Modelling the Four‐Dimensional Power Line Sag Dynamics by Point Signature in Time‐Series Synthetic‐Aperture Radar Images

ABSTRACT Power lines in synthetic‐aperture radar (SAR) images are reflected as point signatures, which fluctuate annually in both the azimuth and range directions. A novel framework is proposed in this paper to evaluate the four‐dimensional dynamic status of power line sag using those point signatures in time‐series SAR images. An inversion model is proposed to reconstruct the three‐dimensional power line sag profile based on the point signature positions and power line suspension points in the SAR coordinates. By combining the elevation function and the power line parabolic function derived in the SAR image, the 3D geometry of power line sag can be reconstructed using the range–Doppler model. A sag–temperature model is proposed to evaluate the 4D dynamic status of power lines in time‐series SAR images. The results show that the proposed model has been able to identify one of the power line sag failure events in the study area, which occurred around September 2016, with an approximate 2.54‐m line sudden sag downward. This case study shows the potential of spaceborne SAR for real‐time power line sag monitoring, particularly under extreme weather conditions.