## Abstract

When operating bioreactors, monitoring the process temperature is of great importance. Ultrasonic temperature measurement enables this without the need for sensors inside the bioreactor. The average speed of sound can be determined from the travel time of ultrasonic waves along the path of known length between a transmitter and a receiver positioned in the reactor wall. Due to the temperature dependence of the speed of sound, the average temperature along the path can be calculated from the average speed of sound. If this measurement is carried out for a sufficiently large number of measurement paths in a slice plane of the bioreactor, the temperature field in this plane can be reconstructed.

In this article, a measurement setup for measuring temperature distributions is presented. Two different methods are then used to reconstruct temperature fields from the travel times measured with this setup. The first method is the Tikhonov regularization, which is widely used to solve inverse problems. The second method is derived from compressed sensing and can only be used under certain conditions. However, if these conditions are met, this method always provides better results than Tikhonov regularization. In addition to reconstruction results from measured travel times, results from simulated travel times are also presented.

## Zusammenfassung

Beim Betrieb von Bioreaktoren ist die Überwachung der Prozesstemperatur von großer Bedeutung. Die Verwendung von Ultraschall ermöglicht die Messung der Temperatur ohne die Notwendigkeit von Sensoren im Inneren des Bioreaktors. Stattdessen kann aus der Laufzeit von Schallwellen entlang des Pfads zwischen in der Reaktorwand positionierten Sendern und Empfängern die mittlere Schallgeschwindigkeit bestimmt werden, wenn die Pfadlänge bekannt ist. Wegen der Temperaturabhängigkeit der Schallgeschwindigkeit kann daraus die mittlere Temperatur berechnet werden. Erfolgt diese Messung entlang einer ausreichend großen Zahl an Messpfaden in einer Schnittebene des Messvolumens, kann daraus die Temperaturverteilung in dieser Ebene rekonstruiert werden.

In diesem Beitrag wird ein Messaufbau zur Messung von Temperaturverteilungen vorgestellt. Aus den gemessenen Schalllaufzeiten mit diesem Aufbau werden mit zwei verschiedenen Methoden Temperaturverteilungen rekonstruiert. Die erste Methode ist die zur Lösung inverser Probleme weit verbreitete Tikhonov-Regularisierung, die zweite entstammt dem Gebiet des Compressed Sensing. Diese kann nur unter bestimmten Voraussetzungen eingesetzt werden, führt dann aber stets zu besseren Ergebnissen als die Tikhonov-Regularisierung. Neben Rekonstruktionsergebnissen aus gemessenen Schalllaufzeiten werden auch solche aus simulierten Laufzeitmessungen präsentiert.

## 1 Introduction

In a bioreactor, substance conversions are carried out under controlled conditions and with the use of biocatalysts. In order to achieve the highest possible yield, the conditions inside bioreactors must be monitored and controlled so that they are optimal for the ongoing reactions [1]. In addition to mixing the reactants and monitoring the pH and oxygen level, monitoring the temperature is of particular importance. The temperature can also deviate only locally due to possible exothermic biochemical reactions in the bioreactor. Hotspots can therefore arise. This makes it necessary not only to monitor the temperature at selected points, but to measure the temperature field.

This contribution deals with the ultrasonic measurement of temperature fields and their reconstruction from the measured data. In the following, first the ultrasonic temperature measurement is treated in general. After that, a setup for measuring temperature fields is presented. In order to estimate the temperature field from the data measured with this setup, an inverse problem has to be solved. Section 4 discusses two reconstruction methods used to do so. Subsequently, various reconstruction results obtained with these methods are presented. Finally, a summary and an assessment of the suitability of the presented method for monitoring temperature fields for application in bioreactors is given.

