On-demand state estimation with sampling time skew in power systems

With new dynamics and uncertainties in today's power grids, traditional fixed-interval State Estimation (SE) may be unable to track the variability and monitor the power grid effectively. This paper presents a new architecture to transfer data and execute SE on demand. A list of situations are summarized to direct the SE-demand generator in system control center. As SCADA and PMU measurements are co-exist in realistic power systems, time skew problem is inevitable. To mitigate the influence of time skew, a state estimator based on time skew oriented weight adaptation is considered. In each SE circle, the weights assigned to the measurements not only correspond to their noise, but also the time offsets relative to the SE-demand point. Numerical examples demonstrate the improved accuracy of our estimator compared with the conventional hybrid SE when measurements time skew is present.