DEVICE AND METHOD FOR CONTROLLING A MICROORGANISM CONTENT

Abstract

Device and method for controlling a content of microorganisms in a liquid, the device carrying out the method and comprising: a measuring unit for measuring the microorganism content, a control unit which, in order to achieve a predefined microorganism content, determines a dosage of a biocide on the basis of the measured microorganism content using a model with at least one parameter, a supply unit, which supplies the biocide to the liquid in the dosage determined by the control unit, and a computing unit, which calculates the at least one parameter from a recording of the microorganism content measured by the measuring unit over at least one past time interval and the amount of biocide supplied in this time interval.

Claims

1. A device for controlling a content of microorganisms in a liquid, the device comprising: a measuring unit for measuring said microorganism content, a control unit, which is connected to the measuring unit and which is configured to determine a dosage of a biocide based on the measured microorganism content in order to achieve a predefined microorganism content, a supply unit, which is connected to the control unit and which is configured to supply the biocide to the liquid in the dosage determined by the control unit, and a computing unit connected to the measuring unit and the control unit, wherein the control unit is configured to determine the dosage using a model with at least one parameter of a relationship between an amount of biocide supplied in a time interval and a change in the microorganism content caused by the supplied amount in this time interval, and wherein the computing unit is configured to calculate the at least one parameter for use by the control unit in at least one later time interval from a recording of the microorganism content measured by the measuring unit over at least one past time interval and the amount of biocide supplied in this time interval.

2. The device according to claim 1, wherein said model is given by an equation $\frac{\mathrm{dX}}{\mathrm{dt}}=-{\mathrm{kB}}^{n}X$ with: $\frac{\mathrm{dX}}{\mathrm{dt}}$ change in microorganism content in a time interval, k parameter of the model, B amount of biocide supplied in this time interval, n dilution coefficient of the biocide and X measured microorganism content.

3. The device according to claim 1, wherein said model is given by an equation $\frac{\mathrm{dX}}{\mathrm{dt}}=-k_1{B}_0^n{e}^{-k_2\mathrm{nt}}X$ with: $\frac{\mathrm{dX}}{\mathrm{dt}}$ change in microorganism content in a time interval, k.sub.1, k.sub.2 parameters of the model, B.sub.0 amount of biocide supplied in this time interval at the time of supply, n dilution coefficient of the biocide and X measured microorganism content.

4. The device according to claim 1, wherein the measuring unit is a flow cytometer which is configured to measure the content of microorganisms of a predetermined type.

5. The device according to claim 1, wherein the computing unit is configured to calculate the at least one parameter after each time interval of a sequence of time intervals from the recording of the microorganism content measured by the measuring unit over at least one immediately preceding time interval and the amount of biocide supplied in this at least one immediately preceding time interval.

6. The device according to claim 1, wherein the supply unit is configured to select the biocide from one or more components according to a predetermined composition, wherein the device further comprises a control unit which is connected to the supply unit and which is configured to predetermine the composition for the supply unit.

7. The device according to claim 6, wherein the device comprises at least one sensor for measuring at least one of the measurands constituted by pH, temperature, pressure and conductivity of the liquid, wherein the control unit is connected to the at least one sensor and is configured to predetermine the composition depending on at least this/these measurand/s.

8. The device according to claim 6, wherein the control unit is configured to predetermine a composition for the supply unit at regular time intervals.

9. A method for controlling a content of microorganisms in a liquid, the method comprising in a control process: measuring said microorganism content by means of a measuring unit, determining a dosage of a biocide with a view to achieving a predefined microorganism content based on the measured microorganism content and a model comprising at least one parameter of a relationship between an amount of biocide supplied in a time interval and a change in microorganism content caused by the supplied amount in that time interval, and supplying the biocide in the determined dosage to the liquid; and in a tracking process parallel or intermittent to the control process: calculating the at least one parameter from a recording of the microorganism content measured by the measuring unit over at least one past time interval and the amount of biocide supplied in that time interval for use in the control process in at least one later time interval.