## 2 Ultrasonic temperature measurement

The basis for ultrasonic temperature measurement is the relationship between the speed of sound and the temperature in the medium under consideration. For ideal gases this relationship is given by

where *c* denotes the speed of sound, *T* the thermodynamic temperature in K, *γ* the isentropic expansion factor, *R* the gas constant, and *μ* the molecular mass of the gas [2]. Equation 1 is also valid to a good approximation for many real gases. For liquids, the dependence of the speed of sound on temperature cannot generally be given exactly in a form similar to Eq. 1. An experimentally found dependence *ϑ* denotes the temperature in °C, is therefore usually used for water. Hence, Bilaniuk and Wong [3] have determined the coefficients of the 5th order polynomial

which are given in Tab. 1. This polynomial is plotted in Fig. 1. The plot shows that the speed of sound reaches a maximum at a temperature of 74 °C and decreases again for higher temperatures. Therefore, with the relationship according to Eq. 2, a temperature value can only be unambiguously assigned to a value of the speed of sound below 50.2 °C. Above this value, additional information is required for a case distinction.

coefficient | value | unit |

A basic arrangement of transducers for measuring temperature acoustically is depicted in Fig. 2. The two piezoelectric transducers shown can be used both as transmitters and receivers, since the piezoelectric effect is reciprocal [4]. The travel times of the ultrasonic waves in both directions are given by

in good approximation. The same applies for *l* is known, the average speed of sound along the path can be calculated from

With the relationships from Eqs. 1 and 2, the average temperature can be determined from this [5].

As can be seen from Eq. 5, the mathematical description of the travel time measurement is the integration over the reciprocal of the speed of sound, which depends on the local temperature distribution. For this reason and because of the non-linearity of

If the average speed of sound is measured along several paths in one slice plane, an estimate for the temperature field in this plane can be reconstructed from the results. The following sections deal with this.

## 3 Measurement of temperature fields

In the following, a setup for the measurement of temperature fields is presented first. Subsequently, the determination of the ultrasonic travel times, which is necessary for the temperature measurement, is dealt with.

### 3.1 Setup for measuring temperature fields

In order to be able to measure the average temperature along a sufficiently large number of measurement paths in one plane, several transducers are required, which are positioned at the boundary of the plane under consideration. For the number of transducers, a compromise must be made between the number of measurement paths and the increasing effort. In the following, a setup is considered in which 16 transducers are attached at regular intervals along the circumference of a round, water-filled basin. This results in a total of 120 theoretically usable measurement paths. However, since the used transducers of the type SMFM21F1000 from *Steiner & Martins* with a resonant frequency of 1 MHz have highly directional emission characteristics [7], [8], not all theoretical paths can be used in reality. Figure 3 (a) shows the remaining usable measurement paths for one transmitter. Consequently, there are 56 usable paths for all 16 transducers. In order to be able to use more information, the paths shown as solid lines in Fig. 3 (b) with backscattering on the opposite basin wall are additionally used. This results in 32 additional paths for all transducers and thus in a total of 88. The paths shown as dotted lines provide no additional information since the location of the backscattering is on the surface of another transducer. Therefore, these paths are not considered.

A photo of the basin with 16 glued-in transducers is shown in Fig. 4. The basin is made of acrylic glass and has an inner diameter of 194 mm and a height of 208 mm. The 16 transducers, which are mounted in one slice plane of the basin, must all be able to be used as both transmitters and receivers. For this purpose, all transducers are connected to electronics for switching between transmission and receiving mode. This electronics consists of 16 relays and a 16:1 multiplexer, which are all controlled by a microcontroller. The relays are used to switch between transmission and receiving mode. The multiplexer is used to select the receiver to be connected to an oscilloscope that is used to record the received signals. Although these electronics only allow sequential measurements to be carried out and are not designed for the shortest possible measurement time, this is sufficient under laboratory conditions. However, the measurement time can be significantly reduced by acquiring the received signals from several transducers in parallel. With suitable hardware and depending on the dimensions of the measurement area, measurement times of a few seconds or even less than 1 s can be achieved. The transmission signals are generated using an STHV748 ultrasonic pulser manufactured by *STMicroelectronics*. The transmission signal has a frequency of 1 MHz and consists of 10 rectangular pulses followed by 5 pulses with a lower magnitude that are delayed by half a period [9], [10]. Due to the number of pulses, the ultrasonic wave package has enough energy to ensure reliable detection. The time-delayed pulses lead to a faster decay of the transducer’s oscillation.

