Abstract

A novel fault-location scheme is designed for transmission lines of high-voltage dc (HVdc) grids equipped with quick-action protections and circuit breakers using very few milliseconds of postfault signals, measurable before the ultrafast fault-isolation stage. A simplified equivalent circuit analysis shows that the first negative overshoot time and average derivative of voltage in the early postfault moments are directly related to the fault location. However, due to HVdc grids' characteristics and several factors involved in reality, the analytical use of these features may not be feasible or may not result in an acceptable accuracy. Hence, a soft computing strategy is adopted to overcome this problem. Based on this strategy, machine-learning-based locators are trained for both line terminals, each responsible for locating faults in the half-line of its side using the two introduced features extracted from the locally captured voltage signals. The statistical evaluations of the performance on a four-terminal grid for various fault cases not considered in the training stage confirm the designed scheme's satisfactory accuracy, generalization ability, tolerance to noisy measurements, and tolerance to changes in the line parameters.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.