10. The method according to claim 9, wherein said model is given by an equation $\frac{\mathrm{dX}}{\mathrm{dt}}=-{\mathrm{kB}}^{n}X$ with: $\frac{\mathrm{dX}}{\mathrm{dt}}$ change in microorganism content in a time interval, k parameter of the model, B amount of biocide supplied in this time interval, n dilution coefficient of the biocide and X measured microorganism content.

11. The method according to claim 9, wherein said model is given by an equation $\frac{\mathrm{dX}}{\mathrm{dt}}=-k_1{B}_0^n{e}^{-k_2\mathrm{nt}}X$ with: $\frac{\mathrm{dX}}{\mathrm{dt}}$ change in microorganism content in a time interval, k.sub.1, k.sub.2 parameters of the model, B.sub.0 amount of biocide supplied in this time interval at the time of supply, n dilution coefficient of the biocide and X measured microorganism content.

12. The method according to claim 9, wherein the measuring unites measures the content of microorganisms of a predetermined type.

13. The method according to claim 9, wherein the tracking process is carried out after each time interval of a sequence of time intervals, wherein the at least one parameter is calculated from the recording of the microorganism content measured by the measuring unit over at least one immediately preceding time interval and the amount of biocide supplied in this at least one immediately preceding time interval.

14. The method according to claim 9, wherein the biocide is selected from one or more components according to a predetermined composition.

15. The method according to claim 14, wherein at least one sensor measures at least one of the measurands constituted by pH, temperature, pressure and conductivity of the liquid and the composition is predetermined depending on at least this/these measurand/s.

16. The method according to claim 14, wherein a composition is predetermined at regular time intervals.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The disclosed subject matter is explained in greater detail below with reference to exemplary embodiments shown in the accompanying drawings. In the drawings:

[0039] FIG. 1 shows a device according to the disclosed subject matter for controlling a content of microorganisms in a liquid in a block diagram;

[0040] FIG. 2 shows a diagram of the fluorescence properties of microorganisms detected with a measuring unit of the device from FIG. 1;

[0041] FIG. 3 shows a graph of the biocide supply and the microorganism content in the liquid over time using the device from FIG. 1; and

[0042] FIG. 4 shows an exemplary method executed by the device from FIG. 1 for controlling the microorganism content in the liquid in a flow chart.

DETAILED DESCRIPTION

[0043] A device 1 according to FIG. 1 controls a content X of microorganisms M in a liquid 2 in order to achieve a predefined microorganism content X*. The microorganisms M are, for example, algae, bacteria, fungi, in particular yeasts, etc. The liquid 2 is used, for example, in an industrial process for the production, processing, cooling, cleaning or the like of an object or a substance or is the actual process objectfor example in the form of a beverage in a filling plant. The industrial process may be a batch process here, in which the liquid 2 is produced or used in batches, or it may be a continuous process in which the liquid 2 is also used continuously.

[0044] The liquid 2 is in a container (here: an open tank) 3, and microorganisms M havein this exampleformed an unwanted deposit 4 in a corner 5 of the tank 3 as a result of their multiplication. Alternatively, microorganisms M multiply in the liquid without forming a deposit 4 and contaminate the liquid 2 or are undesirable for other reasons.

[0045] A measuring unit 6 measures said microorganism content X in the liquid 2. The content X indicates the number of microorganisms M per reference quantity of the liquid 2, i.e. per volume (for example millilitre), per mass (for example gram) or per quantity (for example mole) of the liquid 2. To measure the microorganism content X, the measuring unit 6 may use different methods, for example a quantitative PCR (qPCR), an adenosine triphosphate (ATP) test, a next generation sequencing (NGS) measurement method or a Raman spectroscopy. In the case shown, the measuring unit 6 is a flow cytometer which measures the content X of microorganisms M of a predetermined type A, as explained in greater detail below using the example of FIG. 2.

