A Covariance Consistent Data Fusion method for Power Networks with Multirate Sensors

The quasi-static assumption has been often employed in the analysis of a power system for state moni-toring/estimation. Accordingly, only static state estimates can be obtained. However, the increased penetration of renewable generation, especially photovoltaic generators and wind farms, introduces significant variability in the system and challenges this quasi-static assumption. Thus, it is crucial to extract information about the system dynamic states, which influences the stability of the system. Therefore, dynamic schemes for state estimation, particularly Kalman filtering, have been introduced for the power systems to perform dynamic state estimation. However, power systems usually have a low degree of instrumentation, which renders it necessary to exploit all the available information and measurements in a power network. There are different sources of measurements in distribution grids, such as SCADA and Phasor Measurement Units (PMUs). These sensors, however, provide different rates of data. Hence, multi-rate data fusion is required in a power system containing different types of sensors. Considering the demonstrated consistency with the covariance intersection method (CI), we propose an unscented Kalman filter (UKF)-based CI data fusion approach to fuse the estimates based on sensors with different data rates. This method is then compared to an existing multi-rate data fusion algorithm for power systems. The results show that the proposed dynamic approach is effective and provides robust state estimates for the power systems.