Development of methods of processing sensor signal

Abstract

The object of research is the process of processing the signal of the primary converter by a certain method. In carrying out measurements on the primary converter of the measuring system, in addition to the action parameter measurement, there are many secondary influences. One of them is the change in the temperature of the environment. This leads to the appearance in the measured signal of such a component as an additional constant level. It is caused by the generation of the primary converter in the absence of influence on the object of measurement of additional charge, potential. That is, in the absence of action on the side of the measurement parameter there is a shift in the zero level of the signal of the primary converter. Thus the fluctuations of the temperature of the medium result in the appearance of a non-informative component in the measurement signal. That is, we get an increase in the error of the measurement result. The principle of the method used to process the signal of the primary converter is presented on the example of measuring the vibration with a piezoelectric accelerometer. The proposed method of processing the signal of measurement allows to distinguish a constant component of the complex signal of measurement. This is achieved by supplying a measuring signal to two parallel channels, in one of which the signal is delayed by phase for half of the period relative to the second, followed by processing on the adder. Since the signal from the primary converter has a wide frequency band, it is necessary to ensure the accuracy of the delay in the phase. This is achieved by reconfiguring a constant time phase filter phase. The reconfiguration is carried out using a frequency-voltage converter that tracks the frequency of the measured signal. Since changes in temperature over time compared with the frequency of measured vibrations are much slower, then the inaccuracy of the scheme is practically absent. That is, the inaccuracy of determining the constant component in the complex signal will be determined by the difference in the amplitudes of adjacent half-lives. The obtained results allow to assert that this method can be used both for correction of the primary transformer itself (by influence on its transfer function), and for correction of the results of measurements in general.