[0046] In a flow cytometer, particles are passed individually in rapid succession past a high electric voltage and/or a laser beam and their respective specific reaction to the electric field (for example deflection) and/or the laser beam (for example reflection, colour, fluorescence, etc.) is detected and evaluated. Flow cytometers allow a parallel evaluation of several measuring channels at high speed, i.e. also in real time, and distinguish particles according to their properties or reactions; optionally, auxiliary substances, for example fluorescent dyes, are used here. In most cases, multi-dimensional data fields are generated, in which the detected particles may be automatically classified, for example by (multi-dimensional) cluster formation, and then optionally counted, for example by class. In this way, the type A of microorganisms M of which the content X is to be (selectively) measured in the liquid 2 may optionally be predetermined.

[0047] The exemplary two-dimensional diagram in FIG. 2 shows microorganisms M that were detected by flow cytometry, differentiated according to their properties (in this case two properties, namely red and green fluorescence) and recorded. In the diagram, the green fluorescence is plotted horizontally and the red fluorescence vertically (here: in each case on a logarithmic scale). Further physical or chemical properties (for example other fluorescences, etc.) may be detected on further measuring channels and for example evaluated in further dimensions. Microorganisms M with similar properties (here: in these two wavelength ranges) each form clusters C.sub.1, C.sub.2, . . . , generally C.sub.k, in the diagram. By means of known clusters C.sub.k or those determined with the aid of a learning cluster evaluation, different microorganisms M may be distinguished from each other; for example, in the diagram of FIG. 2 algae are marked by a dashed rectangle E.sub.1 and distinguished from bacteria (dot-and-dash rectangle E.sub.2). Therefore, either algae or (in this case) bacteria, etc. may be predetermined as the abovementioned type A. Alternatively, certain algae, for example blue-green algae, or certain bacteria, for example lactic acid bacteria, etc., or a type A of microorganisms M with special properties may be predetermined as the abovementioned type A, insofar as these may be distinguished in the flow cytometry, for example it could be differentiated by the predetermined type A whether a microorganism M is viable or not, and only the viable microorganisms M could be predetermined as type A. Lastly, different microorganisms M, for example algae and bacteria, may be predetermined together as type A.

[0048] When, in the case of FIG. 2, for example, the stated bacteria are predetermined as the type A, the measuring unit 6 measures their content X in the liquid 2, i.e. other microorganisms M or particles that are not microorganisms M at all are not taken into account further by the measuring unit 6 when measuring the microorganism content X in this example. From diagrams of successive time intervals t, a change in the microorganisms M, in particular their content X, in the liquid 2 over time may be determined.

[0049] It is understood that the measuring unit 6 measures the microorganism content X generally from a sample of liquid 2 taken from the container 3.

[0050] Returning to the device 1 of FIG. 1, a control unit 7 is connected to the measuring unit 6 so that it has access to the microorganism content X measured by the measuring unit 6 in each case. The control unit 6 has, for example, a proportional integral (PI), a proportional integral differential (PID) controller or the like for controlling the microorganism content X. For this purpose, the control unit 7 determines a dosage D of a biocide B which is to be added to the liquid 2 in order to achieve the predefined microorganism content X*. The predefined microorganism content X* may be predefined here to be constant on the one hand or to vary over time on the other hand. The temporal variation of the predefined microorganism content X* is based, for example, on an adaptation to a previously planned course and/or a measured state of the industrial process. For example, in the context of a so-called fed-batch process successively supplying liquid 2for example in a fermentation processnot only the amount of liquid 2 depends on the process progression, i.e. varies over time, but usually also the predefined microorganism content X*; in particular, the quantity of liquid may be small and a, for example, high predefined microorganism content X* may be desired in an initial stage of the process, whereas in a later stage of the process the quantity of liquid is larger and the predefined microorganism content X* is to be reduced.

[0051] The device 1 further comprises a supply unit 8 with one or more reservoirs from which it supplies biocide B to the liquid 2, for example by means of a controllable pump or a control valve. The supply unit 8 is connected to the control unit 7 to supply the biocide B to the liquid 2 in the dosage D determined by the control unit 7.

