Search

Locating defects in an underground cable connection can be based on the propagation of transient signals emerging from faults or weaknesses in the insulation. The accuracy depends on the precise knowledge of the propagation characteristics of high-frequency components in the signals. These characteristics are analysed for multi-conductor cables in case of non-perfect rotational symmetry. Deviations can arise from the positioning of the conductors within the cable cross section or from an inhomogeneous distribution of dielectric materials. It is investigated to what extent the signal propagation modes, applicable for a symmetric configuration, remain a suitable representation for signal interpretation when the symmetry is slightly disturbed. This is judged by means of Raleigh-Schrödinger perturbation analysis in comparison to electromagnetic field simulations. Cable designs for the medium-voltage level incorporate measures to confine the electric field in designated regions. Slight asymmetry does not lead to strongly perturbed modes. For low-voltage cables, electric field control is not needed, and the conductors are positioned within short distance of each other. Modes can change significantly upon a relatively small asymmetry. Inhomogeneous heat dissipation, e.g. from a short-circuit current upon a fault, can cause a temporarily asymmetric permittivity distribution over the cable cross section. The feasibility to detect such effect from changing signal velocities of the propagation modes is discussed.

Locating defects in an underground cable connection can be based on the propagation of transient signals emerging from faults or weaknesses in the insulation. The accuracy depends on the precise knowledge of the propagation characteristics of high-frequency components in the signals. These characteristics are analysed for multi-conductor cables in case of non-perfect rotational symmetry. Deviations can arise from the positioning of the conductors within the cable cross section or from an inhomogeneous distribution of dielectric materials. It is investigated to what extent the signal propagation modes, applicable for a symmetric configuration, remain a suitable representation for signal interpretation when the symmetry is slightly disturbed. This is judged by means of Raleigh-Schrödinger perturbation analysis in comparison to electromagnetic field simulations. Cable designs for the medium-voltage level incorporate measures to confine the electric field in designated regions. Slight asymmetry does not lead to strongly perturbed modes. For low-voltage cables, electric field control is not needed, and the conductors are positioned within short distance of each other. Modes can change significantly upon a relatively small asymmetry. Inhomogeneous heat dissipation, e.g. from a short-circuit current upon a fault, can cause a temporarily asymmetric permittivity distribution over the cable cross section. The feasibility to detect such effect from changing signal velocities of the propagation modes is discussed.

Traditionally, Extra High Voltage (EHV) transmission systems consist exclusively of Overhead lines (OHL) but in the recent years the utilization of High Voltage AC (HVAC) underground power cables (UGC) is increasing due to environmental, political and operational aspects. The application of EHV cables will influence the operation of the transmission system in both steady and transient state. Therefore, formulation of accurate cable simulation models is of great importance in order to understand the new phenomena introduced. This paper presents a comparison between two cable models, a detailed and a more simplified one, as well as a sensitivity analysis on several cable parameters. In this way, guidelines can be extracted concerning the parameters that have larger impact on the cable response and must be modeled in detail and the ones where simplifications can be made for achieving better computational speed. The simulations are performed using software package PSCAD/EMTDC.

Traditionally, extra high voltage (EHV) transmission systems consist exclusively of overhead lines (OHL) but in the recent years the utilization of high voltage AC (HVAC) power cables is increasing due to environmental, political and operational aspects. To study the impact of HVAC power cables on temporary overvoltages and slow front transients, a simulation model of the large-scale network is required. Guidelines concerning the modeling detail for the surrounding network need to be established to ensure sufficient accurate results. Existing guidelines for purely OHL transmission networks may not be adequate since the integration of HVAC power cables shift the harmonic impedance of the network to lower frequencies and that can affect the extent of the needed simulation model. Much effort can be saved from obtaining detailed data for an extensive but proper model, but a less detailed model might not describe the network behavior with sufficient accuracy. In this study, simplification options are considered, for which the time to construct and analyze a simulation model can be reduced. The simplification options are compared with the reference model (full detail), and judged based on three indicators for the level of accuracy: frequency and amplitude of the main resonance, and maximum over-voltage. As a result, guidelines on the extent of the simulated network and in which detail the model has to be built are proposed. The inverse approach, i.e. increasing level of detail until simulation results hardly change leads to similar results, but requires equivalent short-circuit networks for the rest of the grid. The method is tested for the Dutch 380 kV grid where an underground cable connection of 10.8 km length is introduced in 2013 and a second connection is planned for 2017.

Traditionally, extra high voltage (EHV) transmission systems consist exclusively of overhead lines (OHL) but in the recent years the utilization of high voltage AC (HVAC) power cables is increasing due to environmental, political and operational aspects. To study the impact of HVAC power cables on temporary overvoltages and slow front transients, a simulation model of the large-scale network is required. Guidelines concerning the modeling detail for the surrounding network need to be established to ensure sufficient accurate results. Existing guidelines for purely OHL transmission networks may not be adequate since the integration of HVAC power cables shift the harmonic impedance of the network to lower frequencies and that can affect the extent of the needed simulation model. Much effort can be saved from obtaining detailed data for an extensive but proper model, but a less detailed model might not describe the network behavior with sufficient accuracy. In this study, simplification options are considered, for which the time to construct and analyze a simulation model can be reduced. The simplification options are compared with the reference model (full detail), and judged based on three indicators for the level of accuracy: frequency and amplitude of the main resonance, and maximum over-voltage. As a result, guidelines on the extent of the simulated network and in which detail the model has to be built are proposed. The inverse approach, i.e. increasing level of detail until simulation results hardly change leads to similar results, but requires equivalent short-circuit networks for the rest of the grid. The method is tested for the Dutch 380 kV grid where an underground cable connection of 10.8 km length is introduced in 2013 and a second connection is planned for 2017.

Traditionally, Extra High Voltage (EHV) transmission systems consist exclusively of Overhead lines (OHL) but in the recent years the utilization of High Voltage AC (HVAC) underground power cables (UGC) is increasing due to environmental, political and operational aspects. The application of EHV cables will influence the operation of the transmission system in both steady and transient state. Therefore, formulation of accurate cable simulation models is of great importance in order to understand the new phenomena introduced. This paper presents a comparison between two cable models, a detailed and a more simplified one, as well as a sensitivity analysis on several cable parameters. In this way, guidelines can be extracted concerning the parameters that have larger impact on the cable response and must be modeled in detail and the ones where simplifications can be made for achieving better computational speed. The simulations are performed using software package PSCAD/EMTDC.