Research on Fast Detection and Identification of Short Circuit Fault Based on Voltage in Multi-level Low Voltage AC System

Inhalt:
The existing fast or early detection and identification methods of short-circuit faults in low-voltage systems are all based on the sharp increase in current after a short-circuit to extract characteristics of faults. It has the problem of misjudgment caused by the interference of the noise of the power distribution system and the non-fault operation state such as load starting or switching. According to the characteristics that the terminal voltage of the low-voltage AC system, which is the voltage at the end of each branch, almost suddenly drops to zero when the short-circuit fault occurs, and remains near zero until the fault is eliminated, the wavelet packet detail decomposition method is introduced to extract the fault characteristics of voltage. And in order to eliminate the influence of line noise, load start, etc. on voltage signal interference, a new method for fast detection of voltage short-circuit faults is formed, which is a combination of voltage wavelet packet detail decomposition method and terminal voltage instantaneous value method. A real simulation model of single-phase short circuit fault in multi-level low voltage AC system is established. The waveforms of the faulty terminal voltage in the full phase angle range are analyzed, and the principle and technical realization of the new method for fast detection of voltage short-circuit faults are explained. In addition, disturbances such as start-up of loads of different capacities are analyzed, and the threshold value of the new method for fast detection of voltage short-circuit faults is studied. Simulation and experimental analysis not only verify the effectiveness and rapidity of the early detection of short-circuit voltage, but also ensure the reliability of the existing rapid or early detection methods of short-circuit current.