Machine Learning (ML) algorithms continually improve the accuracy of the models by automatically learning and improving from the gathered data across the global FiberSense network. This helps to reduce the number of false positives, allowing the model to accurately identify different threats and categorize accordingly, distinguishing between vibrations caused by excavation activity, as opposed to a non-threatening event.
Historical faults have left a legacy of losses and downtime for operators, which Transgrid expects to avoid with its DigitalAsset™ monitoring service. Previously, fault detection and location was a time-consuming and complex process that often required extensive testing and significant repair work, which could further stress the cable insulation. This made it challenging to efficiently identify and fix faults in the system.
With DigitalAsset™, calibration tests are automatically conducted, saving time when locating faults. Unplanned repairs due to faults or damage pose a serious inconvenience to the utility. What’s more, cable faults can eventuate several years after a mechanical strike. With the construction contractor long gone, and no record of past activity, cost recovery becomes almost impossible.