Deterioration Rule And Diagnostic Method For Partial Discharge In Transformer Oil-pressboard Insulation

Partial discharge (PD) is one of the main reasons for transformer oil-pressboard insulation deterioration. The research about deterioration rule and diagnostic method for PD in transformer oil-pressboard insulation has theoretical significance and practical value for promoting the evaluation reliability of transformer insulation performance. Based on the phenomenon of pressboard aging during the operation of transformer, this paper analyses the deterioration rule for PD in transformer oil-pressboard insulation under the influence of pressboard aging. Besides, the PD signal interference suppression method and pattern recognition method for transformer oil-pressboard insulation is raised. What’s more, the purpose for judging the PD development degree for transformer oil-pressboard insulation under the influence of pressboard aging is achieved. The main contents are as follows:The surface discharge and point discharge model for transformer oil-pressboard insulation was designed, and the pressboard samples with different aging degrees were got by heat aging. Constant voltage was applied to transformer oil-pressboard insulation typical defeat test samples for long time experimental process. According to different discharge phenomena, the transformer oil-pressboard insulation PD development degree can be divided into three stages, namely initial discharge stage, development discharge stage and dangerous discharge stage. The total discharge quantity, the total discharge number and the maximum discharge quantity for oil-pressboard insulation with different aging degrees of pressboard were analyzed. The results show that the pressboard aging has no obvious impact on the PD development process at the initial discharge stage. However, at the development discharge stage and dangerous discharge stage, the pressboard aging has great impact on the PD development process. By the microscopic test and simulation analysis, it point out that the effect is caused by the voids due to the aging of the pressboard.A transformer oil-pressboard insulation PD signal denoising method based on sparse decomposition was proposed. The PD signal correlated overcomplete dictionary were designed, whose atoms were correlated to the original PD pulse signal while uncorrelated or weak-correlated to noise. Denoising the noisy PD signal by sparse decomposition based on matching pursuit (MP) algorithm, only the original PD signal can be expressed by the best atoms, thus the goal of denoising was achieved. Besides, the searching progress of best atoms was accelerated by improved quantum-behaved particle swarm optimization (IQPSO). The denoising method presented in this article was applied to the simulated and laboratory measured PD signals. The results show that the denoising method is superior to traditional wavelet methods, which has less amplitude error as well as less waveform destination.A transformer oil-pressboard insulation PD pattern recognition method based on sparse decomposition was presented. The PD statistical overcomplete dictionary was built by statistical vectors extracted from each training sample signals. Nonlinear mapping was conducted to this dictionary, and the nonlinear PD statistical overcomplete dictionary can be obtained. Decomposing the nonlinear PD statistical vector extracted from the PD signal to be recognized, this vector can only be sparse represented by the atoms extracted from the corresponding pattern sub-dictionary. Thus, the goal of PD pattern recognition can be achieved. Besides, a kernel improved matching pursuit algorithm (KIMP) was raised to obtain sparse decomposition result without accurate form of the nonlinear mapping. The kernel function and its parameters can be determined based on similarity measuring coefficient. PD signals were measured in two different experimental environments by different artificial discharge models, which were used as training samples and testing samples respectively. The recognition results are obtained by the presented method. The recognition results based on neural network, K nearest neighbor (KNN) and support vector machine (SVM) were obtained for comparison. The experimental results show that the presented method has a higher accuracy and more stable recognition result.A transformer oil-pressboard insulation PD development degree judgment method considering the influence of pressboared aging was raised. The Statistical parameters for PD development judgment is evaluated and selected for the pressboard with different aging degrees based on sparse reconstitution, and the oil-pressboard insulation PD development judging overcomplete dictionary was built as well. A transformer oil-pressboard insulation PD development degree judgment method based on weighted voting sparse decomposition was presented, which can judge the PD development degree considering the influence of pressboard aging accurately without predicting the aging degree of pressboard.