WO2008113375A1

WO2008113375A1 - Method and device for measuring flow velocities

Info

Publication number: WO2008113375A1
Application number: PCT/EP2007/002358
Authority: WO
Inventors: Kurt Dirk Bettenhausen; Wolfgang Ens; Herbert Grieb; Peter Hauptmann; Robert Schäfer
Original assignee: Siemens Aktiengesellschaft
Priority date: 2007-03-16
Filing date: 2007-03-16
Publication date: 2008-09-25
Also published as: EP2137541A1

Abstract

The invention relates to a method and a device operating according to said method for measuring flow velocities. A basic calibration is performed independently of the device, by means of which coefficients are determined that represent an influence of variables to which the device is subjected during measurement, for example, properties of the flowing medium and flow velocity. For operating the device, a device-specific operational calibration is performed, wherein a device-specific transmission factor is determined. Thus, during operation of the device, no further calibration, even when changing the medium, is necessary, due to the device-independent coefficients and the device-specific transmission factors being known.

Description

description

Method and device for measuring flow velocities

The invention relates to a method and a device operating according to the method for measuring flow velocities and as such a device, in particular a so-called hot-wire anemometer. Specifically, the invention relates to a method of calibrating such a device as well as to an apparatus in which such a calibration is provided. After calibration, both the method and the device work with the calibration found, so that insofar as the invention as a whole also relates, as stated above, to a method and a device for measuring flow velocities which operates according to the method.

Hot wire anemometers are known per se. Therein, a hot wire is arranged as a heating element whose temperature-dependent electrical resistance or its thermally induced change in length is used to measure physical variables, in particular mass flow, flow velocity, etc. Mass flow and flow rate are used in the present application as synonymous terms. However, it is clear to the person skilled in the art that the mass flow over the density and the flow cross section is related to the flow velocity. For calibration and operation of an anemometer, therefore, a uniform choice of the respective relevant physical quantity mass flow or flow velocity is required.

In the above-mentioned devices, ie in particular hot-wire anemometers, the amount of heat dissipated by convection depends on a flowing medium - hereinafter also referred to as material - both its flow velocity and its material properties, in particular specific heat capacity, density, viscosity, etc., from. A precise mass Current measurement accordingly requires a suitable calibration for such material properties. This is laborious because until now an individual calibration was required for each substance to be measured. In addition, calibration data in tables or the like is available only for individual, known substances; for unknown connections, such values must be determined. At each change of a material to be detected by the device, the calibration data must also be updated.

Accordingly, it is an object of the invention to provide a method of operating a device of the above mentioned type and a device operating in accordance with the method, with which the abovementioned disadvantages are avoided or reduced in terms of their effects.

This object is achieved according to the invention by the features of claim 1, insofar as a calibration referred to below as basic calibration is concerned. For this purpose, in a method for basic calibration of a device for measuring flow velocities by evaluating the effect of convection of a flowing medium on a device covered by, fed by a power source hot-wire provided that in the context of the method, the device, eg as a Mass flow measuring device acting hot-wire anemometer, with a plurality of known with respect to their material properties x (s) media and by each medium with a plurality of predetermined or predetermined flow velocities m is applied. Furthermore, for each medium and flow velocity m, the hot wire is heated in accordance with a predetermined or predeterminable heating curve and an associated electrical impedance Z _e , (s) determined from a supply current I (s) and a voltage U (s) across the hot wire. The system responses obtained in this way allow a statement on the effect of certain substances at certain flow velocities on the respective convection at the heat wire and thus allow a calibration of the device with respect to device-independent, ie in particular anemometer independent, substance-specific influencing factors. Accordingly, this first form of calibration is also referred to as base calibration. For this purpose, a sufficiently determined equation system is formed on the basis of the parameters resulting for the majority of the media, flow velocities m and the measured values for the electrical impedance Z _el (s), and a coefficient matrix A is determined therefrom. The basic calibration supplies data in the form of a coefficient matrix A which, in brief, comprises a description of the substance properties previously stored in tables or the like for a plurality of different substances, ie the calibration instrument-independent calibration data.

A particular advantage of the invention with regard to the basic calibration is that the description of material properties available in the coefficient matrix A is no longer individual substances but characteristics (characteristic curves) which depend on the convection and the individual Represent material properties, acts. Thus, in principle, even in a priori unknown substances can be measured exactly without the need for a renewed calibration with respect to the respective substance; the characteristic curves described by the coefficient matrix A and the model coefficients encompassed thereby permit an "interpolation" when a substance is investigated which, with regard to its substance properties, is to be located "between" those substances which are used in the determination of the coefficient matrix A. were.

