BN
B.K.Y. Naduvinamani
info
Please Note
<p>This page displays the records of the person named above and is not linked to a unique person identifier. This record may need to be merged to a profile.</p>
1 records found
1
A Reduced Scale Test Bench for Investigating DC Spacer Flashover under Thermal Gradients
Design, Simulation, and Experimental Method
High-voltage direct current gas-insulated switchgear (HVDC GIS) uses a grounded enclosure and gas insulation to provide compact high-voltage equipment for DC transmission. Solid spacers support the high-voltage conductor and form the gas–solid insulation interface. Under DC conditions, the electric-field distribution is governed mainly by the electrical conductivity of the insulation materials. The conductivity of epoxy is strongly temperature dependent. A temperature gradient across the spacer therefore changes the conductivity distribution and redistributes the electric field. The field also evolves during DC pre-stress as charge redistributes. This motivates investigation of how temperature gradient, gradient direction, voltage polarity, and DC pre-stress duration affect surface-flashover behaviour.
This thesis develops and numerically assesses a reduced-scale gas–epoxy test bench for future DC surface-flashover measurements. The design operates in atmospheric air near 1 bar with a maximum operating voltage of 50 kV DC. Either electrode can be heated to impose a temperature difference of 40–50 K without changing the field-defining electrode–spacer geometry. Two heating concepts were developed, together with a transparent chamber, diagnostic access, qualification requirements, and a test method. The electrode–spacer geometry was numerically verified against a reduced-scale coaxial GIS reference. The scientific contribution is an integrated methodology that combines the common test geometry, controlled thermal and electrical conditions, time-dependent electrothermal calculation, and a defined basis for future flashover measurements.
At 60 kV, the final electrode–spacer geometry reproduces the coaxial reference with differences of 0.8% in representative peak interface field and 2.2% in mean interface field. For the hot-oil numerical screening case with the HV electrode heated and the grounded electrode colder, the thermal gradient decreases the immediate hot-HV-edge field by about 13%, while the central minimum and grounded-side regional maximum increase by about 38% and 30%, respectively. The HV-to-GND regional-maximum ratio decreases from 1.77 to 1.34, showing a substantial redistribution of the DC interface field while the largest regional field remains on the HV side.
The time-dependent calculations further show that DC field formation is strongly position dependent. Between 1 and 12 h, the interval-based apparent relaxation times are about 3.2 h on the hot HV side and 15.8 h on the colder grounded side. The HV-side field is close to its stationary value after 12 h, whereas the grounded-side deviation remains about 3.9% after 24 h. Together with the non-monotonic central response, these results show that one common relaxation time cannot represent DC field formation along the complete spacer interface. Thermal condition and DC pre-stress duration must therefore be controlled in future flashover measurements. The developed test bench and test method provide the basis for the experimental investigation of RQ1–RQ3 after manufacture and physical qualification. ...
This thesis develops and numerically assesses a reduced-scale gas–epoxy test bench for future DC surface-flashover measurements. The design operates in atmospheric air near 1 bar with a maximum operating voltage of 50 kV DC. Either electrode can be heated to impose a temperature difference of 40–50 K without changing the field-defining electrode–spacer geometry. Two heating concepts were developed, together with a transparent chamber, diagnostic access, qualification requirements, and a test method. The electrode–spacer geometry was numerically verified against a reduced-scale coaxial GIS reference. The scientific contribution is an integrated methodology that combines the common test geometry, controlled thermal and electrical conditions, time-dependent electrothermal calculation, and a defined basis for future flashover measurements.
At 60 kV, the final electrode–spacer geometry reproduces the coaxial reference with differences of 0.8% in representative peak interface field and 2.2% in mean interface field. For the hot-oil numerical screening case with the HV electrode heated and the grounded electrode colder, the thermal gradient decreases the immediate hot-HV-edge field by about 13%, while the central minimum and grounded-side regional maximum increase by about 38% and 30%, respectively. The HV-to-GND regional-maximum ratio decreases from 1.77 to 1.34, showing a substantial redistribution of the DC interface field while the largest regional field remains on the HV side.
The time-dependent calculations further show that DC field formation is strongly position dependent. Between 1 and 12 h, the interval-based apparent relaxation times are about 3.2 h on the hot HV side and 15.8 h on the colder grounded side. The HV-side field is close to its stationary value after 12 h, whereas the grounded-side deviation remains about 3.9% after 24 h. Together with the non-monotonic central response, these results show that one common relaxation time cannot represent DC field formation along the complete spacer interface. Thermal condition and DC pre-stress duration must therefore be controlled in future flashover measurements. The developed test bench and test method provide the basis for the experimental investigation of RQ1–RQ3 after manufacture and physical qualification. ...
