Patent Application
20250198917
The present application is related to and claims the priority benefit of German Patent Application No. 10 2023 135 103.6, filed on Dec. 14, 2023, the entire contents of which are incorporated herein by reference.
The present disclosure relates to a spectrometric measuring method for determining at least one measurement variable of a liquid medium and to a measuring device that is designed to carry out the measuring method.
Measuring devices for performing spectrometric measuring methods are used in a number of different types of applications, for example in wastewater treatment plants, in laundries, and in plants carrying drinking water, to measure different measurement variables.
These measuring devices usually comprise a spectrometric sensor having a spectrometric unit that, during measurement operation, receives measurement radiation resulting from interactions with the medium, such as absorption, reflection, and/or scattering, and uses the measurement radiation to determine measurement spectra, each of which reproduce spectral values for the measurement radiation in a predetermined measured wavelength range. These measurement spectra are provided to an evaluation apparatus that uses the measurement spectra to determine measured values for at least one measurement variable, such as a concentration of an analyte in the medium, for example a nitrite content and/or a nitrate content of the medium, a chemical or biological oxygen demand of the medium, and/or a degree of turbidity of the medium.
A number of applications have the problem that interferences that may occur in the medium at the application location, such as air or gas bubbles and/or larger particles, temporarily impair the quality of the measurement spectra. Air or gas bubbles can occur, for example, if compressed air cleaning is carried out at or near the spectrometric sensor. In addition, they can also be caused by turbulent flows of the medium flowing past the spectrometric sensor. The occurrence of air or gas bubbles usually has an immediate effect on the measurement spectra. This in turn results in a corresponding falsification of the measured values determined based upon the measurement spectra. Accordingly, certain measured values may sometimes have significant measurement errors as a result of measurement spectra affected by interferences. Similarly, impurities in the medium, such as macroscopic particles, such as small stones and/or lumps of sand, can also impair the quality of the measurement spectra and lead to correspondingly erroneous measurement values. Erroneous measured values are particularly problematic if the measured values are used to monitor, control, and/or regulate a property of the medium, a plant, and/or a process, such as a manufacturing process.
What is particularly problematic in this respect is that interferences such as air or gas bubbles and macroscopic particles naturally occur at unpredictable times and in each case only last for a short amount of time. In contrast to interfering analytes permanently contained in the medium, whose influence on the measurement spectra that interferes with the measurement can be at least partially compensated for by subtracting a reference spectrum of the interfering analyte, it is not readily possible to correspondingly compensate for disturbing influences that occur sporadically at unknown times for a short amount of time.
The adverse effects of interferences occurring in the medium on the measured value quality can be reduced, for example, by a method described in the German patent application filed on Dec. 17, 2022 with the file reference DE 102022130510.4. This method makes it possible to identify measurement spectra whose quality is impaired by interferences in the medium as outliers. This offers the advantage that the measurement spectra identified as outliers can be discarded and thus incorrect measurement values caused by outliers can be avoided. The disadvantage, however, is that no reliable measured values can be determined during the periods in which measurement spectra identified as outliers were recorded.
Another way to address this problem is to reduce the number of possible interferences by technical measures, such as bubble traps or particle filters, and/or a suitable selection of spectrometric sensor installation locations. However, this does not eliminate interferences in the medium and their adverse effects on the availability of reliable measured values.
JP 2012 093352 A describes a device having an optical or electrical bubble detector, which can be used to detect any bubbles that may occur in a liquid medium flowing through a supply line to an analyzer. When bubbles occur, the medium is supplied to a bubble removal channel via a switch, where the bubbles are removed by a bubble removal device. However, the use of corresponding devices is complex and associated with additional costs.
One object of the present disclosure is to provide a spectrometric measuring method that makes it possible to extend the measuring periods in which reliable measured values can be determined despite interferences occurring sporadically in the medium for limited periods of time.
