Power lines can be hazardous to low-flying aircraft. Fortunately, it may be possible to use microwave and millimeter-wave radar systems to accurately detect power lines, and prevent accidents. Even under conditions of fog, rain, and snow, it should be possible to measure the polarimetric radar backscattering from a power line using microwave or millimeter-wave radar systems. A main concern, addressed in this study, is the effect of thick ice on these measurements.
Studies have been performed on the use of electro-optical laser radar for collision warning.1 Unfortunately, the laser radar has numerous shortcomings, including limited range, significant atmospheric attenuation under inclement weather conditions, and difficulty in automating a detection system to warn a pilot of oncoming power lines which make it impractical for power-line collision-warning systems.
Millimeter-wave radar systems, on the other hand, can be used to detect thin objects like power lines in foggy, cloudy, snowy and rainy conditions. While all microwave and millimeter-wave radars are not suitable for this application, some have been used effectively.2 In one such case, the millimeter-wave radar used linearly polarized waveforms and modeled transmission lines as long, perfectly conducting cylinders.3 Since power lines are conducting cylindrical wires, they represent difficult targets for many radars. But because a high-voltage power line is made up of strands of wires in a helical arrangement, backscattering detection can be used to detect the power-line cables.
Due to the radar resolution possible at millimeter-wave frequencies, the helical geometry of the power lines is an important factor in influencing the scattering behavior of the electromagnetic (EM) waves, and this can be used in detecting power-line field emissions in off-specular directions. The surface of the power-line cables is periodic along the axis of the cables and usually the period is only a fraction of the helical pitch.
High-voltage power-line cables are usually constructed from a number of aluminium strands twisted helically around a central core of one or more steel strands. The current-carrying capacity of the cable depends on the number of layers and diameter of aluminium/steel strands. However, in an electrical power distribution network, low-tension and high-current cables are used, made of either aluminium or copper strands.
A typical power-line cable geometry can handle current loads to 420 A (Fig. 1). Although research has been performed on polarimetric radar backscattering measurements of a variety of power line cables,4-6 no known literature has reported on the effects of thick ice layer on backscattering measurements. For the current research, measurements were made of the radar cross section versus angle of rotation of a 2-cm-diameter power-line cable at 5, 10, 40, and 82 GHz. The scatterometer systems are based on 8753 vector network analyzers from Agilent Technologies (model 8753A for the 5- and 10-GHz bands and model 8753C for the 40- and 82-GHz bands). The measurement systems feature phase and amplitude measurement capabilities and 100-dB dynamic range. The scatterometers used at microwave frequencies (5 and 10 GHz) are slightly different than those used at the millimeter-wave bands (40 and 82 GHz). The microwave system employs a single radar antenna; the millimeter-wave system has dual antennas.
Figure 2 shows a microwave system capable of measuring the backscatter of electrical cables with a good signal-to-background-noise ratio for all incident angles. The electrical cable is mounted on a Roha-cell foam pedestal in an anechoic chamber. For the radar cross-section measurements, the position of the electrical cable with respect to the antenna coordinate system is accomplished by an azimuth-over-elevation positioner. The azimuth turntable is driven by a computer-controlled stepper motor with accuracy of a fraction of a tenth of a degree; the elevation controller is a precise analog positioner.
A transmit/receive antenna for the microwave system consists of an orthomode transducer and a dual-polarized square horn. The scatterometer provides receive and transmit signals to a switch and a microwave amplifier assembly. The microwave receive and transmit signals pass to the 8753A network analyzer and a computer. The computer controls the turntable stepper motor and sends data to the printer.
Figure 3 shows the block diagram of the millimeter-wave radar cross-section measurement system. The electrical cable is mounted on a turntable with Roha-cell foam on one side of the anchoic chamber. The turntable operates in both azimuth and elevation planes with rotational information controlled by the computer. The scatterometer, which is installed at one end of the chamber, operates in coherent mode. The scatterometer, contains two corrugated horns (for transmit and receive), two isolators, a polarizer and a otrthomode transducer. A Faraday rotator is used to achieve polarization in the transmitter while the orthomode transducer is used to establish receiver polarization. Two separate millimeter-wave oscillators are employed, operating at 40 and 82 GHz.
Radar cross-section measurements on the cable were set up in a 7-m-long tapered anechoic chamber; the cable was set a distance of 5.5 m from the receive and transmit antennas. The power-line cable, which was 25 cm long and 2 cm in diameter, was mounted on a Roha-cell foam structure, with the foam structure rigidly attached to the turntable platform. The permittivity of the Roha cell s 1.07 at microwave frequencies and 1.1 at millimeter-wave frequencies. The loss tangent of the Roha-cell structure is less than 0.001 at both microwave and millimeter-wave frequencies. Its radar cross section is below −35 dBm.
