Simultaneous distributed acoustic and temperature sensing for early detection and localization of tracking phenomena in polymeric insulators
This paper presents a novel distributed fiber-optic sensing framework that simultaneously monitors acoustic and thermal signatures to enable the early detection, continuous tracking, and precise localization of surface tracking degradation in polymeric insulators under high-voltage conditions.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
High-voltage power lines stretch across landscapes, carrying electricity over vast distances. To keep this energy flowing safely, the lines are supported by insulators, often made of durable polymer materials that resist weather and contamination. Over time, however, these insulators can begin to fail. When dirt and moisture settle on their surfaces, tiny electrical sparks can jump across the material. These sparks are not just harmless flickers; they generate heat and create carbon tracks that slowly eat away at the insulator. If left unchecked, this process, known as tracking, can lead to a complete breakdown, causing power outages or even fires. The challenge for engineers has always been how to spot this slow, creeping damage before it becomes catastrophic. Traditional methods often rely on checking specific points or looking for visible signs, but these approaches struggle to watch the entire length of a power line continuously or to see the problem developing in its earliest, quietest stages.
Researchers at the Indian Institute of Technology Madras have developed a new way to watch these insulators, using a single strand of standard glass fiber to act as a continuous, sensitive ear and thermometer along the entire length of the equipment. Instead of attaching dozens of separate sensors, they wrapped a standard optical fiber around the test setup and used it to listen for the faint sounds of electrical sparks and feel the subtle rise in temperature caused by the damage. This approach allows them to see exactly where a problem starts and how it grows, turning the fiber itself into a long, unbroken line of virtual sensors. By listening and feeling at the same time, the team could catch the very first signs of trouble long before the insulator was in danger of failing.
In their experiments, the team subjected polymer insulator samples to a rigorous test that mimics years of wear and tear in a matter of hours. They applied a high voltage and flowed a conductive, salty solution over the surface to encourage the formation of electrical sparks. As the insulator began to degrade, the fiber optic system recorded everything. It captured the moment the first tiny sparks appeared, the brief pause when dry patches formed on the surface, and the eventual development of bright, arcing spots that signaled serious damage. The system recorded 1,750 of these discharge events, tracking the entire journey from a single spark to a permanent, carbonized path that could no longer hold back the electricity.
What made this detection possible was the ability to hear the specific sound of the sparks amidst the background noise of the laboratory. The electrical discharges created a distinct acoustic signature, a sharp, high-pitched sound centered around 14.8 kilohertz. The researchers used a special signal-processing technique to filter out the low-frequency rumble of the environment, making these sharp, transient sounds stand out clearly. This allowed them to pinpoint exactly when and where a spark occurred. They confirmed the accuracy of their listening by comparing the acoustic signals with measurements of the electrical current flowing through the insulator. The timing matched perfectly: every time the electrical current spiked, the fiber heard a corresponding acoustic burst. Furthermore, by moving the distance between the spark and the fiber, they proved that the sound traveled through the air to reach the sensor, confirming that the system was truly listening to the discharge and not just picking up random vibrations.
While the acoustic sensor acted as a fast listener, catching the immediate sparks, a second part of the system acted as a slow, steady thermometer. This part of the fiber measured the temperature along its length, detecting the gradual heat buildup that occurs when sparks persist in one spot. The researchers found that the areas where the acoustic activity was most intense were the exact same spots where the temperature began to rise. This dual view provided a complete picture: the acoustic data showed the rapid, violent nature of the electrical discharge, while the temperature data revealed the slower, cumulative damage that leads to permanent failure. Together, they offered a comprehensive view of the insulator's health that neither method could provide alone.
The study demonstrates that this combined approach can locate multiple problems along a single fiber and distinguish between different stages of damage. Unlike older methods that might only tell an engineer that a problem exists somewhere on a line, this system can identify the precise location of the fault. It is also immune to the electromagnetic interference that often confuses traditional electronic sensors, making it ideal for the noisy environment of a high-voltage power station. The researchers tested this on a fiber that was over a long distance, proving that the technology can scale to monitor vast distances without needing to install new, dedicated sensing networks. By using the same type of fiber already found in many power lines for communication, this method offers a practical, non-invasive way to keep the grid safe. The findings suggest that by listening to the sounds and feeling the heat of the insulators, utilities can detect degradation early, allowing for maintenance before a failure occurs, and ensuring a more reliable flow of electricity.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.