[0052] In order to determine the dosage D, the control unit 7 uses a model with at least one parameter k. The model represents a relationship between an amount P of biocide B supplied in a time interval t and the change in the microorganism content X caused thereby in this time interval t, i.e. it is a kinetic model of the reaction of the liquid 2 and thus of its microorganism content X to the biocide B supplied. It is understood that the amount P of biocide B supplied in the time interval t coincides with the dosage D of biocide B to be supplied to the liquid 2 if this has been determined for the same time interval t. The control unit 7 uses the model to adapt one or more time constants and/or gains of its controller to the reaction of the liquid 2, i.e. the change in its microorganism content X, to biocide supply.

[0053] The model may be of various types, in particular a model that assumes a linear, quadratic or exponential decrease in the microorganism content X with biocide addition. In one embodiment, the model is given by the following equation 1:

[00005]$\begin{array}{cc}\frac{\mathrm{dX}}{\mathrm{dt}}=-{\mathrm{kB}}^{n}X& \left(1\right)\end{array}$

with:

[00006]$\frac{\mathrm{dX}}{\mathrm{dt}}$ [0054] change in microorganism content X in a time interval t (here: the infinitesimal time interval dt), [0055] k parameter of the model concerning a deactivation of the microorganisms M, [0056] B amount P of biocide B supplied in this time interval t, [0057] n dilution coefficient of the biocide B and [0058] X measured microorganism content X.

[0059] In an alternative embodiment, the model is given by the following equation 2:

[00007]$\begin{array}{cc}\frac{\mathrm{dX}}{\mathrm{dt}}=-k_1{B}_0^n{e}^{-k_2\mathrm{nt}}X& \left(2\right)\end{array}$

with:

[00008]$\frac{\mathrm{dX}}{\mathrm{dt}}$ [0060] change in microorganism content X in a time interval t (here: the infinitesimal time interval dt), [0061] k.sub.1 parameter of the model concerning a deactivation of the microorganisms M, [0062] k.sub.2 parameter of the model concerning an inactivation of biocide B after its supply to the liquid 2, [0063] B.sub.0 amount P of biocide B supplied in this time interval t at the time of its supply, [0064] n dilution coefficient of the biocide B and [0065] X measured microorganism content X.

[0066] The extended model of equation (2) makes it possible, through the parameter k.sub.2, to take into account an inactivation of the biocide B after its supply to the liquid 2.

[0067] In both of the above-mentioned models, the dilution coefficient n of the biocide B is either known in advance if the biocide B used and the liquid 2 and the microorganisms M contained therein are known; otherwise, the dilution coefficient n is another parameter of the model, so that the model given by equation (1) has two parameters k and n and the model given by equation (2) has three parameters k.sub.1, k.sub.2 and n.

[0068] If the model is also to represent an influence of any side reactions of the industrial process on the relationship between the amount P of biocide B supplied and the change in the microorganism content X caused thereby, the model may optionally be given, for example, by a system of coupled differential equations, which in each case describe the influence of the industrial process itself or that of its side reactions.

[0069] In order to adapt the model or its at least one parameter k, k.sub.1, k.sub.2, n and thus the control unit 7 to the reaction of the liquid 2 to biocide supply and also to track it in the event of dynamic changes in this reaction, the device 1 has a computing unit 9 which is connected to the measuring unit 6 and the control unit 7 and which tracks the control unit 7 via a control path 10. The computing unit 9 calculates the parameter/s k or k.sub.1, k.sub.2 etc. (and, if necessary, the dilution coefficient n as a further parameter) of the respective model and, for this purpose, uses a recording of the microorganism content X measured by the measuring unit 6 over at least one past time interval t and an amount P of biocide B which was supplied to the liquid 2 in this time interval t, as will be explained in greater detail below with reference to the example of FIG. 3, which relates to a model with a parameter k. It is understood that in the case of a model with two or more parameters k.sub.1, k.sub.2, n, . . . the computing unit 9 may calculate them from recordings of at least a corresponding number of previous time intervals t.