The above object is achieved according to the invention with the features of claim 3, as far as a calibration referred to below as operating calibration is concerned. For this purpose, in a method for operating calibration, a method according to the method outlined above and further described below is already (pre-) calibrated and to that extent a device provided with a coefficient matrix embossed in a device known per se in a manner known per se, eg a hot wire anemometer acting as a mass flow meter, that a relationship between a metrologically determined electrical impedance Z _el (s) and an operational thermal impedance Z _therm {s) of the hot wire is modeled by a transmission factor φ (s). For operating calibration, the device is then subjected to a predetermined or predeterminable flow rate m with a medium known with regard to its material properties x (s). Thus, from a supply current I (s) and a falling over the hot wire voltage U (s), the electrical impedance Z _el (s) and the electrical impedance Z _el (s) and the known parameters material properties x (s) flow velocity m of Transmission factor φ (s) are determined. The transfer factor φ {s) describes device-dependent, ie in particular anemometer-dependent, influencing factors on the measurement result. The calibration of the operation is very fast and easy to carry out at any time on site. The calibration of the operation presupposes comparatively complex basic calibration, which, however, has to be carried out only once for a particular group of substances and not in relation to a specific measuring device or a group of measuring devices.

The results of the basic calibration can be used universally for any measuring device and the operating calibration provides, as it were, the adaptation of the basic calibration to the respective measuring device.

With the coefficient matrix A and the transmission factor φ (s) calibrated devices, ie in particular anemometer, are now ready for use at a specific site. The coefficient matrix A also permits, without recalibration, a measurement of previously non-empirically detected substances, provided that such substance properties are detected by "interpolation" of the characteristics described by the coefficient matrix A. Transmission factor φ {s) takes into account the special features of individual measuring devices, eg the length of the hot wire, which always varies at least to a small extent between different measuring devices of the same series.

Advantages and details emerge from the subclaims. This used backlinks point to the further development of the subject matter of the main claim by the features of the respective subclaim; they should not be construed as a waiver of obtaining independent, objective protection for the feature combinations of the dependent claims. Furthermore, with a view to an interpretation of the claims in a closer specification of a feature in a subordinate claim, it is to be assumed that such a restriction does not exist in the respective preceding claims.

According to a preferred embodiment, thereafter e.g. provided that the coefficient matrix A with a numerical estimation method, e.g. Ordinary Least Squares (OLS),

Principle Component Regression (PCR), Partial Least Squares (PLS), etc., is determined. In practical use, the determination of the flow velocity m is based on the transmission factor φ (s) determined in the context of the operating calibration.

Hereinafter, an embodiment of the invention will be explained in more detail with reference to the drawing, which is not to be understood as limiting the invention. Rather, numerous modifications and variations are possible within the scope of the present disclosure, in particular those variants and combinations, for example, by combination or modification of individual described in conjunction with the general or specific description part and in the claims and / or the drawing Characteristics or elements or method steps for the expert with regard to the solution of the problem can be removed and by combinable features to a new object or new Procedural steps or procedural steps lead, even if they concern testing and work procedures. Corresponding objects or elements are provided in all figures with the same reference numerals.

Show in it

1 shows a schematically simplified schematic representation of a hot wire anemometer and Figure 2 shows another, also schematically simplified representation of a hot wire anemometer.

1 shows a schematically simplified schematic representation of a hot wire anemometer 10 as an example of a device for measuring flow velocities or mass flows. The hot wire anemometer 10 comprises a hot wire 12 (FIG. 2) which is fed by a current source 14 with a current I (s) (the transformer shown in FIG. 1, also designated by the reference numeral 12, is not with the hot wire 12, but merely represents its electrical behavior as a function of the thermal impedance acting: from the electrical side, ie in the sense of an equivalent circuit diagram, the hot wire 12 thus corresponds more to the totality of thermal impedance transducers. During operation, a voltage U (s) drops across the hot wire 12. With these two variables, an electrical impedance Z _el (s) of the hot wire 12 can be determined.

FIG 2 shows the hot wire anemometer 10 with some further details. Thereafter, the hot wire 12 is arranged in a volume, for example a tube 16. The tube 16 is in operation of a medium, not shown separately but only illustrated by the block arrow 18, flows through. The medium also passes through the hot wire 12. According to respective material properties x (s) of the medium, eg heat capacity, etc., the hot wire 12 releases heat generated by the flow of current to the medium. The resulting temperature drop influenced can be expressed in the resistance of the Hitz ^¬ wire 12 above the wire Hitz ^¬ by the quotient of 12 sloping in a known manner voltage U (s) and supply current I (s).

