RU2782885C1 - Method for adjusting a differential transducer - Google Patents

Abstract

FIELD: measuring equipment.

SUBSTANCE: method relates to measuring equipment and can be used when setting the transducers in analogue feedback systems of automatic process machines. Method consists in setting the reference point of the transducer, followed by grading the output signal thereof. The method is characterised by predetermining the probability distribution of the parameter measured by the transducer over the entire operating range thereof; then selecting the most probable value of the parameter and resetting the reference point of the transducer to this value.

EFFECT: ensured operation of the method with a minimum absolute error (maximum precision) and the highest probability.

1 cl, 2 dwg

Description

The present invention relates to the field of measuring technology and can be used when setting up measuring transducers in analog feedback systems of technological automatic machines.

Currently, methods for adjusting measuring transducers in feedback systems, and in general in measuring technology, are known. So, in particular, incremental displacement transducers such as pulse sensors used in CNC machines, etc. are adjusted so that their origin coincides with the middle of the pulse (displacement quantum) (see, for example, the book "E.I. Gitis, E.A. Piskulov. Analog-to-digital converters. M .: Energoizdat. 1981").

In analog systems, adjustment is done differently. In such systems, measuring transducers are used not incremental, but differential (absolute). Their adjustment is carried out with the help of the reference offset present in them, by setting (adjusting) the reference point to zero values of the measured parameter, followed by calibration of its output signal. (See the site "en.ppt-online.org. Characteristics of the simultaneous impact of additive and multiplicative errors on the transform function"). It is this method of adjustment adopted as a prototype.

The existing adjustment method is quite simple, however, it has a significant drawback. It proceeds from the idea of a uniform distribution of the probabilities of the measured parameter over the entire working measurement range, which in reality is not always the case.

As a rule, any analog measuring transducer with a nominal characteristic Y = f (X), where X is the measured parameter, and Y is the output signal of the transducer, produces a signal with errors.

These errors are divided into two types: additive and multiplicative. The first, or transducer reference error, is usually constant along the entire length of the transducer characteristic and has some value α. The second one is proportional to the value of Y. As a result, the absolute value of the total error of the signal Y is equal to:

,

,

where в is the relative multiplicative error.

Since Y = f (X), then Δ depends on X, and each value of X in the working measurement range has its own value of Δ.

By setting the reference point of the converter to the zero value of the measured parameter, the value of Δ is made minimal, since Y in this case is equal to zero. With an increase in X, and hence Y, Δ increases, but this is inevitable, and with a uniform distribution of the probabilities of the measured parameter X, all possible values of Δ are equally probable. But, if any value of X has a greater probability than other values, then the converter outputs a Y signal with the corresponding error more often than with other Δ. With repeated measurements of the X parameter, which takes more probable values in the operating range, it will turn out that the converter is not working accurately with a high probability. Meanwhile, this is not always acceptable. It is often allowed that the converter sometimes still outputs a Y signal with large ∆, but in general it is required that it work as accurately as possible, with the smallest ∆.

In accordance with the foregoing, the problem to be solved using the proposed method is precisely that the converter, adjusted by the prototype method, works with non-minimum possible errors with high probability.

Technically, the solution to this problem is provided due to the fact that the method of adjusting the differential measuring transducer, including setting (setting) the reference point and subsequent calibration of the output signal of the transducer, differs from the prototype in that the probability distribution of the measured parameter is preliminarily determined in the entire operating measurement range, then the the most probable value of the parameter and the reference of the transducer are set (adjusted) to this value. In this case, mechanical displacement or, for example, gas pressure is used as the measured parameter, and electrical voltage or mechanical displacement, or similar physical parameters, respectively, are used as the output signal.

, 4 – график возможной функции плотности вероятности параметра Х, Х_НО – начало отсчета Х, Δ – обозначение зоны распределения суммарной погрешности, соответствующей наиболее вероятному (модальному) значению Х=Х_М. In FIG. 1 shows an illustration of the prototype alignment method. In FIG. 2 illustration of the proposed method. On both figures 1 is the nominal characteristic of the converter Y = f (X), 2 is the zone of distribution of the additive error α, 3 is the zone of distribution of the multiplicative error

, 4 - graph of the possible probability density function of the parameter X, X _BUT - the origin of X, Δ - designation of the zone of distribution of the total error corresponding to the most probable (modal) value X=X _M .

Let's show the implementation of the method with an example. Let there be a measuring transducer of mechanical displacement into voltage of a rheostat type with a nominal characteristic:

,

,

where K =5 V/mm is the sensitivity of the transducer.

The working range of the transducer is 0-100 mm. Error b = 0.1 V, and error c = 5%. The output parameter X is distributed according to the normal law with a probability density:

,

,

;where

;

y=15 mm;

When adjusting by the prototype method, the most probable error:

In terms of X, this is 2.52 mm.

то градуировка выходного сигнала сводится к введению поправки на

и, соответственно, на

. То есть, выходной сигнал Y преобразователя будет соответствовать

, откуда When adjusted by the proposed method, D=b=0.1 V, which in terms of X is 0.02 mm. Since the latter is obtained by setting (adjusting) the reference point to

then the calibration of the output signal is reduced to the introduction of a correction for

and, accordingly, on

. That is, the output signal Y of the converter will correspond to

, where

. Характеристика

или

будет являться результатом градуировки.

. Characteristic

or

will be the result of the calibration.

Another example of the implementation of the method. There is a gas pressure converter into the angular mechanical movement of the pointer (barometer), with a nominal characteristic

,

,

where K is the transducer sensitivity equal to 0.1 bar/degree.

. Operating range of the barometer -2 bar…+4 bar. Error b = 0.01 deg, and error c = 3%. The output parameter is also distributed according to the normal law, but with

.

When adjusting by the prototype method, the most probable error is:

which in terms of X will be 0.19 bar.

и градуировки выходного сигнала преобразователя получим характеристику последнего Y=

, откуда

. When adjusted by the proposed method, D=b=0.01 deg, which in terms of X is 0.1 bar. After setting the origin to

and calibration of the output signal of the transducer, we obtain the characteristic of the last Y=

, where

.

As can be seen from the above, the technical result of applying the proposed method is to ensure its operation with the minimum absolute error (maximum accuracy) with the highest probability.

Claims (1)

A method for adjusting a differential measuring transducer, including setting a reference point and subsequent calibration of the output signal of the transducer, characterized in that the probability distribution of the measured parameter is preliminarily determined over the entire measurement range, then the most probable value of the parameter is selected and the reference point of the transducer is set to this value.

Images