How to locate a serial arc fault in a domestic electrical network?

Significance 

Electrical faults are the leading cause of fire in homes and even in some industries. In particular, series arc fault that occurs when damaged electrical wires come in contact with each other is considered the main cause of electrical faults. Unlike other faults, series arc fault is more difficult to detect in series configuration due to its random complex behaviors since the current level is limited by the impedance load rather than by the fault itself. Despite the availability of different protection systems, there are no single circuit breakers with the ability to locate series arc faults when the line is operational. Besides, series arc fault detection methods are based on reflectometry that requires disconnection of the power supply before insertion of the signals. Nevertheless, these methods require impedance matching between the load and the line to prevent undesirable reflections and the complexity involved in designing the special sensor, which remains the biggest challenge and hindrance to their practical applications.

Lately, the potential use of algorithms has drawn significant research attention. To date, several algorithms for locating parallel faults have been developed. Despite the good progress, little effort is devoted to locating faults on short low-voltage lines. The available detection methods have several challenges associated with the difficulty of using mesh Kirchhoff equations, high sensitivity to noise and the need for precise synchronization of the measured current. Consequently, the accuracy of the existing algorithms for locating AC series-arc faults is significantly influenced by the dynamic variation of the fault and adequate frequency band selection that degrade their performance. Therefore, developing effective algorithms for locating series arc faults on short power lines without disconnecting the power supply load is highly desirable.

To address the aforementioned challenges, Mr. Edwin Calderon-Mendoza, Professor Patrick Schweitzer, Dr. Hien Duc Vu and Professor Serge Weber from the University of Lorraine developed a new and more robust series arc fault detection method for short low voltage lines. The new algorithm was based on virtual decoupling of the mutual capacitance of the line achieved by the interaction of the impedance parameters and fault map trace generated by signature coefficients calculated from Kirchhoff’s equations considering the hypothetical fault distances. The method was tested using different short line lengths (30 m – 200 m) supplied by AC 230V 50 Hz network source with inductive loads and different series arc fault data. The original research article is now published in the International Journal of Electrical Power and Energy Systems.

The research team findings showed that the generation of fault-map traces at different arc-fault locations allowed the creation of a baseline that acted as the main reference points. As a result, the unknown fault distances were calculated from the intersection of the linear trend curve (the coefficients of the fault trace map) and the coefficients obtained by subtracting the estimated currents. The unknown fault distances were precisely and accurately estimated for different line lengths, line models and inductance-load values. Besides, the results showed its ability to effectively analyze electrical faults induced by either carbonized paths or opening contacts. It was worth noting that the feasibility of this approach was only tested on an electrical network with a single line, and the task will be more difficult for complex networks with lateral branches that require placing a locating device in all the branches.

In summary, the authors presented an innovative and more robust method for locating series arc faults in short low voltage AC power networks. The algorithm performance and feasibility were successfully validated for different lines lengths with different inductive loads. In a statement to Advances in Engineering, Professor Serge Weber, the lead author said the promising results of series arc fault location method is a potential candidate for efficient health monitoring of low-voltage power networks to locate damaged cables and prevent potential fire outbreaks and other related accidents.

Reference

Calderon-Mendoza, E., Schweitzer, P., Vu, H., & Weber, S. (2021). Series arc fault location algorithm based on impedance parameters and fault map trace generationInternational Journal of Electrical Power & Energy Systems, 130, 106652.

Go To International Journal of Electrical Power & Energy Systems

Check Also

28.3% Efficient III–V Tandem Solar Cells Fabricated Using a Triple-Chamber Hydride Vapor Phase Epitaxy System - Advances in Engineering

28.3% Efficient III–V Tandem Solar Cells Fabricated Using a Triple-Chamber Hydride Vapor Phase Epitaxy System