Investigation on Direct Lightning Strike to High Voltage Transmission Lines

Abstract

Lightning continues to be the single largest natural cause of line outages. With high voltage lines spanning millions of kilometres around the globe, it has been a serious concern to transmission line engineers. Quite unfortunately, the lightning performance of ultra-high voltage lines falls far below that estimated from the available models. There are two reasons for this difference, namely, inadequacy in the present standards in evaluating the lightning attachment process and in ascertaining the surge response of these tall lines. The recent progress in simulating the attachment process is dealt with in the latest CIGRE technical brochure. It is shown that the simplified model proposed over there can account for the observed differences between estimated using older models and the field data. Different modelling approaches have been employed to evaluate lightning surge response. The most popular approach in power engineering is a lumped current source in parallel with an impedance. However, there is no consensus among researchers regarding the value of impedance to be used in such models. Based on observations over tall towers, it has been inferred that the peak value of the lightning current depends on the strike object. An alternative modelling approach for evaluating the lightning surge response of lines is to make use of the return stroke models. Among the different categories of return stroke models available, only the transmission line and electrodynamic models can, in some sense, be employed for modelling a direct strike to the transmission line. Even though current is not assumed in the transmission line models, the mode of propagation is assumed as TEM mode, which is not true on the channel, at least in the initial stages of current evolution. Even though the mode of propagation is not assumed in electrodynamic models, usually, a lumped excitation is assumed. Therefore, a model that does not assume the excitation and the mode of propagation is essential. A self-consistent return stroke model recently proposed by the group would be an ideal tool for the required investigation. It basically emulates basic essential physical processes along with accurate tracing of dynamic electromagnetic fields using time-domain thin-wire formulation. In order to apply this model to simulate a direct strike scenario to a transmission line, the field computation methodology had to be extended from the axisymmetric to a fully three-dimensional wire geometry. Further, wire junctions need to be considered, which involve modifications in the spatial interpolation functions for evaluating current. These were achieved and for the latter, spatial basic function is borrowed by the Numerical Electromagnetic Code. Using this extended self-consistent return stroke model, detailed simulations are carried out for a direct strike to the transmission line. The peak current during a strike-to-phase conductor and strike-to-ground (or simplified ground wire-tower structure) are deduced. Using this consistent modelling approach, it has been clearly shown by modelling that the peak current value for the same conditions reduces to almost half of that for a strike to ground/ground wire-tower structure, in line with the field observations. The physical reason for this difference has also been identified. Furthermore, the mode of propagation of the stroke current along the phase conductor is also assessed and was found that in the initial stages of surge propagation, the mode is TM and later on, as the surge progresses, the mode changes from TM to TEM. In summary, this work has delved into some of the fundamental aspects of lightning strikes to overhead lines.