Skip to Main Content
The paper deals with developing and testing frequency-adaptive PMU algorithms with wider linearity range than specified in IEEE Std C37.118-1. This goal is achieved by means of three different concepts encompassing robust state-of-the-art design approaches: 1) FIR bandpass filtering, 2) extended Kalman filtering (EKF), and 3) discrete Fourier transform (DFT) demodulation with FIR low-pass smoothing. While FIR-based PMUs are linear phase with no overshoot in either phase or amplitude step responses, the adaptive EKF PMU is more computer-intensive but allows for a reduced group delay and better out-of-band interference rejection at the cost of a phase step response with overshoot. Frequency measurement performances of the various PMUs are assessed in detail. It turned out that FIR PMUs are best for meeting Std C37-118-1 metrics but they are outperformed by EKF under changing harmonics. Test results on three recent commercial PMU models further confirm that PMU algorithms meeting standard C37-118-1 can behave quite differently under dynamic conditions.