the method can distinguish internal and external fault current under the condition of TA saturation and timely prevent possible damage, as well as can avoid transformer differential protec-tion mal-operation.%TA饱和是变压器差动保护中待解决的关键问题之一。通过对变压器TA饱和的深入分析及仿真研究，得出结论：TA饱和时，区外故障电流含有大量明显的突变点，而区内故障电流连续。据此结论，文章提出了一种利用小波包变换区分变压器内外部故障的新方法。首先小波包变换分别对TA饱和情况下的内外部故障电流进行分解，然后对分解后的信号进行能量计算，将问题量化，可以直观地观察信号的差异，最后提出判据。仿真验证证明，该方法能够正确区分TA饱和情况下的内外部故障电流，能及时阻止可能造成的损害，并可防止变压器保护误动。