In this paper, a new testing method for the diagnose of winding faults based on the resonant frequency of the transformer is proposed. The system which consists of the high-voltage DC power and high-voltage IGBT could charge the transformer periodically. The response measured from the bushing of the winding is the oscillating wave that can be used for evaluating the status. Carrying out the new test on 220kV three winding, three-phase transformer in the field, the tests of insulation and the winding have been completed in one test. The calculated frequency verifies the feasibility and applicability of the method in theory and the voltage does not affect the resonant frequency in the test. The three winding, three-phase model transformer was built to simulate the various parameters of the power supply, which proves that the features of the oscillating waves are brought about by the properties of the transformer. And that, the new testing method provide a new opportunity to do parameter identification of axial displacement and short circuit (SC) fault. The above research shows that the proposed method can conduct the test of winding and insulation in one test for a short time, which proves a reference in the field.

Oil-immersed transformer is one of the most important electrical equipment in power distribution and transmission systems. Dielectric response method is a well-recognized method to diagnose the insulation defect of oil-immersed transformers. However, the applicability of this method is restricted due to the long testing time. Under some special field conditions, the method is not even applicable. In this paper, a novel testing method is proposed based on the following ideas: 1) first, the low frequency dielectric parameters are extracted by using mixing frequency excitation; 2) then, parameters of the extended Debye equivalent circuit are determined based on cuckoo search (CS) optimization algorithm; 3) finally, the specific parameters are used to the established simulation model and obtain the recovery voltage curve. Compared with the traditional method, the testing time of the proposed method has been greatly reduced. Besides, dielectric parameters in both frequency domain and time domain can be obtained simultaneously. The applicability of the proposed method is verified by the dielectric response measurements on a laboratory transformer and a real power transformer in a substation.

Frequency response analysis (FRA) is an effective approach for detecting mechanical damage in the windings of a transformer. However, the use of FRA to estimate the winding condition of autotransformers (ATs) employed in China High-Speed Railway is problematic because the split windings of the ATs are connected inside the tank, such that it is impossible to measure the FRA curves for each split winding outside the tank directly. Therefore, it is difficult to identify fault winding and effectively diagnose the overall condition of the windings. This paper explores the features of FRA curves under different fault conditions to help distinguish fault winding. A verified distributed circuit model with turn unit is built based on split windings, and the electrical parameters of the windings are determined using finite element method. The model is then employed to simulate typical axial displacement faults. The results show that the first anti-resonance and resonance of the frequency response for connections involving series windings or series and common windings are visible features for diagnosing winding faults and the fault level. Specifically, the first and second resonances of the frequency response for the common winding connection can be regarded as the primary feature for identifying fault winding.

Autotransformer (AT) is the most core power supply equipment, overvoltage and short circuit fault may lead to winding deformation, which will have a negative impact on its insulation and even affect the operation of train. Frequency response analysis (FRA) is widely used for detecting winding faults in a transformer. However, the direct measure of FRA for each split winding fails, because the split windings are adopted to satisfy the impedance requirement of high-speed railway, where the windings are connected inside the tank. A novel fault interpretation method based on image features and binary tree support vector machine is proposed, which can get the condition of three windings in one measurement. Winding faults caused by different windings are simulated including short circuit defect, axial deformation and series capacitance variation, and the FRA curves are measured under various faults. Then the features of the gray-level gradient co-occurrence matrix and the gray-level difference statistics are got from the polar plot of FRA. Finally, the image features are used as the inputs to binary tree support vector machine for fault type and faulty winding classification. The results show that the proposed method has high accuracy for identifying fault type and faulty winding in AT.