For this purpose, the present disclosure comprises a spectrometric measuring method for measuring at least one measurement variable for a liquid medium in which interferences can occur, which include air bubbles, gas bubbles and/or particles whose dimensions are larger than wavelengths in a predetermined measured wavelength range, wherein:
The measuring method offers the advantage that all measurement spectra during the recording of which no optical saturation occurred are provided for determining the measured values for the measurement variable(s). The verification method offers the advantage that any interferences that may have occurred in the medium when recording the measurement spectra are detected thereby and their influence on the measurement spectra is accordingly compensated for by subtracting the offset. The latter offers the advantage that high measurement accuracy is achieved not only in the absence of interferences, but also in the presence of interferences.
This results in an extension of the measurement periods in which reliable measured values can be determined despite interferences occurring sporadically in the medium for limited periods. These measurement periods are only interrupted by those periods in which the effects of the interferences occurring in the medium are so significant that optical saturation occurs.
Another development is that the measured wavelength range comprises wavelengths in the ultraviolet range, the visual range and/or the near-infrared range, and/or wavelengths in a wavelength range from 190 nm to 2500 nm or in a section of this wavelength range, and
A first variant provides that
Developments of the first variant consist in the fact that
According to a further development of the first variant, the reference value expected in the reference wavelength is determined in advance on the basis of spectral values for reference spectra recorded in the reference wavelength by the spectrometric sensor in the absence of interferences.
A second variant consists in the fact that
Developments of the second variant consist in the fact that
According to a further development of the first variant, the reference value expected in the absence of interferences in the reference wavelength is determined in advance using spectral values ascertained by extrapolation for reference spectra recorded by the spectrometric sensor in the absence of interferences in the reference wavelength.
A third variant provides that
A development of the third variant consists in the fact that
According to at least one embodiment, measured values for a concentration of an analyte contained in the medium, a nitrite content of the medium, a nitrate content of the medium, a chemical oxygen demand of the medium, a biological oxygen demand of the medium, a color of the medium, an organic carbon content of the medium, an organic carbon content of the medium dissolved in the medium and/or a degree of turbidity of the medium are determined on the basis of spectral values for the medium spectra.
Furthermore, the present disclosure comprises a measuring device for performing the measuring method according to the present disclosure, which:
According to a further development, the measuring device comprises an output apparatus connected to the evaluation apparatus, wherein the output apparatus:
Furthermore, the present disclosure comprises a computer program comprising computer-readable program code elements that, when executed on a computer, prompt the computer to use the measurement spectra determined by the spectrometric sensor to carry out the method steps of the measuring method that are carried out in the measuring method according to the present disclosure on the basis of the measurement spectra or at least to determine the medium spectra and to provide the medium spectra and/or the measured values of the or each measurement variable.
Furthermore, the present disclosure also comprises a computer program product comprising the above-mentioned computer program and at least one computer-readable medium on which at least the computer program is stored.
The present disclosure and its advantages will now be explained in detail using the figures in the drawing, which show several examples of embodiments. The same elements are indicated by the same reference numbers in the figures.
The present disclosure comprises a spectrometric measuring method for determining at least one measurement variable of a medium, as well as a measuring device for performing this method. A flowchart for the measuring method is shown in
As shown in
In the exemplary embodiment shown in
The spectrometric unit 9 shown here by way of example comprises a detector 13 that receives the measurement radiation S and a measuring electronics unit 15 connected to the detector 13, which determines the measurement spectra A(λ) using the radiation intensities of the measurement radiation S measured by the detector 13 at different wavelengths λi. The measurement spectra A(λ) can be determined, for example, as intensity spectra provided by the measuring electronics unit 15, for example in the form of raw digital or analog spectrometric signals, which comprise pairs of spectral values corresponding to the intensity Igem(λi) measured for the particular wavelength λi and the associated wavelength λi. Alternatively, the measuring electronics unit 15 is designed to determine the measurement spectra A(λ) as absorption spectra. In this case, the spectral values a(λi) of the measurement spectra A(λ), which spectral values each occur for a specific wavelength λi and are formed as absorption values, are, for example, each determined in accordance with: a(λi):=−Log [Igem(λi)/I0(λi)] as the logarithm of the ratio of the radiation intensity I0(λi) entering the medium 3 at this wavelength λi to the intensity Igem(λi) of the measurement radiation S impinging on the detector 13 measured by the detector 13 at this wavelength λi.
