The UHVDC Transmission Line Transient Protection And Lightning Disturbance Identification Research

UHVDC transmission line have the long distance and in the certain condition of the natural environment is complex. The above characteristic not only greatly increases the probability of line short-circuit and lightning but seriously threatening the safe operation of UHVDC system, lightning is the main factor of line fault in the usually. For this, the UHVDC transmission line protection and lightning disturbance identification method is imperative. Based on yun-guang UHVDC transmission system model, the ±800kV UHVDC transmission system model is established. The lightning interference and short-circuit in the time domain and the transient energy characteristics were analyzed and put forward the lightning disturbance identification criterion. The frequency characteristics of the border of UHVDC transmission lines and lines were analyzed, then the transmission line transient protection criterion is proposed. In this paper, the main research work includes:(1) The research of lightning disturbance identification based on the morphology for UHVDC transmission line.This paper used mathematical morphology to analysis ±800 kV UHVDC transmission lines. The multi-scale morphology decomposition was used to decompose the transient current of the non-fault lightning stroke, fault lightning stroke and line short circuit. Extract the spectrum energy from the high and the low frequency bands, is to identify the lightning disturbance. Further according the first head of current decomposition waveform of maximum amplitude ratio in which the second scale and sixth scale, to distinguish the fault lightning stroke and the line short circuit. Through extensive simulations demonstrate the approach correct and effective.(2) The research of lightning disturbance identification based on grey correlation degree-morphology spectrum entropy.The multi-scale morphological decomposition was performed for the different fault types, its power spectral entropy and singular spectral entropy in different scale domains were computed. The two indicators were taken as character-vector. Then, the standard fault mode matrix was built. The grey relational analysis between the character-vector and standard fault mode matrix of the sample signal to be detected was conducted to distinguish the different fault types according to the value of grey relevancy.(3) The research of single-ended current direction transient based protection for UHVDC transmission line.The method of single-ended current direction transient based protection used morphology for UHVDC transmission line is proposed. First, the fault occurred in the rectifier side or contrary side can be judged by the electric wave direction of the fault. Second, the fault occurred in the inside or outside of contrary can be judged by the decay characteristics of the line boundary and line to the high frequency signal. The transient fault signal detected by the lateral protection components which can be used and combining multi-resolution morphological gradient (MMG) to extract the polarity of current signal, then the current direction element principle and criterion is proposed. The transient signals which detected by rectifier side protection installation were did spectrum calculations, normalized and transform it to the frequency domain, the characteristics in the frequency domain were compared to judgment the fault is inside or outside of contralateral. The simulation results show that the approach is proposed can distinguish the rectifier side fault, the inside fault and the contrary side fault accurately, the line protection of UHVDC transmission also can be realized.(4) The research of fault intelligent identification method based on singular spectrum entropy and support vector machine for UHVDC transmission line.The wavelet transform of multi-scale decomposition was adopted to obtain the faults information detected from the protection device. Feature vector which was composed of the singular spectrum entropy of each layer was obtained and divided into training set and testing set. The former was trained to get the parameters of the SVM classifier and the latter was to be tested. The prediction was the fault classification in different position. Simulations show that the transient protection principle based on multi-resolution singular spectrum entropy and support vector machine (SVM) for UHVDC transmission line can reliably identify the external faults in rectifier side, the internal faults and the external faults in the inverter sideThe simulations demonstrate the above proposed approach correct and effective.