Device and method for detecting deposits in a container holding a medium

Abstract

The invention relates to a device for detecting deposits on a surface, which is an inner surface of a conduit through which a medium flows, and comprises a light transmitter, a light detector, and an optical fiber that carries light to and / or from the sensor, the end of which facing away from the sensor forming the surface on which the deposits are deposited. A control and evaluation unit serves to control the light transmitter and evaluate the received signals, wherein the transmitted light emitted by the light transmitter generates fluorescence in the deposits and the fluorescence light is detectable by the light detector.According to the invention, the evaluation unit is configured to learn correction values ​​for at least one environmental parameter, representing an environmental condition for the deposits, in a learning process, and is configured to correct the received signals with these correction values. Furthermore, it is configured to generate a deposit signal representing the presence of deposits from the corrected received signals and output it at an output. The invention also relates to a method for detecting the deposits.

Description

[0001] The invention relates to a device for detecting deposits in a container containing a medium, in particular a line through which the medium flows, with a light transmitter, a light detector, at least one optical fiber which guides light from and / or to the sensor, the front end of which facing away from the sensor forms the surface on which the deposits are deposited, and a control and evaluation unit for controlling the light transmitter and evaluating the received signals, wherein the transmitted light emitted by the light transmitter generates fluorescence in the deposits and the fluorescent light is detected by the light detector.

[0002] Such a device is known from DE 102016109819 B4. This device for detecting deposits on a surface includes a light source that can generate excitation radiation. The light is introduced via an optical fiber into a line carrying a medium. The end of the optical fiber facing away from the sensor is inserted into the wall of the line, so that the optical fiber "looks" into the line. Deposits in the line then deposit on the front end of the optical fiber. The excitation radiation excites fluorescence in the deposits, and the fluorescent light is guided via the optical fiber to a light-receiving element of the sensor, where it is evaluated. Depending on the intensity of the fluorescent light, the thickness of the deposits can be determined.

[0003] Knowledge of deposits and their thickness is particularly important in the food processing industry. Such deposits, known as fouling, occur in so-called closed systems such as heat exchangers (bundle heat exchangers, plate heat exchangers, flash pasteurizers), pipelines (product lines, CIP lines, CIP / SIP systems [CIP = cleaning in place, SIP = sterilization in place]), filter systems, or tanks. State-of-the-art technology involves rinsing containers containing food, particularly pipes, at predefined intervals to prevent these deposits or excessive deposits. Thorough rinsing is then carried out at these intervals, with the rinsing process running for a fixed, predefined time, which is always longer than the time actually required to ensure that all deposits have been rinsed out.

[0004] For the purposes of this description, "container" or the term "line" refers to any type of the aforementioned closed systems. The terms "container" and "line" are used synonymously.

[0005] The aim is therefore, on the one hand, to only start such flushing processes when they are necessary and to end the flushing process as early as possible, namely when the deposits have been flushed out and not to flush for an unnecessarily long time when the pipes are already clean.

[0006] The known sensor works in principle, but only under ideal conditions, which do not exist in reality. There are numerous boundary conditions that make realistic measurements impossible in practice, but only possible under ideal conditions. Ideal conditions would exist, for example, if precise knowledge were available about a number of factors. These include: - the so-called photobleaching, i.e. a decrease in fluorescence over time (bleaching of the deposits), - Homogeneity and density of deposits, - no fluorescent light generation if no deposits are present, - constant temperatures, humidity, pH, turbidity, color, viscosity and similar properties of the medium, because all these properties influence the fluorescence intensity, - constant flow parameters, e.g. flow velocity, - Constituents of the medium, i.e. protein content, fat content and the like, whereby one would also need to know how the corresponding constituent affects the fluorescence properties.

[0007] All of this is not normally the case, which is why the object of the invention is to further develop the known sensor in such a way that real-life use becomes possible.

[0008] This object is achieved by a device having the features of claim 1 and a method according to claim 10.

[0009] The device according to the invention for detecting deposits on a surface which is an inner surface of a container containing a medium, in particular a line through which a medium flows, comprises - a light transmitter, - a light detector for generating reception signals depending on the incidence of light on the light detector, - at least one optical fiber that guides light from and / or to the sensor, the front end of which, facing away from the sensor, forms the surface on which the deposits are deposited, - a control and evaluation unit for controlling the light transmitter and evaluating the received signals, - wherein the transmitted light emitted by the light transmitter generates fluorescence in the deposits and the fluorescent light is detectable by the light detector.

[0010] According to the invention, the evaluation unit is designed to learn correction values ​​for at least one environmental parameter, which represents an environmental condition for the deposits, in a learning process and is designed to correct the received signals with the correction values ​​and is further designed to generate a deposit signal representing the presence of deposits from the corrected received signals and to output it at an output.

[0011] The object is achieved by a method for operating a device according to the invention with the steps: - i) Inserting the device into a wall of the pipe carrying the medium, - ii) passing the medium through the pipe, - iii) emitting the transmitted light and guiding the transmitted light through the optical fiber, - iv) recording the environmental parameter(s), - v) receiving the fluorescent light coupled into the optical fiber and generating reception signals, - vi) evaluating the received signals by correcting them with the correction values ​​determined in the learning process for the environmental parameter(s) detected, - vii) evaluating the corrected received signals and outputting a deposit signal representing the presence of deposits in the line.