The present disclosure is not limited to the spectrometric sensor 1 shown in
In the example shown in
The measuring method described in detail below, and of course analogously the measuring device 100 performing the measuring method, can be used in a number of different applications in which interferences such as air bubbles, gas bubbles or macroscopic particles can sporadically occur in the medium 3. Examples include breweries, water monitoring facilities, wastewater treatment plants, such as municipal wastewater treatment plants or industrial wastewater treatment plants used in certain industries, laundries, or plants carrying drinking water. In these applications, the measurement method shown in
The measured wavelength range Δλm of the spectrometric sensor 1, 1′ comprises, depending on the type of measurement variable(s) to be measured, e.g. wavelengths in the ultraviolet range, in the visual range and/or in the near-infrared range, such as wavelengths in a wavelength range from 190 nm to 2500 nm or in a section of this wavelength range.
Interferences occurring in the medium 3, such as air or gas bubbles and/or macroscopic particles, usually have dimensions in the millimeter range or at least in the micrometer range, which are significantly larger than the wavelengths λi in the measured wavelength range Δλm. In this respect, macroscopic particles are understood to be solid bodies, such as small stones or lumps of sand, whose dimensions are larger than the dimensions of solids suspended in the medium 3, which may be responsible for the turbidity of the medium 3 and are generally microscopically small.
The measuring method shown in
In a second method step V2, the continuously determined measurement spectra A(λ) are used to identify and discard those measurement spectra A(λ) during the recording of which an optical saturation OS of the spectrometric sensor 1, 1′ has occurred.
In conjunction with measurement spectra A(λ) in the form of absorption spectra, optical saturation occurs if the absorption occurring along the optical path 11 exceeds an upper limit value for measurable absorption values, above which further additional absorption no longer results in an increase in the measurement spectral values a(λ) that can be resolved using measurements. In conjunction with measurement spectra A(λ) in the form of intensity spectra, optical saturation occurs if the intensity of the measurement radiation S becomes so low that the measurement spectral values formed as intensity values in this case fall below a lower limit value for measurable intensity values, below which a further additional decrease in intensity no longer causes a reduction in the measured intensity values that can be resolved using measurements. In both cases, the particular limit value is a constant or even wavelength-dependent property of the spectrometric sensor 1, 1′, which can in each case be measured experimentally or determined numerically for a wide variety of spectrometric sensors.
Accordingly, in conjunction with measurement spectra A(λ) designed as intensity spectra, for example, the second method step V2 proceeds such that the occurrence of optical saturation OS is established if at least one or the smallest of the spectral values of the particular measurement spectrum A(λ) falls below the associated lower limit value. Analogously, in conjunction with measurement spectra A(λ) in the form of absorption spectra, for example, the second method step V2 proceeds such that the occurrence of optical saturation OS is established if at least one or the largest of the spectral values a(λi) of the particular measurement spectrum A(λ) exceeds the associated upper limit value.
In order to be able to determine reliable measured values mv of the or each measurement variable both in the presence and in the absence of interferences, in a third method step V3 a medium spectrum Am(λ), which represents the spectral properties of the medium 3, is determined for at least one or each measurement spectrum A(λ) remaining after the second method step V2 has been carried out.
The third method step V3 is based on the consideration that in each measurement spectrum A(λ), a plurality of effects are superimposed and each measurement spectrum A(λ) can be described, at least approximately, as a sum or as a weighted sum of subspectra assigned to the individual effects. These subspectra comprise a first subspectrum T1(λ) corresponding to the wavelength-dependent absorption in the medium 3, and possibly also a second subspectrum T2(λ) corresponding to the scattering caused by the turbidity of the medium 3. In the presence of an interference in the medium 3, such as at least one air bubble entering the optical path 11 in the medium 3, at least one gas bubble entering the optical path 11 in the medium 3, and/or at least one macroscopic particle entering the optical path 11 in the medium 3, the subspectra additionally comprise a third subspectrum T3(λ) caused by the disturbance. Examples of absorption spectra for all three subspectra T1(λ), T2(λ), T3(λ) are shown in
As can be seen from
The spectral progression of the third subspectrum T3(λ) caused by the interference is dependent on the interactions between the radiation and the particular interference. These interactions substantially consist in some of the radiation being absorbed in the interference and/or some of the radiation being reflected and/or scattered by the interference.