High-voltage direct current gas-insulated switchgear (HVDC GIS) uses a grounded enclosure and gas insulation to provide compact high-voltage equipment for DC transmission. Solid spacers support the high-voltage conductor and form the gas–solid insulation interface. Under DC conditions, the electric-field distribution is governed mainly by the electrical conductivity of the insulation materials. The conductivity of epoxy is strongly temperature dependent. A temperature gradient across the spacer therefore changes the conductivity distribution and redistributes the electric field. The field also evolves during DC pre-stress as charge redistributes. This motivates investigation of how temperature gradient, gradient direction, voltage polarity, and DC pre-stress duration affect surface-flashover behaviour.
This thesis develops and numerically assesses a reduced-scale gas–epoxy test bench for future DC surface-flashover measurements. The design operates in atmospheric air near 1 bar with a maximum operating voltage of 50 kV DC. Either electrode can be heated to impose a temperature difference of 40–50 K without changing the field-defining electrode–spacer geometry. Two heating concepts were developed, together with a transparent chamber, diagnostic access, qualification requirements, and a test method. The electrode–spacer geometry was numerically verified against a reduced-scale coaxial GIS reference. The scientific contribution is an integrated methodology that combines the common test geometry, controlled thermal and electrical conditions, time-dependent electrothermal calculation, and a defined basis for future flashover measurements.
At 60 kV, the final electrode–spacer geometry reproduces the coaxial reference with differences of 0.8% in representative peak interface field and 2.2% in mean interface field. For the hot-oil numerical screening case with the HV electrode heated and the grounded electrode colder, the thermal gradient decreases the immediate hot-HV-edge field by about 13%, while the central minimum and grounded-side regional maximum increase by about 38% and 30%, respectively. The HV-to-GND regional-maximum ratio decreases from 1.77 to 1.34, showing a substantial redistribution of the DC interface field while the largest regional field remains on the HV side.
The time-dependent calculations further show that DC field formation is strongly position dependent. Between 1 and 12 h, the interval-based apparent relaxation times are about 3.2 h on the hot HV side and 15.8 h on the colder grounded side. The HV-side field is close to its stationary value after 12 h, whereas the grounded-side deviation remains about 3.9% after 24 h. Together with the non-monotonic central response, these results show that one common relaxation time cannot represent DC field formation along the complete spacer interface. Thermal condition and DC pre-stress duration must therefore be controlled in future flashover measurements. The developed test bench and test method provide the basis for the experimental investigation of RQ1–RQ3 after manufacture and physical qualification.
This thesis develops and numerically assesses a reduced-scale gas–epoxy test bench for future DC surface-flashover measurements. The design operates in atmospheric air near 1 bar with a maximum operating voltage of 50 kV DC. Either electrode can be heated to impose a temperature difference of 40–50 K without changing the field-defining electrode–spacer geometry. Two heating concepts were developed, together with a transparent chamber, diagnostic access, qualification requirements, and a test method. The electrode–spacer geometry was numerically verified against a reduced-scale coaxial GIS reference. The scientific contribution is an integrated methodology that combines the common test geometry, controlled thermal and electrical conditions, time-dependent electrothermal calculation, and a defined basis for future flashover measurements.
At 60 kV, the final electrode–spacer geometry reproduces the coaxial reference with differences of 0.8% in representative peak interface field and 2.2% in mean interface field. For the hot-oil numerical screening case with the HV electrode heated and the grounded electrode colder, the thermal gradient decreases the immediate hot-HV-edge field by about 13%, while the central minimum and grounded-side regional maximum increase by about 38% and 30%, respectively. The HV-to-GND regional-maximum ratio decreases from 1.77 to 1.34, showing a substantial redistribution of the DC interface field while the largest regional field remains on the HV side.
The time-dependent calculations further show that DC field formation is strongly position dependent. Between 1 and 12 h, the interval-based apparent relaxation times are about 3.2 h on the hot HV side and 15.8 h on the colder grounded side. The HV-side field is close to its stationary value after 12 h, whereas the grounded-side deviation remains about 3.9% after 24 h. Together with the non-monotonic central response, these results show that one common relaxation time cannot represent DC field formation along the complete spacer interface. Thermal condition and DC pre-stress duration must therefore be controlled in future flashover measurements. The developed test bench and test method provide the basis for the experimental investigation of RQ1–RQ3 after manufacture and physical qualification.