Method for lightning flashover types identification on transmission lines utilizing OPGW optical polarization state signal characteristics

The Optical Fiber Composite Overhead Ground Wire (OPGW) has undergone rapid development and has been extensively applied in the realm of intelligent power system status monitoring. This article conducts research on lightning signal waveform in optical fibers within the context of lightning fault location using the OPGW optical polarization state method. A transmission line model is simulated employing ATP-EMTP to capture electrical characteristics when lightning strikes the OPGW. Subsequently, by employing the Faraday magneto-optical effect principle, electrical signals are transformed into optical polarization states within the internal fibers of OPGW. Using such data, the optical signal propagation characteristics when lightning strikes OPGW are deduced. Finally, a method for discerning between induced lightning and direct lightning strikes through the analysis of time-frequency domain characteristics in OPGW. This method addresses the limitation of traditional OPGW lightning monitoring, which is unable to deduce electrical signals, thus achieving precise identification of lightning-induced transient faults in transmission lines.


Introduction
Lightning strike localization is the technique that locates the occurrence of lightning strikes on a line when it is struck by lightning, based on relevant electrical or non-electrical parameters.Presently, lightning strike localization methods for power transmission lines primarily employ traveling wave and electromagnetic wave multi-point localization approaches [1] .The traveling wave method entails locating and identifying faults based on the propagation time and characteristics of transient current traveling waves generated during lightning strikes [2] .While this method could yield high localization precision, it is generally highly susceptible to the influence of line parameters, including uncertainties in traveling wave velocity, challenges in identifying wave reflections, and waveform distortions.Practical applications are constrained by various limitations and elevated costs.Lightning location systems (LLS) have gained widespread use by relying on direction finding and time difference techniques [3] However, 2. The principle of OPGW lightning monitoring using the Faraday magneto-optical effect OPGW comprises internal optical fibers enclosed within a protective sheath, serving as the communication pathway for OPGW.Externally, it consists of twisted conductors, which function as an overhead ground wire and provide protection for the internal fibers.During a lightning strike, the lightning current passing through the OPGW can be divided into two components: first, it travels axially through the OPGW, and second, it follows a helical path within the OPGW.
When the lightning current travels along the helical path, it effectively forms a helical coil, generating a magnetic field parallel to the OPGW's axis.In accordance with the Faraday effect, linearly polarized light passing through fibers aligned parallel to this magnetic field and propagating in the direction of the magnetic field undergoes a rotation of its polarization plane.The rotation angle is directly proportional to the product of the magnetic field's strength and the length of the optical fiber traversed, with the proportionality constant being influenced by the properties of the medium and the frequency of the light wave.This rotation angle θ is referred to as the Faraday rotation angle which is illustrated in Figure 1.Since θ elegantly mirrors the magnitude of the magnetic field engendered by the potent lightning current, it can be expressed as follows: where V is the dielectric constant of the medium; L is the distance over which light propagates in the magneto-optical material.According to the formula for calculating the magnetic field intensity around a current-carrying straight conductor, it can be obtained that: where μ0 is the magnetic permeability of the medium; I is the current in the conductor; r is the distance to the center of the conductor.Hence there is: where K=Vμ0 /2πr.According to Malus's law, the amplitude of the current after modulation Im can be expressed as: where βm is the emitted light intensity, β0 stands for the incident light intensity, and α is the angle between the optical axes of the polarizer and analyser.Consequently, if a continuous linearly polarized light is introduced into the internal optical fibers of OPGW, the sudden impact of a lightning current on OPGW will lead to an abrupt alteration in the optical polarization state at that location.Therefore, measuring the change in the optical polarization state at the end of the optical fiber allows for the indirect calculation of the magnitude of the lightning current responsible for generating the magnetic field.This, in turn, facilitates the real-time monitoring of lightning events on OPGW.

