A Fast and Adaptive Fisher Discriminant Analysis Framework for Robust Fault Detection in HVDC Transmission Lines

Since high-voltage direct current (HVDC) systems are among the most crucial ones utilized in connecting nations, they are more likely to failure, which could result in major losses in DC transmission systems, causing extensive damage to electrical equipment and substantial financial losses. In order to provide uninterrupted service continuity and safeguard equipment and devices from potential harm, it was required to develop faster and more accurate methods for detecting faults in DC power transmission systems. A novel method for fault detection in DC power transmission systems is presented in this research. Since adaptive discriminant Fischer analysis (AFDA) is one of the best techniques for fault detection, the suggested method is based on using AFDA and compares its performance with Fischer discriminant analysis (FDA). The approach proposed is based on obtaining DC current signals from the IRS transmitter's current sensor and extracting 500 samples, which are used to train the suggested method. No matter how the system's operating conditions change, the suggested technique for this study has a high degree of fault detection ability because it can identify both external faults that arise in AC networks on the sending or receiving side as well as internal faults of the DC transmission system. The simulation of the DC transmission system was done using PSCAD software, and different scenarios of faults were implemented and used to determine the efficiency of the proposed methodology. The suggested approach was later adopted, and the data obtained were visualized using the MATLAB software. The findings reveal that the proposed methodology is more efficient and quicker in the identification of faults while minimizing false alarms.