According to FIG 2, the hot wire anemometer 10 includes a STEU ^¬ ergerät 20, which in turn comprises a power source 14 (FIG 1) comprising the power unit 22, a processing device, in particular a processor 24, or the like, and a memory 26th The power unit 22 includes not only the power source 14 but also known means for measuring the falling over the hot wire 12 voltage U (s).

With a hot-wire anemometer 10, as already mentioned above, the amount of heat removed by convection from the hot-wire 12 to the mass flow measurement is transferred to a medium, e.g. a gas flowing past the hot-wire, evaluated. The dissipated heat quantity depends both on a flow velocity m of the medium and on its material properties x (s) (specific heat capacity, density, viscosity, etc.). Precise mass flow measurement requires calibration for the properties of the material.

For calibrating the hot wire anemometer 10, a base calibration and an operating calibration are provided, wherein only the operating calibration is specific to the respective hot wire anemometer 10 and the base calibration is basically usable for a wide variety of hot wire anemometers 10.

In the following, the basic calibration will be described first. For this purpose, it is provided that a basically arbitrary hot-wire anemometer 10 is subjected in succession to a plurality of media known with regard to their material properties x (s) and through each medium to a multiplicity of predetermined or predefinable flow velocities m. In addition, for each medium and flow rate m, the hot wire 12 becomes one predetermined or predetermined heating curve (not shown) heated and an associated electrical impedance Z _el (s) from the supply current I (s) and the falling over the hot wire 12 voltage U (s) determined. The heating curve can be stored in the memory 26 and is "driven off" by appropriate variation of the supply current I (s) by the power unit 22 under the control of the processor 24. On the basis of the majority of the media, flow velocities m and the measured values for the electrical impedance Z _el (s) resulting parameters can be set up a sufficiently specific system of equations, based on which a coefficient matrix A can be determined.

Around

Z = 7 = φ ² (a - m + p - x) = [ab] = A - ξ [1]

[Ia]

[Ib]

In equation [1] Z denotes the electrical impedance Z _d (s), U the voltage U (s) dropping across the hot wire 12 and / or the supply current I (s). The dynamics of these and other variables taken into account by introducing the complex frequency s has been omitted from equation [1] for reasons of clarity. Further, in equation [1] m, the (frequency-independent) flow velocity and x denote

Material properties X (^) = [X ₀ (J ¹⁾ •• ^■ JC _j (, s).] ^R, where m and x elements of ξ are (Note: bold symbols indicate vectors or matrices), a and ß are (initially unknown) coefficients describing the influence of the flow velocity m and the material properties x (s) on the electrical impedance Z _el (s), φ finally describes as φ {s) an (initially unknown) relationship between a acting thermal impedance Z _ιherm (s) = Z _ιherm (rh, x (s)) and the metrologically determined electrical impedance Z _el (s). This relationship exclusively models the properties of the hot wire 12 and is therefore independent of the respective medium.

The part multiplied by the part of the equation [1] denoted by [Ia] yields the part designated by [Ib], where the sizes are written as matrices. The multidimensionality of the respective variables thus expressed results, for example, from the fact that for the base calibration, as described above, a plurality of known media and each medium having a plurality of predetermined or predeterminable flow velocities rh are successively selected with respect to their material properties x (s) is used. Under this assumption, the row vector [ab] becomes a matrix hereinafter referred to as the coefficient matrix A. This contains the coefficients a or β used to describe the influence of the different flow velocities rh and material properties x (s).

With the aid of a reference set, S = [^ ₀ ••• ξ _N ], which covers a specified range of values for both the mass flow rh and the material properties x (s), and the respective measured electrical impedance Z = [Z ₀ ••• Z _N \ can be the coefficients of the matrix A as

A = Z (Ξ ^T Ξ) - ¹ Ξ ^T [2]

to be appreciated.

In addition to the Ordinary Least Squares (OLS) represented by Equation [2], alternatively, Principle Component Regression (PCR) and Partial Least Squares (PLS) are suitable as adequate estimators. The respective choice depends on the numerical design of Ξ (rank, characteristic of eigenvalues, etc.). Since the model established by equation [1] separates the properties of sensor (hot wire 12) and medium, the model coefficients, ie the coefficients of matrix A, with known φ (s) within one calibration to be performed - the base calibration - can be done with a large one Number of nodes are determined very accurately. This calibration is then valid for all other hot wire anemometers.