Gaseous interferences have a significantly lower optical density than the liquid medium
3. Accordingly, the partial length of the optical path 11 passing through gaseous interferences, such as air bubbles and/or gas bubbles, is usually too short for the radiation component absorbed in these interferences to be able to lead to the third subspectrum T3(λ) being substantially dependent on the wavelength. On the premise that the gaseous interferences have dimensions that are significantly larger than the wavelengths λi in the measured wavelength range Δλm, the radiation component reflected or scattered at these interferences does not lead to the third subspectrum T3(λ) being substantially dependent on the wavelength either.
Interferences in the form of solids, such as macroscopic particles, have a significantly higher optical density than the liquid medium 3. This leads to some of the radiation within the entire measured wavelength range Δλm impinging on these interferences along the optical path 11 being almost completely absorbed and some reflected and/or scattered. The almost complete absorption in the entire measured wavelength range Δλmprovides an at least approximately constant contribution to the third subspectrum T3(λ) within the entire measured wavelength range Δλm. On the premise that the macroscopic particles have dimensions that are significantly larger than the wavelengths Aj in the measured wavelength range Δλm, the radiation component scattered and/or reflected by these interferences does not lead to the third subspectrum T3(λ) being substantially dependent on the wavelength either.
Accordingly, the third subspectrum T3(λ) has a spectral progression that is at least approximately equal to a constant offset ΔA for interferences that include air bubbles, gas bubbles and/or macroscopic particles entering the beam path 11 in the medium 3, the dimensions of which are larger than the wavelengths λi in the measured wavelength range Δλm.
The medium spectra Am(λ) representing the spectral properties of the medium 3 can be described at least approximately as the sum of the first subspectrum T1(λ) and the second subspectrum T2(λ). The third subspectrum T3(λ) is additionally superimposed by this sum spectrum caused by the medium 3 in the measurement spectra A(λ) in the presence of interferences, which third subspectrum, on the premise that the interferences have dimensions that are significantly larger than the wavelengths λi in the measured wavelength range Δλm, corresponds at least approximately to a constant offset ΔA.
Based on this knowledge, in a first sub-step V3.1 in which the determination of the associated medium spectrum Am(λ) is carried out in the third method step V3 for the or each remaining measurement spectrum A(λ), a verification method is carried out in which the particular measurement spectrum A(λ) and a previously determined reference for a spectrum property expected in the absence of interferences within a predetermined spectral range comprising at least one reference wavelength λri are used to verify whether or not the particular measurement spectrum A(λ) contains an offset ΔAcaused by interference occurring in the medium 3 while recording the corresponding measurement spectrum A(λ). If an offset ΔA exists, its size is preferably simultaneously also determined in the first sub-step V3.1, e.g. in the form of an offset value.
In the verification method, a range in which the absorption coefficient of the medium 3 varies by less than a predetermined spectral variance depending on the wavelength and/or is below a predetermined limit value is used as the predetermined spectral range.
The spectral variance is calculated, for example, using the sum of the squares of the differences between the values for the absorption coefficient in the wavelengths within the spectral range and the mean of these values within the spectral range.
As can be seen from the first subspectrum T1(λ) shown in
Accordingly, a spectral range that comprises at least one wavelength in the visual range, such as at least one wavelength in the range from 380 nm to 800 nm, and/or at least one wavelength in the near-infrared range, such as at least one wavelength in the range from 800 nm to 2500 nm, is particularly suitable as the predetermined spectral range. Where the predetermined spectral range lies in relation to the measured wavelength range Δλm depends on the size of the measured wavelength range Δλm.
A suitable spectral range is, for example, a range that comprises at least one reference wavelength Δλr1 that lies within a subrange ΔλR of the measured wavelength range Δλm, a range corresponding to a subrange ΔλR of the measured wavelength range Δλm and/or a range comprising at least one reference wavelength Δλr2 that lies outside the measured wavelength range Δλm. Each subrange ΔλR of the measured wavelength range Δλm is, depending on the size of the measured wavelength range Δλm, for example a long-wave edge wavelength range, a short-wave edge wavelength range or a middle wavelength subrange of the measured wavelength range Δλm.
