Abstract
Nowadays due to the modern wave of intellectualization of measuring instruments the digital methods of correction not only for primary measuring transducers, but also for secondary sensors are practiced on a large scale. Since the conversion function of secondary measuring transducers is linear, the method of model measures by two points is used for automatic correction of systematic components of the errors of such converters. The method of digital compensation, which provides a more significant reduction in the errors of measuring transducers compared with the method of analog compensation has been described. Features and technical indicators of this method are considered on the example of measuring pressure transducer with foil strain gauges. This method is universal, allows to adjust not only the errors of measurement channel nonlinearity and additional errors, but also the errors associated with the effect of interferences of general type due to ground resistance, which induces the connection between measuring channels of the main and destabilizing factor. The disadvantages of this method include a significant amount of computations, which sharply increases with the increase of the order of approximating polynomials. The aim of the paper is to develop a method and means of measuring stress-strain state using strain gauge, free from the above mentioned shortcomings. The main destabilizing factors that limit the measurement accuracy using strain gauge are the following: - random processes (noises, obstacles, etc.); - changes in parameters of measuring transducers due to aging and physical degradation; - effects of external climatic and mechanical factors (temperature, humidity, etc.). Regarding systematic components, the errors of nonlinearity and temperature component of the error are the most important in statistical measurements. We have studied the influence of temperature range variations, the spread of values of temperature error (± 10 %) on the mean square error of approximation by power polynomials. Using the NUMERY Program, the dependence of approximation error on the order of approximating polynomial has been determined. It is established that in a wide temperature range, the error for constantan has a weak relation with the order of a polynomial.
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call