Comparison of Frequency Measurement Methods Under Transient Conditions in Electrical Power Networks

The signals of modern electrical power networks have a large number of features that must be taken into account when performing frequency measurements. The most common deviations from the sinusoidal shape include: the presence of signal harmonics, frequency deviation relative to the nominal value, the presence of random noise (usually uniform or normal distribution density), flicker (sinusoidal or rectangular shapes). In addition to the above-mentioned features, different types of transient processes occur in electrical networks. All this leads to errors in frequency measurement and it seems relevant to search for a measurement method that would have the best metrological characteristics. The paper discusses the currently most popular digital measurement methods: the signal zero crossing method, the signal phase increment method, the amplitude spectrum approximation method, the method based on three input signal samples. A comparison has been made of these methods in terms of the achieved error for the case of sinusoidal and polyharmonic input signals, as well as under conditions of transient processes occurring in the electrical power network. The comparison results were obtained by simulation in Matlab package. For each of the considered methods, modifications have been proposed to reduce the final measurement error of power frequency.