Abstract: In solving modern problems of the electric power industry, when measuring the signal frequency, the most common
method is based on determining the zero crossings of a given signal. The advantage of the method is the simplicity of the analysis
and implementation of measurement algorithms, which makes it possible to impose lower requirements on the element base used in
the implementation. The disadvantage of this method is the significant influence of harmonics and noise on the frequency
measurement error. In this article, a new version of the modification of this method for measuring the signal frequency is proposed.
The analysis carried out in the article shows that the proposed modification to reduce the dependence of the metrological
characteristics of the method on the minor spectral components of the measured signal and noise. The essence of the modification
consists in the additional use of another value instead of the zero level, for which the application of the method for signal zero
crossings will be more effective. The signal level is determined by calculating the torque that will correspond to the maximum
derivative of the measured signal, i.e. its maximum tilt angle. The derivative value of the signal is determined using a first-order
digital differentiator. The paper considers options for constructing digital differentiators and their influence on the frequency
measurement error. The moments of signal transition through the level corresponding to the maximum derivative are refined using
piecewise linear approximation of the signal in the vicinity of the transition. In the course of the work, a computer simulation of the
zero-crossing method and its modifications was carried out using the Simulink software package, and analytical expressions were
obtained that characterize the frequency measurement error. As a result of the simulation, a comparison was made between the
metro-logical characteristics of the “classical” zero-crossing method and the proposed modification.
Index terms: signal frequency, interpolation, differentiation, measurement error, simulation modeling.