Assessment of Feeder Voltage Regulation Using Statistical Process Control Methods

Power-quality voltage and current transient waveform data have been explored rather extensively as the primary input data in predictive maintenance, automatic root-cause analysis, and evaluating system performance to indicate potential problems. Unfortunately, very few efforts have been directed toward making use of the voluminous steady-state data collected alongside waveform data. Therefore, this paper proposes to use steady-state data, particularly, rms voltage data to detect abnormal trend behavior that may be indicative of a problem. Specifically, this paper develops a statistical analysis algorithm based on the well-known statistical process control methods for assessing feeder voltage regulation performance. The assessment results can be used to indicate potential regulator problems as well. The efficacy of the method is demonstrated by applications to two sets of actual RMS voltage data.