The next section deals with the further processing of the received signals to determine the travel times.

### 3.2 Determination of travel times

The received signals from all 88 measurement paths are processed with an FIR bandpass filter. Figure 5 shows the filtered received signals of a direct measurement path at two different temperatures. The shorter travel time at the higher temperature is clearly visible.

To determine the ultrasonic travel times for all 88 measurement paths, the filtered signals are compared with reference signals of known travel times. For this purpose, the cross-correlation of the received signal and the reference signal is calculated for each path. With the shift against 0 of the maximum of the resulting cross-correlation function, the delay of the signal with respect to the reference signal can be estimated [12]. Since the noise that remains in the relevant frequency range around 1 MHz after filtering the signals is approximately white, cross-correlation is a matched filter. Therefore, the use of cross-correlation to determine the travel times yields very good results. The standard deviation of the travel time measurement is less than 1.2 ns for all paths. With the measurement setup described above, this corresponds to a standard deviation of the temperature measurement of less than 5 mK at room temperature.

For the reconstruction dealt with in the next section, the average speed of sound values along all measurement paths are calculated from the travel times determined with cross-correlation according to Eq. 6.

## 4 Reconstruction algorithms

For the reconstruction of the temperature field from the average speed of sound values along all measurement paths, the reconstruction area is discretized. The reconstruction is thus formulated as a general discrete inverse problem. To solve this inverse problem, two methods, Tikhonov regularization and compressed sensing, are used.

### 4.1 Discretization of reconstruction area and formulation as inverse problem

The circular reconstruction area is subdivided into cells of equal area using a method from [13]. Thereby, the reconstruction area is divided into a small inner circle and *N* rings. These rings are then again divided into individual cells. This results in a total of

The inverse problem to be solved for the reconstruction of the distribution of the speed of sound, from which the temperature field is then determined, is

The solution vector *x* contains the values sought for the individual cells of the discretization. Its elements are assigned to the cells according to the designations in Fig. 6. The measurement vector *b* contains the average speed of sound values of all measurement paths weighted with the respective path length. The system matrix *A* connects the two vectors *x* and *b*. Its elements each contain the length of the measuring paths per cell of the discretization. The measurement matrix *A* is thus given by the geometry of the measurement setup.

To solve the inverse problem in Eq. 7, the two methods described in the next two sections are used. The temperature field is then calculated from the reconstructed distribution of the speed of sound. Since the 5th order polynomial in Eq. 2 cannot be solved unequivocally for the temperature, this is done iteratively.

### 4.2 Tikhonov regularization

The difficulty in solving inverse problems is that they are usually ill-posed and therefore even small perturbations in the measurement vector *b* lead to large errors in the solution *x* [14]. Since *b* contains noisy measurement data, these perturbations are always present. Therefore, the naive solution of the inverse problem usually yields very poor results. The Tikhonov regularization tries to counteract this. The Tikhonov solution

The first term *b*. By the term *λ* weights the two terms in Eq. 8 with respect to each other [15]. For *λ*, which is why the best possible value for it has to be found for good reconstruction results.

### 4.3 Compressed sensing

Methods known as compressed sensing [16] can also be used to solve inverse problems given certain a priori knowledge. The condition for using these methods is that the solution vector *x* is sparse or at least compressible or can be transformed into a respective representation [17], [18], [6]. A vector *x* is called sparse if

Minimizing the

A temperature field is typically not sparse or compressible, and neither is the corresponding speed of sound distribution. However, it can be transformed into a compressible representation if it can be assumed that the distribution has only a few hotspots and is otherwise approximately constant. This is often the case in bioreactors. The transformation into a compressible representation is then possible by using the gradient of the solution vector. The

with the remaining error *ϵ*. It specifies the termination condition of the optimization [21].