Waveform characteristics of lightning-stroke OPGW optical signals using ATP-EMTP
This study utilizes ATP-EMTP to establish a simulation model for power transmission lines.It performs calculations for lightning strikes on OPGW under various lightning current amplitudes, resulting in the acquisition of OPGW current at different distances from the lightning strike point along the transmission towers.Subsequently, the Faraday effect formula is utilized to calculate the rotation angle θ, leading to the generation of lightning current waveforms represented in the form of changes in light intensity.
The transmission line model is based on an actual 110 kV power transmission line, with OPGW grounded in each tower.The transmission line utilizes the Jmarti frequency model, representing the towers as single-circuit cup-shaped structures.The lightning current is modelled according to the standard 2.6/50 μs wave-shape.Relevant electrical parameters are obtained through ATP simulations.Finally, the lightning-induced optical signal is calculated by integrating Equation ( 4), leading to the generation of lightning current waveforms.The simulation experiments encompass scenarios where lightning currents with amplitudes of 20 kA, 60 kA, 100 kA, 150 kA, and 200 kA strike the OPGW.Specifically, for lightning currents with amplitudes of 20 kA, 60 kA, and 100 kA, no flashover occurs in the line.However, for lightning currents of 150 kA and 200 kA, flashover faults occur in the C-phase of the line.
The simulation results obtained from Figures 2 and 3 reveal that when the lightning current is relatively low, at 60 kA for instance, the corresponding rotation angle θ is accordingly small.As a result, the lightning waveform exhibits a gradual rise in the wavefront slope, with no noticeable oscillations.However, when the lightning current is higher, the corresponding rotation angle θ increases, resulting in a steeper slope of the lightning waveform.Additionally, the waveform exhibits oscillations.This is due to the relatively large value of θ, leading to an angle of rotation that surpasses one full cycle, resembling a saturation effect.The waveform's tail exhibits oscillatory decay, with different oscillation frequencies between the wavefront and wavetail.The wavefront oscillates at a higher frequency, while the wavetail oscillates at a lower frequency.When the lightning current amplitude is significantly high, it may lead to an insulator flashover within the transmission line.Under such circumstances, the Faraday rotation angle becomes exceptionally large, and there are noticeable differences in the lightning waveform compared to non-flashover conditions.The resulting lightning signal waveform exhibits continuous oscillations and an extended duration.The wavefront portion of the waveform displays densely packed oscillations.Moreover, the waveform exhibits transverse wave characteristics, with the transverse wave frequency being the power frequency or one of its harmonic frequencies.The simulation demonstrates a substantial correlation between the lightning waveform of the transmission line based on OPGW optical sensing and the amplitude of the lightning current.The intensity of wavefront oscillations is positively associated with the magnitude of the lightning current, and the duration of waveform oscillations also strongly correlates with the lightning current's amplitude.Flashover waveforms exhibit distinct differences in waveform characteristics when compared to nonflashover scenarios.Flashover waveforms have a longer duration, densely packed oscillations in the wavefront, and exhibit transverse wave characteristics, with a transverse wave frequency at the power frequency or one of its harmonic frequencies.

