Power supply: Drones automatically inspect overhead power lines
Drones can detect damage to overhead power lines. This is largely automatic.
An inspection drone provides a camera image of the test site for the Oak Ridge National Laboratory's power line inspection system.
(Image: Jason Richards / ORNL, U.S. Dept. of Energy)
Scientists at the US Department of Energy's Oak Ridge National Laboratory (ORNL) have developed an automated inspection system for above-ground power lines using drones in collaboration with the US power utility ETB. Sensors in the power lines and transformers detect deviations in the power grid and send the information to the drone system. Inspection drones then take off automatically for an inspection flight and, if necessary, request a maintenance team for repairs.
In the joint project "Autonomous Intelligent Measurement Sensors and Systems" (AIMS), the research team uses commercially available drones, sensors and software. In addition, techniques, algorithms and automated protocols have been developed by ORNL specifically for the system. The extensive use of commercial technology is necessary to keep the costs of such an automated drone inspection system low, according to ORNL and ETB.
However, the system does not work entirely without in-house developments: The researchers had to develop a special ultraviolet camera, for example. Commercial systems would have been too heavy, weighing around 4.5 kg, and would have cost 25,000 US dollars. By developing their own camera, the researchers were able to reduce the weight to around 450 g and the costs to 250 dollars, so that the camera could be housed in a drone at low cost.
The control system begins with continuous monitoring of voltage and current in the power lines and transformers. Such sensors are located throughout the grid and continuously send their data to the central management system of the electricity supply company. This compares the voltage and current values supplied with the waveforms in the Grid Event Signature Library, in which information on grid data is stored. In the event of irregularities, a reconnaissance drone from a substation can be sent out to detect any damage. Localization is carried out using GPS information on the location of the damage, which is transmitted to the drone. The system automatically decides which drone is closest to the potential damage site and has sufficient battery capacity for the inspection flight.
Damage assessment by drone
The drone then flies autonomously to the potential damage site and collects data using a high-frequency sensor, visual cameras, infrared cameras and a sound detector. Arcs, for example, can be recorded visually and detected and evaluated using an algorithm. The drone transmits the measured values to a ground control system in real time via mobile networks. Depending on the data situation, it automatically sends other drones with special inspection capabilities to – depending on what further measurements are required. Depending on the damage situation, a repair team is then sent to the site.
The system allows the automatic inspection decisions to be overridden by human personnel at any time. For example, the drones can be forced to land immediately and sent back to a charging station ahead of schedule.
The researchers have already tried out the system in a test setup at the training facility of the electricity grid operator ETB in Chattanooga, Tennessee, and have achieved good results. ETB is therefore interested in using the drone inspection system for routine and emergency inspections, for example after storms. The drones could quickly locate the worst storm damage after storms and speed up the restoration of the power grid. The largest 20 of ETB's 100 substations are to be equipped with the technology to begin with in order to be able to monitor an initial part of the approximately 1,550 square kilometer supply area. The company assumes that this will save high inspection costs.
(olb)