Load Protection Detection Method for 10 kV Furnace Transformer in 35 kV Substation Based on Wavelet Transform
Too high or too low load may lead to instability of transformer for 10 kV electric furnace,and even lead to system collapse or equip-ment failure.Therefore,a detection method of transformer load protection for 10 kV electric furnace in 35 kV substation based on wavelet transform is proposed.Through empirical wavelet and wavelet transform,the load data of electric furnace transformer with noise is denoised and decomposed into multi-frequency modal functions.Based on kurtosis value,the scale domain of denoised modal functions is divided to re-construct the load data of electric furnace transformer.Through KD-Tree algorithm,the load data of electric furnace transformer is divided into several evenly distributed partitions,and the reachable distance is calculated to set the threshold of electric furnace transformer load determina-tion.The local cluster association relation set is established,and the clustering results of electric furnace transformer load calculation nodes are merged and processed,and the global abnormal sample point set is divided to complete the abnormal detection of electric furnace trans-former load data.The experimental results show that the proposed method has a good denoising effect,and the recall and precision of abnormal load of electric furnace transformer are kept above 95％,and the detection time is kept within 600ms,which can effectively and accurately de-tect the load protection of electric furnace transformer.
electric furnace transformerKD-Tree algorithmwavelet transformmodal functionreachable distance
万方数据