In contrast to Tikhonov regularization, minimization of the total variation does not have a parameter comparable to *λ* on which the quality of the solution depends significantly. Instead, a priori knowledge is required for total variation minimization so that the gradient of the solution vector can be used for the reconstruction.

## 5 Reconstruction results

In the following, reconstruction results from simulated and measured data with both methods discussed above are presented. Subsequently, measurement uncertainties are discussed briefly.

### 5.1 Results from simulated data

To simplify testing and validation of the reconstruction, the travel time measurement is simulated for assumed artificial temperature fields. For this purpose, the distribution of the speed of sound is derived from the temperature field and the travel times are calculated for all measurement paths according to Eq. 5 by numerical integration using Gaussian quadrature. For a more realistic reconstruction, the simulated travel times are overlaid with noise of the order of magnitude of the travel time measurement in Section 3.2.

Figure 7 shows the reconstruction results of a constant temperature distribution with

Another assumed temperature distribution is shown in Fig. 8 (a). The distribution consists of a hotter inner ring with temperature

In Fig. 8 (b) the result with Tikhonov regularization can be seen, the result with minimizing the total variation in (c). The reconstruction result obtained by minimization of the total variation is again very good both qualitatively and quantitatively. Both rings are also reconstructed very well with Tikhonov regularization. However, the rest of the reconstructed distribution shows the artifacts seen in the previous example. Moreover, the outer area (outer two rings of the discretization) has a slightly too low value.

This example shows that rotationally symmetrical distributions can be reconstructed using both methods. However, if the radii of the rings are increased, i. e. if they are positioned closer to the outer edge, the reconstruction no longer works. This is due to the fact that data from more different measurement paths are available for the reconstruction near the midpoint. This is also the reason why the outer area has a slightly incorrect value with Tikhonov regularization.

After the reconstruction results from simulated data above, a result from measured travel times follows in the next section.

### 5.2 Results from measured data

For the reconstruction of a temperature field from real data, the travel times are measured using the setup from Section 3.1. The distribution to be measured has a hotspot that is generated by locally introducing heat. For this purpose, a coiled resistance wire made of constantan is used, to which current pulses are applied. The center of the spot where the heat is introduced has approximately the coordinates

The reconstruction results achieved with Tikhonov regularization and minimizing the total variation are shown in Fig. 9. The position of the generated hotspot is reconstructed correctly and unambiguously with both methods. The result obtained by minimizing the total variation in (b) again shows a significantly lower scattering of the reconstructed distribution outside the hotspot and the hotspot can be recognized much more clearly. With Tikhonov regularization, the values in the outer area are again somewhat too low, as already seen in the example in Fig. 8. However, Tikhonov regularization is also well suited for localizing the hotspot.

### 5.3 Measurement uncertainties

As already discussed in Section 3.2, the standard deviation of the travel time measurement is less than 1.2 ns for all paths, which corresponds to a standard deviation of the temperature measurement of less than 5 mK, assuming the size and shape of the investigated vessel. Additional deviations result from the non-linearity of the relationship between speed of sound and temperature.

The results from simulated data in Section 5.1 showed, that for small hotspots with small temperature deviations from the mean, a reconstruction of temperature fields for water with an uncertainty of about 0.1 K is possible with minimization of the total variation and with an uncertainty of about 0.5 K with Tikhonov regularization. Further errors can arise from the extent of hotspots not being congruent to the discretization. However, since these errors depend on the distribution to be reconstructed, they cannot be generally quantified. In the case of measurements in real bioreactors, the reactions forming products from reactants with their inherently different speeds of sound can also cause small additional uncertainties.

## 6 Conclusion

The temperature dependence of the speed of sound enables the reconstruction of a temperature field based on a sufficiently large number of travel time measurements in one slice plane. By installing the required arrangement of transducers at several heights, the temperature field can be measured in all corresponding planes. With Tikhonov regularization and minimization of the total variation of the solution vector, two methods to reconstruct the temperature field from the measured travel times were shown. If the condition for its applicability is met, minimizing the total variation leads to better results. This necessary condition is met if the temperature field deviates from an otherwise approximately constant value only in small individual areas, which is often the case in bioreactors.