For calibration purposes, the chamber and turntable including attachments were measured in the absence of the power-line cable. These measurements were then subtracted from the measured power-line responses. In these measurements, the scatterometers were calibrated using the single-target calibration technique.7 A 25-cm metallic sphere was used for the microwave 5- and 10-GHz calibration target, while 4- and 2-cm metallic spheres were used for the 40- and 82-GHz calibration targets, respectively. This technique provides less than 0.5 dB magnitude error and less than 5 deg. phase error. In all cases, the signal-to-noise ratio (SNR) was better than 25 dB.
During measurements, the power-cable line is placed in the H-plane of the antenna system and radiation patterns are plotted. The power-cable orientation during the measurement is similar to that of a radar system mounted on a low-flying aircraft, with power lines in the horizontal plane. Figure 4 shows the plot of measured radar cross-section versus angle of rotation in the co-polar and the cross-polar planes at 5 GHz. In Fig. 4, the co-polar and cross-polar plots are shown by full and dotted lines, respectively. Figure 5 shows the measured co-polar and cross-polar at 10 GHz for the braided power line cable (Fig. 1). Figures 4 and 5 show one peak value of radar cross section at an incident angle of 0 deg. There is no significant backscatter at incident angles greater than zero deg. The cross-polar level is less than −35 dB for all angles at 5 and 10 GHz.
Figure 6 shows measured co-polar and cross-polar radar cross sections versus the angle of incidence at 40 GHz. The co-polar and cross-polar radar cross-section patterns appear in Fig. 7. The radar cross-section response of the electric cable at both polarizations (co-polarization and cross-polarization) has peaks at normal incidence and certain discrete incident angles. In a periodic structure such as a power-line cable, Bragg backscatter can be predicted by:
λ = the wavelength and
L = the period (shown in Fig. 1).
In Figs. 6 and 7, the Bragg scattering ceases at incidence angles larger than 15 deg. Higher-order Bragg modes occur at very low levels at angles away from normal incidence, since most of the scattered energy is in the specular direction.
In the anechoic chamber, backscattering measurements were performed on a power line covered with a layer of thick ice, at 5, 10, 40, and 82 GHz. A radar transmitter and a power line positioner were set up in the anechoic chamber. During these measurements, the outdoor temperature was about −20°C; the doors and windows of the anechoic chamber were opened during the measurements to maintain the temperature of the chamber at about −50°C. Ice was formed over the power line by spraying it with water and allowing it to freeze. A layer of ice with an average thickness of about 0.5 cm was formed on the horizontal surface of the power-line sample. To simulate real-world ice on outdoor cables, water was dripped onto the braided power-line cable and allowed to freeze. The thick lines in Figs. 4-7 represent the co-polar radar cross-section measurements for the ice-covered power line. The measurements show that ice layers modify the cable-surface reflectivity and the effective surface roughness.
These measurements show that the radar cross section of the ice-covered power line have slightly lower values (a few dB) than the dry power line at angles close to normal incidence, but the values of these radar cross-section measurements increase for higher angles of incidence. These measurements provide a guideline for creating a detection algorithm for power lines using microwave and millimeter-wave radar systems. In conclusion, at microwave frequencies for small diameter power lines (D/λ << 1), there is no significant backscattering at angles away from normal incidence.
- M. Savan, and D.N. Barr, "Reflectance of wires and cables at 10.6 micrometer," MSEL-NV-TR-0063, Center for Night Vision and Electro-Optics, January 1988.
- B. Rembold, H.G. Wippich, M. Bischoff, and W.F.X Frank, "A MM-wave collision-warning sensor for helicopters," Proceedings of Military Microwave (London, England), 1982, pp. 344-351
- B. Rembold, "Radar cross section of long wires," IEEE Transactions on Antennas & Propagation, Vol. 32, No. 10, October 1984, pp. 1124-1126.
- K. Sarabandi, L. Pierce, Y. Oh, and F.T. Ulaby, and M. Park, "Power lines: Radar measurements and detection algorithm for polarimetric SAR images," IEEE Transactions on Aerospace and Electronic Systems, Vol. 30, No. 2, April, 1994, pp. 632-643.
- K. Sarabandi and M. Park, "Millimetre-wave radar phenomenology of power lines and a polarimetric detection-algorithm," IEEE Transactions on Antennas & Propagation, Vol. 47, December, 1999, pp. 1807-1813.
- K. Sarabandi and M. Park, "A radar cross-section model for power lines at millimetre-wave frequencies," IEEE Transactions on Antennas & Propagation, Vol. 51, No. 7, September 2003, pp. 2353-2360.
- K. Sarabandi, and F. T. Ulaby, "A convenient technique for polarimetric calibration of radar systems," IEEE Transactions on Geoscience and Remote Sensing, Vol. 28, No. 6, November 1990, pp. 1022-1033.