Identification of fault based on field measurements of OPGW lightning current waveforms
This paper takes data collected from the OPGW optical sensing lightning localization system installed on the 220 kV transmission line in Jiaxing as an illustrative example.Over the operational period of the system, an extensive dataset has been gathered.Compared with meteorological data and the LLS, several sets of characteristic lightning data with varying lightning current amplitudes have been meticulously chosen.An examination of the field-measured lightning data is exemplified in Figure 4.In this scenario, the OPGW lightning waveform is captured when the lightning current reaches 100 kA.The lightning current amplitude is significantly elevated, resulting in a waveform with a substantial amplitude.The wavefront demonstrates prolonged and intense oscillations, with oscillation peak amplitudes remaining consistent.This phenomenon aligns with the previously discussed saturation of the rotation angle θ.Subsequently, the waveform gradually enters a phase of oscillatory decay, mirroring the simulated waveform presented in Figure 3.As the lightning current amplitude progressively increases, it becomes evident that the characteristics of the field-measured lightning waveforms closely mirror those of simulated lightning waveforms without flashover.
Figure 5 illustrates the comprehensive waveform of the field-measured OPGW lightning signal during flashover events.During flashovers, the signal exhibits a relatively protracted duration, spanning approximately 20 ms, accompanied by a substantial signal amplitude.The wavefront section displays tightly clustered oscillations, including distinct 50 Hz or 100 Hz harmonic components.The fieldmeasured lightning-induced OPGW flashover waveform demonstrates remarkable similarity to the     Moreover, extensive analysis of various lightning signals has revealed that specific lightning events manifest clear temporal and spectral waveform characteristics.As illustrated in Figure 6, typical induced lightning signals tend to exhibit comparatively smaller amplitudes.They show limited oscillations near the wavefront and possess relatively extended wave tails.Their spectral distribution is predominantly concentrated within the 0 kHz to 10 kHz frequency range.In contrast, Figure 7 depicts direct lightning waveforms characterized by consistently substantial amplitudes, indicating their high intensity.In the vicinity of the wavefront of direct lightning signals, typically within the range of 0.1 ms to 0.5 ms, one can observe pronounced, rapid, and vigorous oscillations.This phenomenon results from the large magnetic fields generated by intense lightning currents, which, in turn, induce significant Faraday rotation of polarized light.When the rotation angle becomes substantial, it leads to rapid and vigorous oscillations within the signal waveform.The spectral content of direct lightning signals primarily spans the 0 kHz to 30 kHz frequency range.When compared to induced lightning spectra, the high-frequency components in direct lightning signals are more pronounced, particularly in the 10 kHz to 20 kHz range.This is mainly attributed to the rapid oscillations occurring near the wavefront of direct lightning signals.
On the other hand, flashover signals exhibit spectral content mainly in the low-frequency range of 0 to 500 Hz, with prominent peaks occurring at or near power frequencies and their harmonics, including 50

Conclusion
This paper utilizes the ATP-EMTP to simulate the distribution of electrical currents across individual transmission line towers when subjected to lightning strikes.The application of the Faraday effect integration formula enables the computation of the polarization state of light.This methodology facilitates the exploration of lightning waveform characteristics under various lightning current magnitudes.Significantly, this approach adeptly translates electrical signals into optical signals, reproducing the lightning waveforms as reflected by the optical signals when lightning strikes OPGW.
The findings affirm a substantial correlation between the lightning waveforms observed in transmission lines utilizing OPGW optical sensing and the magnitude of the lightning current.The intensity of oscillations near the wavefront of the waveform is positively correlated with the lightning current's magnitude, as is the waveform's duration.Notably, flashover signals exhibit pronounced disparities when compared to non-flashover signals in terms of waveform features.These disparities encompass intense oscillations near the wavefront, extended signal duration, sometimes spanning several tens of milliseconds, significant signal amplitudes, and the conspicuous presence of harmonics at 50 Hz or 100 Hz within the oscillatory waveform.
simulated waveform.Thus, similarity encompasses densely packed wavefront oscillations, prolonged durations reaching several tens of milliseconds, substantial signal amplitudes, and the presence of distinct 50 Hz or 100 Hz harmonic components within the oscillatory waveform.This corroborates the precision of the previous simulation results.The study of field-measured lightning-induced OPGW flashover waveform characteristics underscores pronounced disparities between flashover and nonflashover waveforms.The most conspicuous contrast lies in the flashover waveforms' extended duration, extensive oscillations, and the inclusion of power frequency harmonic frequencies.

Figure 4 .
Figure 4. Waveform of the lightning strike OPGW without flashover.

Figure 6 .
Figure 6.Characteristics of time-domain waveform (left) and spectral distribution (right) of typical induced lightning strike signals.

Figure 7 .
Figure 7. Characteristics of time-domain waveform (left) and spectral distribution (right) of typical direct lightning strike signals.
Hz, 150 Hz, 200 Hz, 250 Hz, and 300 Hz.Flashover signals primarily result from insulator breakdown, which generates low-frequency continuous current signals.Consequently, flashover signals prominently feature a power frequency characteristic.By employing these discriminative criteria and associated thresholds, it is feasible to differentiate between induced and direct lightning events.