Research And Realization Of Wavelet Algorithm For Transformer Protection Device

All along, differential protection is used as the main protection for transformer protection device. Discrimination for inrush current is the core problem of transformer differential protection. Differential protection can distingusih internal faults and external faults very reliably, but often causes mal-operation during inrush current. The traditional methods of discrimination between inrush current and internal fault current exist certain shortcomings on theory completeness and computer simulation, which can not completely meet the requirement of transformer microcomputer protection. Wavelet transform has the characteristics of time-frequency localization and multi-resolution analysis, it is particularly well adapted to process the break signals and to extract characteristics of these signals. Wavelet modulus maximum can measure dead-angle of inrush current, and identify inrush current from internal fault current correctly, then transformer protection device can brake reliably during inrush current. Therefore, the research and design of transformer protection device based on wavelet algorithm has great academic significance and practical value.In this paper, the various familiar algorithms for discrimination between inrush current and internal fault current are summed up and compared firstly, the merits and demerits of these algorithms are also analyzed, wavelet modulus maximum is used in order to detect chop point of inrush current covering several parties include theory completeness and hardware implementation. After that by use of Power System Blockset in MATLAB, the simulation models of inrush current and internal fault current are established, which provide raw data for wavelet algorithm design. Inrush current and internal fault current are three-scale decomposed by db5 wavelet, and a new algorithm of discrimination between inrush current and internal fault current of transformer based on wavelet transform is proposed and proved. Based on this, the implementation idea of protection algorithm is determined through FIR filter, the application program is developed by using assemble language and C language, DSP implementation of differential protection algorithm based on wavelet transform is successful finally through compile, link and debug.