[0070] FIG. 3 shows a sequence of time intervals t.sub.1, t.sub.2, . . . , generally t.sub.n, which each lie between two times t.sub.1, t.sub.2, . . . , generally t.sub.n. At least at each time t.sub.n, the measuring unit 6 measures the microorganism content X.sub.1, X.sub.2, . . . , generally X.sub.n (here: a continuously measured progression of the microorganism content X). In each time interval t.sub.n of the sequence, an amount P.sub.1, P.sub.2, . . . , generally P.sub.n, of biocide B is supplied in each case (here: continuously during each time interval t.sub.n). If, for example, an amount P.sub.4 of biocide B is added to the liquid 2 in a time interval t.sub.4 between the times t.sub.4 and t.sub.5, this causes a change in the microorganism content X from a value X.sub.4 at the time t.sub.4 to a value X.sub.5 at the time t.sub.5.

[0071] Said parameter k of the model may be calculated by the computing unit 9 after any or after each time interval t.sub.n of the sequence. If, for example, the parameter k is to be calculated for the time interval t.sub.4, the computing unit 9, after this time interval t.sub.4 has elapsed, takes the microorganism content X.sub.4 measured by the measuring unit 6 over this time interval t.sub.4 from the recording (FIG. 3) and the amount P.sub.4 of biocide B supplied during this time interval t.sub.4. The parameter k is calculated on the basis of this amount P.sub.4 and the change X (or infinitesimal: dX) of the microorganism content X (for example according to X=X.sub.5X.sub.4) caused thereby and is transmitted to the control unit 7 for use by it in one or more later time intervals t.sub.n (here: for example t.sub.5 and/or t.sub.6).

[0072] Optionally, the computing unit 9 may contain an adjustment device which averages or adjusts the calculated parameter k with values of the parameter k calculated for previous time intervals t byfor example weightedaveraging before transfer to the control unit 7. The adjustment device is, for example, an additional computing module of the computing unit 9 or a neural network, in particular a Long Short-Term Memory (LSTM), which during the adjustment optionally additionally takes into account changes in the microorganism content X caused in previous time intervals t. In this case, the control unit 7 uses the adjusted parameter k of the model to adjust its controller, whereby past parameter changes are taken into account and a sudden adjustment of the controller is prevented.

[0073] Referring back to FIG. 1, in an optional embodiment the delivery unit 8 is configured to select the biocide B from one or more components, each of which the delivery unit 8 holds in a reservoir, in order to achieve a predetermined composition Z of the biocide B. In this embodiment, the device comprises a control unit 11, which is connected to the supply unit 8 and predetermines said composition Z for the supply unit 8. The control unit 11 thus intervenes in the supply of the biocide B, for example by predetermining or changing the composition Z of the biocide B via a control path 12.

[0074] The composition Z may optionally be predetermined here randomlyat least occasionally; alternatively, the components of the composition Z are selected heuristically, for example, as will be explained in greater detail below. Optionally, the device 1 comprises for this purpose one or more (in the example of FIG. 1: two) sensors 13, 14, which measure at least one measurand (usually one measurand per sensor 13, 14), for example the pH W, the temperature T, the pressure and/or the conductivity of the liquid 2. In the example of FIG. 1, one sensor 13 of said sensors 13, 14 measures the pH W and the other sensor 14 measures the temperature T of the liquid 2. In this case, the control unit 11 predetermines the composition Z depending on at least said measurands W, T.

[0075] When predetermining the composition Z of the biocide B, the control unit 11 is further optionally connected to the measuring unit 6 to predetermine the composition Z also depending on the microorganism content X. Furthermore, the control unit 11 is optionally connected to the control unit 7 to obtain the dosage D determined by the latter, so that the control unit 11 may predetermine the composition Z also depending on the dosage D or the amount P of biocide B supplied in the time interval t, depending on the parameter/s k, k.sub.1, k.sub.2, n calculated therefrom, and/or depending on the change in the microorganism content X as a result of the biocide supply. Moreover, the control unit 11 may take into account other information, for example an environmental compatibility and/or the cost of each of the components, and may predetermine the composition Z in dependence thereon.

