A recursive state estimation approach to mitigate false data injection attacks in power systems

摘要

Estimating the power system states is essential for monitoring, protection and control functions. Recent studies have shown that the traditional bad data detection techniques can be bypassed by an adversary. One such scenario is the false data injection attack, where the adversary can compromise the measurement devices, to create a false set of measurements, that cannot be detected by the conventional bad data detection algorithms. The power system loads normally follow a regular time pattern and their time history, can be, utilized to provide accurate information to the estimator. In this case, the historical data has been combined with the current measurement set recursively to generate more accurate estimates, which in turn, can detect the false data attack in power systems. IEEE 14-bus test system is used as the test case for the study. The proposed method has been tested for two attack scenarios, where an adversary compromises certain meters or where the current measurements is replaced using a previous measurement set.