In conjunction with predetermined spectral ranges comprising at least one single reference wavelength λr1, λr2, the reference for the spectrum property expected in the absence of interferences comprises, for example, a reference value r(λr1), r(λr2) for the spectral value a (λr1), a(λr2) expected in the absence of interferences for the corresponding reference wavelength λr1, λr1.
In conjunction with the measured wavelength range ΔΔλm shown as an example in
In conjunction with measured wavelength ranges Δλm, which comprise wavelengths in the visual and/or near-infrared range, the subrange ΔλR is, for example, a long-wave edge wavelength range, a short-wave edge wavelength range or a middle wavelength range of the measured wavelength range ΔΔm, depending on the size of the measured wavelength ranges Δλm. In these cases, a wavelength greater than or equal to 380 nm, greater than or equal to 500 nm, greater than or equal to 800 nm, or greater than or equal to 830 nm is suitable as the reference wavelength λr1 that lies within the measured wavelength range Δλm, depending on the size of the measured wavelength range Δλm.
Because the reference wavelength λr1 lies within the measured wavelength range Δλr1, the reference value r(λr1) expected in the absence of interferences in the reference wavelength λr1 is, for example, determined in advance using spectral values a(λr1) for reference spectra recorded by the spectrometric sensor 1, 1′ in the absence of interferences in the reference wavelength λr1.
In this exemplary embodiment, method step V3.1 proceeds such that a difference between the spectral value a(λr1) of the particular measurement spectrum A(λ) at the reference wavelength λr1 and the reference value r(λr1) and the existence of an offset ΔA is established if the difference is greater than a predetermined tolerance. The tolerance is, for example, a tolerance that is predetermined on the basis of the sensor-specific measurement accuracy of the spectral values a(λ) for the measurement spectra A(λ). In this embodiment, the size of the offset ΔA is given by the value of the difference and is determined according to: ΔA:=a(λr1)−r(λr1), for example.
The embodiment shown in
In conjunction with measured wavelength ranges Δλm that comprise wavelengths λi in the near-infrared range, the reference wavelength λr2 that lies outside the measurement wavelength range Δλm is, for example, much smaller than the wavelengths λi in the measured wavelength range Δλm. In this respect, a suitable reference wavelength λr2 that lies outside the measured wavelength range Δλm for these measured wavelength ranges Δλm is, for example, a wavelength in the visual range, such as a wavelength greater than or equal to 380 nm or even greater than or equal to 500 nm.
In the exemplary embodiment shown in
In this exemplary embodiment, the reference value r(λr2) expected in the absence of interferences in the reference wavelength λr2 is, for example, determined in advance using spectral values a(λr2) established by extrapolation for reference spectra recorded by the spectrometric sensor 1, 1′in the absence of interferences in the reference wavelength λr2.
Analogous to the previously described embodiment, in this embodiment, too, the difference between the spectral value a(λr2)—established here by extrapolation using the measurement spectrum A(λ)—for the reference wavelength λr2 and the reference value r(λr2) is determined and the existence of an offset ΔA is established if the difference is greater than the predetermined tolerance. The size of the offset ΔA is given by the value of the difference and is calculated according to: ΔA:=a(λr2)−r(λr2), for example.
This embodiment is advantageous in particular in applications where the spectral values a(λ) for measurement spectra A(λ) recorded in the absence of interferences in the edge range ΔλRR of the measured wavelength range Aam facing the reference wavelength λr2 that lies outside the measured wavelength Δλm can be described, at least approximately, by a straight line.
Depending on the size of the measured wavelength range Δλm, the subrange ΔλR is also formed in this case, for example, in the manner previously described in connection with the exemplary embodiment shown in
In conjunction with measured wavelength ranges Δλm, which comprise wavelengths in the visual and/or infrared range, the subrange ΔλR is, depending on the size of the measured wavelength range Δλm, for example a long-wave edge wavelength range, a short-wave edge wavelength range or a middle wavelength subrange of the measured wavelength range Δλm and/or a range comprising wavelengths greater than or equal to 380 nm, greater than or equal to 500 nm, greater than or equal to 800 nm, or greater than or equal to 830 nm.