[0076] All of these dependencies are stored, for example, in a database as heuristics, which belong to the control unit 11 so that it may predetermine a suitable composition Z based on heuristics. Alternatively, the control unit 11 optionally has a self-learning system, for example a neural network, in particular a Long Short-Term Memory (LSTM), which learns from compositions Z predetermined in the past the resulting amounts P of biocide B added or the changes in the microorganism content X caused thereby. On the basis of the respective measured values of the sensors 13, 14 or the measuring unit 6 and/or any further of said information and the heuristics, the control unit 11 then predetermines a particularly suitablefor example particularly effective, environmentally friendly, cost-effectivecomposition Z of the biocide B in each case. For this purpose, the system may at least occasionally predetermine random compositions Z.

[0077] With reference to the example shown in FIG. 4, various examples of a method 15 carried out by the device 1 for controlling the microorganism content X in the liquid 2 are explained in greater detail below.

[0078] In an initialisation step 16, the parameter/s k or k.sub.1, k.sub.2 of the model including the dilution coefficient n and the microorganism content X* to be achieved are predefined and the microorganism content X of the liquid 2 is measured. A control process R of the method 15 is then either continuously executed, for example with the aid of analogue controller components, or is run through cyclically, in particular periodicallyas in the example shownas a digital control process R. As described above, the control process R comprises at least one step 17, in which the control unit 7 determines the dosage D of the biocide B, a step 18, in which the supply unit 8 supplies the biocide B to the liquid 2 in the dosage D determined in step 17 and optionally selects the composition Z, and a step 19, in which the measuring unit 6 measures the microorganism content X in the liquid 2.

[0079] If the control process R is run through cyclicallyas shownthe supply unit 8 may in step 18 either immediately supply the entire dose of the dosage D of biocide B determined by the control unit 7 to the liquid 2 or may cause the biocide B to be added to the liquid 2 continuously, for example with the aid of a pump, in the determined dosage D.

[0080] In order to take into account a merely slow change in the microorganism content X after the biocide B has been supplied 18, the control process R optionally contains a waiting step 20 so that a changed microorganism content X may already be expected during the following measurement 19.

[0081] In a tracking process F, the computing unit 9 calculates the parameter/s k, k.sub.1, k.sub.2, n of the model in step 21 as previously described. The calculated parameter/s k, k.sub.1, k.sub.2, n of the model is/are used to track the control unit 7 or its controller via path 22 (to step 17). The tracking process F may be carried out in parallel with the control process R oras shownintermittently, i.e. interrupting the control process R in each case. In the variant shown, the tracking process F immediately follows the control process R, i.e. it interrupts the control process R if a positive check is carried out in a branch 23 (branch Y of branch 23); if a negative check is carried out (branch N of branch 23), the control process R continues to be executed. The test criterion used in branch 23 is, for example, a number of runs of the control process R since the last run of the tracking process F, the expiry of a period of time, for example the time interval t, or a particularly high value for, for example, the dosage D determined by the control unit 7 or the microorganism content X measured by the measuring unit 6. Alternatively, the tracking process F may immediately follow the control process R in each cycle, i.e. the process 15 may have no branching 23.

[0082] In this context, it should be noted that the time interval t may be chosen as required, for example equal to the period of the control process R if this is run periodically, or longer, in particular a multiple thereof.

[0083] In an optional step 24, which is carried out in parallel with the control process R and the tracking process F or intermittently theretoas shownthe control unit 11 determines the composition Z of the biocide B as described above and stipulates this to the supply unit 8 via path 25. In the variant shown, the predetermining 24 immediately follows the tracking process F if a positive check is carried out in a branch 26 (branch Y of branch 26); if the check is negative (branch N of branch 26), the control process R continues. In general, the composition Z is predetermined at regular time intervals, i.e. reaching the time interval from the last point of being predetermined 24 is used as a test criterion in branch 26. These time intervals are usually considerably largerfor example 10 times or 100 times largerthan the time intervals t considered in the control process R and the tracking process F. As an alternative or supplementary test criterion, for example a decrease in the microorganism content X below the threshold valueand thus lower than expecteddespite a high dosage D or an exceeding of the threshold value of the microorganism content X or an increase of the latter may be used.

[0084] The disclosed subject matter is not limited to the embodiments set out, but encompasses all variants, modifications and combinations thereof which fall within the scope of the accompanying claims.