PD measurements on a DC gas-insulated transmission line (DC GIL) conducted in the frame of the Prototype Installation Test according to recommendation of CIGRE JWG D1/B3.57

There is a growing demand for DC long distance transmission, as the power generation increasingly takes place far away from the load centres. In many countries the transmissionlines in question have to be installed underground, due to environmental concernsandpolitical decisions. Besides DC cables DC gas-insulatedlines are an option. To gain long-term experience with this type of technology, CIGRE JWG D1/B3.57 suggests a PrototypeInstallation Test with a test duration of one year, by which the dielectric DC and overvoltage stress under real service conditions are reproduced and investigated. The gas-insulated line considered in this paper is designed for a rated DC voltage of #550 kV and a current carrying capability of 5000 A. It is installed at a HVDCtest facility at the Technical University of Darmstadt. For a long-term test with sufficient significance the insulating conditions of the test arrangement before and during long-term testing have to be monitored to observe the insulation performance. For this purpose an ultra-high-frequency partial discharge (UHF PD) monitoring system is installed, which is adapted for recording of PD under DC voltage stress. This system shall also be proven for condition monitoring of future DC GIL installations. With this regard, the UHF PD monitoring system installed and the method to detect and identify possible PD defects under DC voltage will be discussed. The measuring sensitivity of the installed UHF monitoring system was tested according to the recommendations given in CIGRE TB 654. As the DC GIL under considerationis a new design, sensitivity verification step 1 had to be performedfirst on a typical test assembly. Thesensitivity verification step 2 was carried out within the frame of the commissioningtests of the test arrangement. The UHF PDsignalattenuation and the sensitivity obtained between two adjacent sensors will be given. By means of the attenuation profile the sensor distance for future DC GIL installations can be estimated. During commissioning of the 100 m DC GIL prototype with DC test voltage a probable PD defect was detected by the monitoring system. Time of flight measurements were carried out in order to localise the defect. As no phase correlation of the PD pulses exists at DC voltage, pulse sequence analysis (PSA) had to be applied for PD defect identification. For this purpose investigations on the 100 m long DC GILinstallation were performed using PSA methods at rated voltages. Based on pre-investigations of typical PD defects under DC voltage, briefly given in this report, the PD defect detected onsite could be identified by PSA analysis. The analysis of the measured PD signal showed a good agreement with one reference signal analysis gained in the preinvestigations.