Optimized Robust Adaptive Networks in Supervisory Control and Data Acquisition Systems

This technology enables power grid stability regardless of network background traffic.

The Electric Power and Communica - tion Synchronizing Simulator (EPOCHS) system could provide great benefits to private industry and Department of Defense (DoD) power systems infrastructure by allowing simultaneous, synchronous simulation of communication and power system simulators to better understand the power grid under anomalous situations. Supervisory Control and Data Acquisition (SCADA) systems and the power grid infrastructure have recently been the focus of attention given the recent disclosure of potential cyber attacks. Applying Special Protection Schemes like the one used by the EPOCHS systems could provide a solution not only to cyber attack, but to possible cascading failures typical to power systems on current high-load demands. Utilizing an Internet-like architecture to implement such a backup and communication system is likely to be the most logical solution to providing a power grid intranet.

This work investigates the application of an Exponential Weighted Moving Average (EWMA) to be used as a memory buffer in conjunction with Special Protection Schemes (SPS) using EPOCHS. It is proposed that using an SPS incorporating EWMA can compensate for the network layer lack of guarantee of packet delivery, and provide for the stability and integrity of the power grid under a catastrophic event.

The work determined that the EWMA can successfully be applied to the SPS in EPOCHS, increasing its reliability under all simulated network background conditions. Using EWMA to create a buffer provides the capability to retain some information of the previous data, while still retaining information of current state of the system. However, using EWMA for disturbance size calculation introduces some error and thus steady state after disturbance is not comparable with the case not using EWMA. Regardless of the error introduced, in all cases the error is considered to be acceptable and is within threshold value to be considered the normal operation state of a power grid.

The SPS with EWMA algorithm provides statistically significant performance gains over the previously used SPS algorithm. The results indicate that the proposed EWMA SPS ensures the protection of the grid. The EWMA SPS has a significant impact on performance when applied to a heavy background traffic network without router reservation, enabling it to be stable without the additional hardware cost. Overall, in the tested configuration, the new SPS system successfully maintained steady-state operation under all traffic intensities.

The largest performance gains are seen when applying SPS with EWMA while heavy background traffic and no router reservation is present. In this scenario, the SPS with EWMA improves the minimum frequency from 58.35 Hz to 59.08 Hz while the system is in transient or anomalous state, and stabilizes to normal operating state of 60.35 Hz. This effect demonstrates that the utility intranet could be establish using standard UDP protocols to transmit while avoiding expensive router reservation schemes deployment. In all cases, statistical analysis of the injection of the EWMA to the SPS metrics found that it provided for a quick reaction to system disturbances while minimizing errors in steady-state stabilization.

This work was done by Luis A. Oquendo Class of the Air Force Institute of Technology. For more information, download the Technical Support Package (free white paper) at www.defensetechbriefs.com/tsp under the Information Sciences category. AFRL-0139

This Brief includes a Technical Support Package (TSP).

Optimized Robust Adaptive Networks in Supervisory Control and Data Acquisition Systems (reference AFRL-0139) is currently available for download from the TSP library.

Please Login at the top of the page to download.