What is it about?

discrete wavelet transform and deep neural network are utilized for fault location in a series-compensated three-terminal transmission line. Features extracted from synchronous measurements of fault currents at the three terminals using discrete wavelet transform are fed to the deep neural network. Faulted section determination and fault distance calculation are carried on using a single intelligent network simultaneously. Faulted section is determined with 100% accuracy, and the efficiency of algorithm is validated for symmetrical and unsymmetrical faults, and different values of fault resistance, inception angle, and location. The accuracy of the algorithm is acceptable for large fault resistances (above 100 ohm) and fault inception angles near zero. Total mean error for test data is 0.0458% which is much improved with respect to other similar works.

Featured Image

Why is it important?

The series compensation causes challenges in fault location schemes of the three-terminal transmission lines. Fault location in three-terminal lines compensated with series Flexible AC Transmission System (FACTS) has not been investigated, which is a more challenging task than the fault location in uncompensated lines.The error in fault location with conventional neural networks was high, as the network was not able to capture the complex nonlinear nature of the optimization problem. Hence, using deep neural network (DNN) with a special output encoding is proposed which is an efficient mathematical tool for learning complex nonlinear relationships and also has not been utilized for fault location in previous works.

Perspectives

Fault location in series-compensated teed-feeders is considered for the first time. Utilizing Deep Neural Network has improved fault location accuracy greatly compared to previous works. The accuracy of the algorithm is independent of line parameter estimation. The accuracy of the algorithm is independent of series FACTS device modelling. The algorithm performance is evaluated for a great number of fault scenarios. The distributed-parameter Bergeron line model is considered for a more accurate approximation of the dynamic behavior of the real-world system.

Mahdi Mirzaei
Amirkabir University of Technology

Read the Original

This page is a summary of: Fault location on a series-compensated three-terminal transmission line using deep neural networks , IET Science Measurement & Technology, April 2018, the Institution of Engineering and Technology (the IET),
DOI: 10.1049/iet-smt.2018.0036.
You can read the full text:

Read

Contributors

The following have contributed to this page