Calculation of the Admissible Time Delay of Discrete Samples During Real-Time Measurements

An increase in the accuracy of measurements in automatic systems of operational control and regulation by taking into account the error due to the time delay of a discrete reading of the measured value in real time is considered. It is shown that in dynamic measurements of random processes in servo systems, it is necessary to take into account the subjective factor in determining the dynamics of the measured quantities, which consists in the method of assigning the maximum frequency of the spectrum of the initial processes. When calculating the admissible time delay in obtaining the results of discrete measurements in real time, the extrapolation error was taken as a dynamic error. Analytical expressions are obtained to determine the required efficiency of discrete measurements of continuous quantities, taking into account the admissible extrapolation error and the subjective factor of assigning the maximum frequency of the spectrum of the initial process. Three common models of the measured quantity are considered, in which the energy spectrum and the correlation function of the desired random variable are presented as a response to white noise of the ideal, RC and Gaussian low-pass filters. The admissible time delays of discrete samples are calculated depending on the admissible extrapolation error, the extrapolation method, and the maximum frequency of the spectrum of the measured quantity in real-time measurements. It is shown that with a priori uncertainty about the model of the measured value and extrapolation from one or two samples, it is necessary to use a zero-order extrapolator as a prediction algorithm when calculating the admissible time delay. The scientific results obtained in the article will be of interest to specialists in the field of tracking systems, telemechanics, guidance systems, operational control and regulation during real-time measurements in the return information channel.