identification, has been written to use real-time switch status and current and voltage measurements to determine the present system operating condition. A system dispatch package [15], using linear programming techniques, determines if the actual load conditions warrant rescheduling of the experiments. If the actual load is in excess of the scheduled load then experiments are delayed. If the actual load is less than the scheduled load then experiments may be rearranged to reduce the test time. If load has to be shed then a priority order is used to weight each experiment. The load models, used to predict the power required for each experiment, may be updated by exponential smoothing techniques. The power system dynamic simulator is presently used to evaluate system abnormal conditions. The results will be used to determine the current and voltage waveforms that a microprocessor-based relay system would have to recognize to detect abnormal conditions. Relay algorithms, based on digital signal processing techniques, have been written and tested for various abnormal conditions. Present research is under way to detect incipient faults such that preventive action can be taken. Once these results are available, a more general package will be written to identify the appropriate relay settings. The instrumentation model parameter update uses state estimator results to update the conversion curves which translate the sensor readings to actual units. Summary This paper outlines a power management system that is presently being assembled and tested based on terrestrial power system technology. Extensive experience has shown that such techniques are easily implemented and respond quickly to protect equipment and to maintain system performance. The proposed implementation based on microprocessor technology would minimize the demands on space, weight and operational power. The proposed hardware implementation, based on distributed embedded controllers, would not require the use of complex operating systems (e.g. multitasking) and would result in a highly redundant, and therefore secure, system. ACKNOWLEDGEMENTS This research was supported by the Center for the Commercial Development of Space Power with funds from NASA Grant NAGW-1192-CCDS-AL, Auburn University, and the Center’s Industrial Partners. REFERENCES [1] Stott, B., Alsac, O. & Monticelli, A.J. (1987) Security analysis and optimization, Proceedings of IEEE, December, p. 1623. [2] Gaushell, D.J. & Darlington, H.T. (1987) Supervisory control and data acquisition, Proceedings of IEEE, December, p. 1645. [3] Thorp, J.S., Phadke, A.G. & Karimi, KJ. (1985) Real time voltage-phasor measurements for static state estimation, paper 85 WM 082-3 presented at the IEEE/PES 1985 Winter Meeting, New York, NY, 3-8 February. [4] Athay, T.M. (1987) Generation scheduling and control, Proceedings of IEEE, December, p. 1592. [5] Bose, A. & Clements, K.A. (1987) Real-time modeling of power networks, Proceedings of IEEE, December, p. 1607.
RkJQdWJsaXNoZXIy MTU5NjU0Mg==