Production-related variances are determined by individual φ (s) determined as part of an operating calibration. In each case a φ (s) is determined for each measuring device. For this purpose, the measuring device is acted upon with a reference medium, that is to say a medium with known material properties x (s), with a predetermined or predeterminable flow velocity m. It is important here that both m and x (s) correspond in value to at least one element of the matrix Ξ used for the basic calibration (the operating calibration can also be carried out for several elements of Ξ, but at least for one element of Ξ) , This ensures that the measured change in the electrical impedance, Z, during operation relative to the measured electrical impedance, Z _ref , in the basic calibration is due exclusively to a changed transmission behavior of the device (anemometer), since both measurements at the same mass flow and same material properties were performed. The change in the transmission behavior manifests itself in a transmission factor φ {s) in the present case in the case of operating calibration compared to a transmission factor φ _ref (s) in the case of basic calibration. Since a relative change of φ is to be detected, by definition φ _ref (s) = 1 can be set. Since the term (αr-zw + β-x) from equation [1] is the same for both the base calibration and the operating calibration (same defined mass flow, same defined substance), the device-specific transmission factor and thus the result the operating calibration according to equation [3].

In operation, from the known coefficients of the estimated coefficient matrix A and known transmission factor φ (s), finally, a relationship for determining the flow velocity m can also be derived from the left-hand part of equation [1]: m = * [4] a

Equation [4] applies to any substances, insofar as the material properties x (s) lie within the range of the base calibration covered by A (interpolation). This opens up the advantage of this method, since a recalibration is not required in this case.

On the basis of equation [4], it is thus possible with a measuring device of the type described at the outset, if the estimated coefficient matrix A is impressed on it, e.g. by being stored in the memory 26, and if the transmission factor φ {s) is determined by means of an operating calibration with a reference medium, for fundamentally any media, provided that their properties depend on the substance properties x (x) which are used in the determination of the coefficient matrix A. s), which calculate the respective flow velocity and / or the respective mass flow.

Based on the approach of the invention, an improved mass flow measurement, i. a method and a device operating according to such a method possible. For this purpose, the control unit 20 with processor 24 and memory 26 was shown schematically simplified in FIG. The functionality of the device results from a software implementation of the described method according to the invention or a preferred embodiment which may be stored in the memory 26. By accessing the memory 26 and the software implementation of the method stored there, the control unit 20 executes the method by means of the processor 24.

Thus, the present invention as a whole can be briefly described as follows: The invention relates to a method and a device for measuring flow velocities, wherein a device-independent basic calibration is carried out, with which coefficients are determined that have an influence of variables, With where the device is exposed during the measurement, such as characteristics of the flowing medium and flow velocity, and wherein a device-specific operating calibration is performed to operate the device, with a device-specific transmission factor is determined, wherein in operation of the device due to the reputation of the device-independent coefficients and of the device-specific transmission factor, no further calibration is necessary, even if the medium is changed.

Claims

claims

1. A method for base calibration of a device for measuring flow velocities by evaluating the effect of convection of a flowing medium on a covered by the device, by a power source (14) fed hot wire (12), wherein the device with a plurality of respect to their material properties x (s) known media and by each medium with a plurality of predetermined or predetermined flow velocities m is applied, heated for each medium and each flow rate m of the hot wire (12) corresponding to a predetermined or predetermined heating curve and an associated electrical impedance Z _e , ( s) is determined from a supply current I (s) and a voltage U (s) dropping across the hot wire (12), and where, for the plurality of media, flow velocities m and the measured values for the electrical impedance Z _el (s) resulting parameters a sufficiently determined Gleichun gssystem formed and from a coefficient matrix A is determined.

2. The method of claim 1, wherein the coefficient matrix A is determined by a numerical estimation method.

3. A method for operating calibration of a calibrated according to the method of claim 1 or claim 2 device with an impressed coefficient matrix A, wherein a relationship between a metrologically determined electrical impedance Z _e! (s) and a thermal impedance Z _therm (s) of the hot wire (12) acting in operation is modeled by a transmission factor φ (s), the device having a medium known with regard to its material properties x (s) with a predetermined or predeterminable flow velocity m is applied, wherein from a supply current I (s) and one above the Hot wire (12) falling voltage U (s), the electrical impedance Z _el (s) is determined and with the electrical impedance Z _el (s) and the known parameters material properties x (s) and flow velocity m, the transmission factor φ (s) is determined ,

4. A method for operating a calibrated according to the method according to claim 3 or 4 device, wherein for determining the flow rate m of the determined in the context of the operating calibration transmission factor φ (s) is used.

A computer program comprising computer executable program code instructions for implementing the method of any one of claims 1, 2, 3 or 4 when the computer program is run on a computer.

6. computer program product, in particular storage medium, with a computer-executable computer program according to claim 5.

7. A device for measuring flow velocities by evaluating the effect of convection of a flowing medium on a by a current source (14) with a supply current I (s) fed hot wire (12), with a control device (20) for measuring an over the hot wire ( 12) decreasing voltage U (s) and either a computer program executed or executable by the control unit (20), in particular a processing unit, according to claim 5 or a computer program product readable by the control unit (20) according to claim 6.