FDTD Simulation of Lightning Surges on Overhead Wires in the Presence of Corona Discharge

Abstract

A simplified model of corona discharge for finitedifference time-domain (FDTD) computations has been applied to analyzing lightning surges propagating along overhead wires with corona discharge. The FDTD computations simulate the experiments of Inoue and Wagner et al.. In Inoue's experiment, a 12.65-mm radius, 1.4-km-long overhead wire was employed, and in Wagner et al.'s experiment, a 21-or 25-mm radius, 2.2-km-long overhead horizontal wire was employed. The critical electric field on the surface of the 12.65-mm-radius wire for corona initiation is set to E 0 = 1.4, 2.4, or 2.9 MV/m, and those for 21-and 25-mmradius wires are set to E 0 = 2.2 and 2.1 MV/m, respectively. The critical background electric field for streamer propagation is set to E cp = 0.5 MV/m for positive voltage application and E cn = 1.5 MV/m for negative voltage application. The FDTD-computed waveforms (including wavefront distortion and attenuation at later times) of surge voltages at three different distances from the energized end of the wire agree reasonably well with the corresponding measured waveforms. Also, the FDTD-computed waveforms of surge voltages induced on a nearby parallel bundled conductor agree fairly well with the corresponding measured waveforms.

FDTD simulation of direct lightning strike to a phase conductor: Influence of corona on transient voltages at the tower

“…Also, it has been shown in [1] that the computed increase of coupling between the energized wire and another wire nearby due to corona discharge agrees well with the corresponding measured one. Further, it has been shown in [2] that computed waveforms (including wavefront distortion and attenuation at later times) of fast-front surge voltages at different distances from the energized end of 1.4-km and 2.2-km long overhead wires agree reasonably well with the corresponding measured waveforms.…”

“…As a result, the corona radius has a nonuniform distribution along the conductor. More details on our corona model can be found in [1,2,4]. of 3 s and time to half-peak value of about 40 s. Additionally, injected current waveforms with a peak of 20 kA and the same waveshape parameters were used in the simulations, although the corresponding current waveforms are not shown here.…”

“…In our previous study, a simplified (engineering) model of corona discharge developed in Thang et al [1,2] for the finite-difference time-domain (FDTD) computations (Yee [3]) has been applied to analysis of transient voltages at the tower of a 60-m-high transmission line, the ground wire of which is struck by lightning, and transient voltages computed with and without corona on the ground wire are compared. It is clear from the results of Thang et al [4] that the transient voltages are reduced by corona discharge on the ground wire, but the reduction of transient-voltage peak is not very significant: the upper-, middle-, and lower-phase-voltage peaks are reduced by 15,14, and 13% for a positive stroke with 50-kA-peak and 3-s-risetime current, and those for a negative stroke with the same current waveform parameters are reduced by 10, 9, and 8%, respectively. In this paper, we apply the same simplified model of corona discharge to the analysis of transient voltages of a 40-m tower for the case of direct lightning strikes to an upper phase conductor, at the midpoint of the span (shielding failure).…”

“…The validity of the corona model of Thang et al [1,2] has been tested against experimental data found in Noda [5], Noda et al [6], Inoue [7], and Wagner et al [8]. Specifically, it has been shown in [1] that the waveform of radial corona current, and the relation between the total charge (charge deposited on the wire and emanated corona charge) and applied voltage (q-V curve), computed using the FDTD method including the corona model for 22 m and 44 m long horizontal wires, agree reasonably well with the corresponding measured ones.…”

A corona discharge model using the transmission-line modelling method

“…Recently, a number of papers have demonstrated simplified methods when modelling corona discharge on wires using the Finite-Difference Time-Domain (FDTD) technique [1,2] using a graded mesh. The aim of this paper is to demonstrate an alternative approach which does not require a graded mesh using another popular modelling technique; the Transmission-Line Modelling (TLM) method This paper utilises an embedded-wire TLM node allowing coupling between field and wire to be achieved using an additional wire network, [3] positioned within the TLM cell in addition to the basic link lines that constitute the General Symmetric Condensed Node (GSCN).…”