Research on Feature Parameter Extraction and Health Analysis Method of UHV DC Converter Valve

Abstract

The converter valve is the core equipment of UHV DC transmission. The failure of converter valve will cause the outage of the DC transmission, which will cause serious economic losses and even cause major safety accidents. However, the current technology can only make logical judgments on whether the converter valve is in a normal state and cannot judge the health status of the converter valve; and then carry out maintenance and replacement plans for the key components of the converter valve according to the health status. To meet the technical requirements for monitoring the health status of converter valves, the method is carried out on the health status of converter valves for UHVDC transmission. Firstly, a multi-source heterogeneous data acquisition program for UHVDC converter valve status parameters is proposed. After that, a data feature extraction program is invented; combined with the data feature extraction method, a single index and multi-index converter valve health status assessment is carried out. Finally, a feature extraction system for high-voltage DC converter valves was built in the laboratory to verify the effectiveness of the feature extraction function. The test results show that the feature parameters and health status analysis proposed in this paper can effectively extract the features of the converter valve and complete it and analysis the health status of the converter valve.