For the actual use of the measurement method presented in this article in bioreactors, a number of factors still have to be examined in more detail. The most important of these factors is the influence of the substances produced in the bioreactor and the reactants supplied for them on the speed of sound and its temperature dependence. Without knowledge of this, no quantitative temperature measurement is possible, but qualitative monitoring of the occurrence of hotspots is nevertheless feasible. Other factors are the gassing of the reactor and the influence of stirring. However, as explained in Section 2, the influence of the flows caused by stirring can be almost neglected if the travel time measurements in both directions along a measurement path are carried out quickly enough in succession.

Both the arrangement of transducers and the two methods used for reconstruction can also be used to measure temperature fields in gases. Thus, the use of the method in boilers of power plants [22], [23] or for spatially resolved temperature measurement in halls [24] is also conceivable. However, a reduced resolution of the temperature measurement compared to the results in water is to be expected.

## About the authors

Michael Schwarz received his B.Sc and M. Sc. degrees in electrical engineering from the Friedrich-Alexander-University Erlangen-Nürnberg in 2014 and 2017, respectively. From 2017 to 2021 he worked as a research and teaching assistant at the Institute for Measurement Technology at the Johannes Kepler University Linz and received his Ph. D. degree from Johannes Kepler University Linz in 2022. His research interests are primarily in the areas of signal processing and ultrasonic measurement technology, especially ultrasonic temperature measurement and acoustic microscopy.

Bernhard G. Zagar heads the Institute for Measurement Technology at the Johannes Kepler University Linz. His research interests are widely spread in the field of measurement technology and range from sensor design to signal processing. Particular emphasis is placed on optical measuring methods, image processing and magnetic tomography.

## References

1. Horst Chmiel, Ralf Takors, and Dirk Weuster-Botz. *Bioprozesstechnik*. Springer, 2018.10.1007/978-3-662-54042-8Search in Google Scholar

2. Reinhard Lerch, Gerhard Sessler, and Dietrich Wolf. *Technische Akustik: Grundlagen und Anwendungen*. Springer, 2009.10.1007/978-3-540-49833-9Search in Google Scholar

3. Nykolai Bilaniuk and George S. K. Wong. Speed of sound in pure water as a function of temperature. *The Journal of the Acoustical Society of America*, 93(3):1609–1612, 1993.10.1121/1.406819Search in Google Scholar

4. Stefan J. Rupitsch. *Piezoelectric sensors and actuators: Fundamentals and applications*. Springer, 2018.10.1007/978-3-662-57534-5Search in Google Scholar

5. Ray C. Asher. *Ultrasonic Sensors for Chemical and Process Plant*. IOP Publishing, 1997.10.1177/002029409703000503Search in Google Scholar

6. Michael Schwarz and Bernhard G. Zagar. Reconstruction of temperature fields from ultrasonic travel time measurements. In *2021 IEEE International Ultrasonics Symposium (IUS)*. IEEE, 2021.10.1109/IUS52206.2021.9593413Search in Google Scholar

7. Michael Schwarz, Daniel Eder, and Bernhard G. Zagar. Delay-and-sum processing of echo data of transducers focused by 3D printed lenses. In *2020 IEEE International Ultrasonics Symposium (IUS)*. IEEE, 2020.10.1109/IUS46767.2020.9251389Search in Google Scholar

8. Heinrich Kuttruff. *Ultrasonics: Fundamentals and applications*. Springer, 1991.10.1007/978-94-011-3846-8Search in Google Scholar

9. Andrew B. Raine, Nauman Aslam, Christopher P. Underwood, and Sean Danaher. Development of an ultrasonic airflow measurement device for ducted air. *Sensors*, 15(5):10705–10722, 2015.10.3390/s150510705Search in Google Scholar
PubMed
PubMed Central

