DC fault detection and classification using time-derivative of butterworth highpass filtered currents without threshold

• Eliminates threshold setting by selecting sensitivity above the transient. • Fault classification using polarity asymmetry driven by cutoff frequency. • Concurrent fault detection and classification in a unified step. • Validation demonstrated by both simulation and hardware prototype. DC transmission is efficient and renewable-friendly, but high-frequency capacitive discharge during faults necessitates fast and reliable fault detection and classification. Therefore, this paper presents a method based on the time derivative of a Butterworth Highpass Filter (HPF), validated through simulation and a hardware prototype. The HPF isolates the high-frequency components of measured pole currents for fault detection, while an optimised cutoff frequency enhances fault classification. The time derivative further accelerates detection by tracking change rates rather than peak magnitude. The Butterworth response is monotonic, stable, and comparable. Thus, both poles’ derivatives are normalised with identical sensitivity above the charging transient, avoiding predefined thresholds. Comparing these normalised derivatives enables simultaneous fault detection and classification.