Since the subrange ΔλR lies within the measured wavelength range Δλm, the reference progression R(λ) expected in the absence of interferences in this wavelength range is determined in advance, for example on the basis of the spectral progressions of the spectral values for reference spectra recorded by the spectrometric sensor 1, 1′ in the absence of interferences in the subrange ΔλR.
In the exemplary embodiment shown in
Analogous to the previously described exemplary embodiments, this embodiment also proceeds in such a way that the existence of an offset ΔA is established when the value Umin of the offset variable U that minimizes the difference is greater than the predetermined tolerance. The size of the offset ΔA here is given by the value Umin of the offset variable U that minimizes the difference and is determined, for example, according to: ΔA:=Umin.
This embodiment is particularly advantageous in applications in which the spectral values a(λ) of measurement spectra A(λ) recorded in the absence of interferences within the subrange ΔλR of the measured wavelength range Alm have a certain degree of non-linear wavelength dependence.
Following the verification method carried out in the first sub-step V3.1, the medium spectrum Am(λ) is determined for each verified measurement spectrum A(λ). As shown in
Lastly, in method step V4, the spectral values of the medium spectra Am(λ) determined in this way are used to determine measured values mv for the measurement variable(s) to be measured, such as a concentration of an analyte in the medium 3, a nitrite content of the medium 3, a nitrate content of the medium 3, a chemical oxygen demand of the medium 3, a biological oxygen demand of the medium 3, a color of the medium 3, an organic carbon content of the medium 3, an organic carbon content of the medium 3 dissolved in the medium 3 and/or the turbidity of the medium 3.
To determine the measured values mv, depending on the measurement variable, methods known from the prior art are used, for example, by which the measured values of the particular measurement variable are determined using measurement spectra in the prior art. For example, measured values mv for the turbidity of the medium 3 are determined using a specific integral total value that is determined on the basis of the spectral values in the entire measured wavelength range Δλm or using spectral values for the medium spectra Am(λ) that occur where there are one or more specific wavelengths λ. Other methods known from the prior art comprise, inter alia, MOD calculation rules, also referred to among experts as chemometric models, with which measured values mv for measurement variables, such as the concentration of an analyte in the medium 3, the nitrite content, the nitrate content, the organic carbon content, the organic carbon content dissolved in the medium 3, as well as the chemical and/or biological oxygen demand of the medium 3 can be calculated using the spectral values for measurement spectra. Since the medium spectra Am(λ) are equal to the measurement spectra A(λ) in the absence of interferences and correspond to measurement spectra A(λ) compensated for with respect to the interference in the presence of interferences, these methods can be easily adopted from the prior art and applied in the fourth method step V4.
The measuring method shown in
In measuring devices, such as the measuring device 100 shown in
The signal processing apparatus 5a comprises, for example, a data processing device, such as a computer or a microprocessor, and a computer program stored in a memory and executable the data processing device, by which the medium spectra Am(λ) are determined using the measurement spectra A(λ). Analogously, the measured value determination apparatus 5b comprises, for example, a data processing device, such as a computer or a microprocessor, and a computer program stored in a memory and executable by the data processing device, by which the measured values mv are determined using the medium spectra Am(λ).
The measured values mv can each be read out, output and/or transmitted to a unit at a higher level than the measuring apparatus 100, such as a control room, a process control system, a distributed control system or a programmable logic controller, wirelessly and/or via wires, for example in the form of data or signals via an interface 19 of an output apparatus 21 of the measuring device 100 connected to the evaluation apparatus 5. Alternatively or additionally, the output apparatus 21 comprises, for example, a display device 23, such as a display, for displaying the measured values mv and/or a display element 25, such as a light-emitting diode, which indicates when the existence of an offset ΔA has been established using the verification method and/or which indicates whether or not the measurement spectrum A(λ), on the basis of which the particular measured value mv was determined, contains an offset ΔA.
| Number | Date | Country | Kind |
|---|---|---|---|
| 10 2023 135 103.6 | Dec 2023 | DE | national |