10. Ajay Singhvi, Aidan Fitzpatrick, and Amin Arbabian. Resolution enhanced non-contact thermoacoustic imaging using coded pulse excitation. In *2020 IEEE International Ultrasonics Symposium (IUS)*. IEEE, 2020.10.1109/IUS46767.2020.9251350Search in Google Scholar

11. Michael Schwarz and Bernhard G. Zagar. Rekonstruktion von Temperaturverteilungen aus Ultraschall-Laufzeitmessungen. In *Tagungsband DAGA 2021 - 47. Jahrestagung für Akustik*, pages 1524–1527, 2021.Search in Google Scholar

12. Julius S. Bendat and Allan G. Piersol. *Engineering applications of correlation and spectral analysis*. Wiley-Interscience, New York, 1980.Search in Google Scholar

13. Irving I. Gringorten and Penelope J. Yepez. *The division of a circle or spherical surface into equal-area cells or pixels*. Number 343. Phillips Laboratory, Directorate of Geophysics, 1992.10.21236/ADA257770Search in Google Scholar

14. Per Christian Hansen. *Discrete inverse problems: insight and algorithms*, volume 7. Siam, 2010.10.1137/1.9780898718836Search in Google Scholar

15. Heinz Werner Engl, Martin Hanke, and Andreas Neubauer. *Regularization of inverse problems*, volume 375. Kluwer Academic Publishers, 2000.Search in Google Scholar

16. David L. Donoho. Compressed sensing. *IEEE Transactions on information theory*, 52(4):1289–1306, 2006.10.1109/TIT.2006.871582Search in Google Scholar

17. Emmanuel J. Candès and Michael B. Wakin. An introduction to compressive sampling. *IEEE signal processing magazine*, 25(2):21–30, 2008.10.1109/MSP.2007.914731Search in Google Scholar

18. Yonina C. Eldar and Gitta Kutyniok. *Compressed sensing: theory and applications*. Cambridge university press, 2012.10.1017/CBO9780511794308Search in Google Scholar

19. David L. Donoho. For most large underdetermined systems of linear equations the minimal l1-norm solution is also the sparsest solution. *Communications on Pure and Applied Mathematics: A Journal Issued by the Courant Institute of Mathematical Sciences*, 59(6):797–829, 2006.10.1002/cpa.20132Search in Google Scholar

20. Leonid I. Rudin, Stanley Osher, and Emad Fatemi. Nonlinear total variation based noise removal algorithms. *Physica D: nonlinear phenomena*, 60(1-4):259–268, 1992.10.1016/0167-2789(92)90242-FSearch in Google Scholar

21. Emmanuel Candès and Justin Romberg. l1-magic: Recovery of sparse signals via convex programming, 2005. statweb.stanford.edu/˜candes/software/l1magic.Search in Google Scholar

22. Mauro Bramanti, Emanuele A Salerno, Anna Tonazzini, Sauro Pasini, and Antonio Gray. An acoustic pyrometer system for tomographic thermal imaging in power plant boilers. *IEEE Transactions on instrumentation and measurement*, 45(1):159–167, 1996.10.1109/19.481329Search in Google Scholar

23. Helmut Sielschott and Willy Derichs. Tomography with few data: Use of collocation methods in acoustic pyrometry. In *ECMI ’94: Student Proceedings*, pages 251–263, 1994.Search in Google Scholar

24. Peter Holstein, Armin Raabe, Roland Müller, Manuela Barth, David Mackenzie, and Eric Starke. Acoustic tomography on the basis of travel-time measurement. *Measurement Science and Technology*, 15(7):1420–1428, 2004.10.1088/0957-0233/15/7/026Search in Google Scholar

**Received:**2022-02-23

**Accepted:**2022-04-04

**Published Online:**2022-04-23

**Published in Print:**2022-09-30

© 2022 the author(s), published by De Gruyter

This work is licensed under the Creative Commons Attribution 4.0 International License.