A Hybrid Fault Event Detection Algorithm Using Fault Recorder Data

Abstract

When fault occurs in power grid, a lot of alarms and records are generated in substations. These are secondary circuit signals, fault recorder and PMU data. Fault recorder data contains the details of fault evolution, which can be used for diagnosis fault occurrence time, circuit breaker operation events. It is crucial for power grid diagnosis and intelligent alarm functions. However, the analysis and utility of fault record data for diagnosis purpose require the operator's participate and is time consuming. In this paper we design a two stages fault diagnosis mode for future development of IEC61850 and digital substation. Based on this, we propose a hybrid algorithm to analyze fault recorder data with abrupt change detection which can obtain major events during a fault, including fault occurrence, circuit breaker operation time. We use a simulated example to prove correctness and efficiency of our proposed algorithm. The algorithm proposed in this paper can be used as intelligent alarm functions for future power grid fault diagnosis.