Sub-Nyquist rate ADC sampling in digital relays and PMUs: Advantages and challenges

Data acquisition unit is the basic building block of any power system protection and monitoring device. Operating principle of data acquisition unit heavily depends on the Nyquist theory. As per Nyquist theorem, the sampling rate of analog-to-digital converter (ADC) should be greater than or equal to the twice of signal bandwidth. Situation of sub-Nyquist rate ADC sampling can arise in relays/PMUs if there is no anti-aliasing filter or if the frequency specification of anti-aliasing filter violates the Nyquist theorem. Traditionally, sub-Nyquist rate sampling has been considered unusable due to aliasing. Digital relays and PMUs always use anti-aliasing filter before ADC to band-limit the analog signals. Compressive Sampling (CS) theory has opened up a new possibility of signal reconstruction from sub-Nyquist rate samples if the signal is `sparse' in nature. This paper discusses the possible use of sub-Nyquist rate ADC sampling in digital relays and PMUs. This concept is comparatively new in the area of power system signal processing. Both advantages and implementation challenges of sub-Nyquist rate sampling are discussed in the paper. Results are also presented to demonstrate the phasor estimation performance with sub-Nyquist rate ADC sampling.