[0012] By correcting for real-world conditions, meaningful and reproducible deposition signals are obtained. This makes the device suitable for real-world applications with diverse environmental conditions.

[0013] The correction values ​​corresponding to the respective environmental parameters were previously learned in a learning process. This learning process comprises the following steps: - a) Inserting the device into a wall of the pipe carrying the medium, - b) Passing the medium through the pipe, - c) Emitting the transmitted light with constant power and guiding the transmitted light through the optical fiber, - d) recording the environmental parameter(s), - e) receiving the fluorescent light coupled into the optical fiber and generating and storing the received signals in a memory together with the environmental parameter(s), - Repeat steps c) to e) for a preselected period of time, with changing environmental parameters, - f) evaluating the received signals and / or evaluating the course of the received signals with changing environmental parameters and - g) determining correction values ​​for the received signals for the associated environmental parameter(s), - h) Saving the correction values ​​together with the environmental parameters

[0014] There are various methods for determining and finding the correction values. If the qualitative fluorescence behavior of a medium with changes in the ambient parameter, such as temperature dependence, is known from databases, then, for example, if the fluorescence decreases by 50% when the temperature increases by 10°C, this knowledge could be used to determine the correction values ​​to compensate for the temperature dependence of the received signals.

[0015] It is also conceivable to determine a certain dependency experimentally in the real application by creating conditions where a constant received signal is present and then changing a selected environmental parameter and compensating for the change in the received signal, which results in the correction values.

[0016] Regarding steps f) and g), it may be important to consider that there may be a time delay between the change in the environmental parameter(s) and the change in the received signal. In this case, this must also be accounted for and taken into account in the correction.

[0017] An important environmental parameter is time and the associated photobleaching, i.e., the bleaching of deposits during measurements. Correction values ​​for this can be advantageously found by observing the fluorescence signal during the formation of deposits. The assumption is that with increasing deposits, the fluorescence signal also increases until it reaches a measurable maximum and should then remain at this maximum (assuming that the other environmental parameters remain constant). The subsequent decrease in the fluorescence signal is then caused by photobleaching. The values ​​required to compensate for this decrease are the desired correction values.

[0018] It is also very useful if a zero signal is known. A zero signal is understood to be a fluorescence signal that results when no deposits are present (clean line condition). In a further development of the invention, such a zero signal is obtained by starting a flushing process from a clean line, recording received signals during this process, and defining the received signal then present as the zero signal after the flushing process is completed. Knowing this zero signal is advantageous for being able to end future flushing processes promptly, namely when the zero signal has been reached, and not using a fixed flushing time interval that is always too long for safety reasons. This saves time and avoids unnecessary consumption of resources.This zero signal does not always have to be the same, but can depend on the currently used detergent due to the possible autofluorescence which is dependent on the detergent used.

[0019] As already indicated above, the correction values ​​are usually dependent on the environmental parameters, for example time-dependent, temperature-dependent, moisture-dependent, etc.

[0020] The deposit signal can be a measure of the thickness of the deposit. However, this is not necessary, as for a meaningful use of the device according to the invention, it is sufficient if there is any knowledge of deposits at all, or, with regard to the flushing process, if the received signal can be used, to shorten the flushing process (with the zero signal).

[0021] In a further development of the invention, other environmental parameters can be taken into account in addition to or in addition to the time, e.g., moisture content, temperature, turbidity, color, flow rate, flow parameters such as flow velocity, pH value, and / or viscosity. Corresponding correction values ​​can be found for the other parameters, preferably in the specific application during corresponding learning processes.

[0022] The ambient parameter can also reflect a substance content of the product, such as protein content, fat content, or the like. Corresponding correction values ​​then take into account a change in the received signal with a change in this parameter.

[0023] Advantageously, one or more sensors are provided for measuring the environmental parameter(s). Such sensors can be, for example, temperature sensors, humidity sensors, pH sensors, flow velocity sensors, or the like.

[0024] It is also conceivable to refine the correction values ​​during operation by continuously recording the ambient conditions: Under defined conditions (same medium, same flow rate, same temperature), the fouling development process, and thus also the signal curve, is expected to be reproducible. Anomalies in the signal curve, such as sudden signal drops or increases, are then not due to a change in the fouling state, but rather to a possibly unknown external influence, and can still be corrected.

[0025] The invention will be explained in detail below using an exemplary embodiment with reference to the drawing. The drawing shows: Fig. 1 a schematic representation of a device according to the invention;

[0026] One in Fig.The device 10 according to the invention shown in Figure 1 is used to detect deposits 12 on an inner surface 14 of a line 18 through which a medium 16 flows. For this purpose, the device 10 has a light transmitter 20, a light detector 22, and an optical fiber 24. The light transmitter 20 emits excitation light 26, which is coupled into the optical fiber 24. The optical fiber 24 is inserted into the line 18 with its end 28 facing away from the light transmitter 20, such that the front side of the end 28 is flush with the inner surface 14 of the line 18 and forms part of the inner surface 14. The end 28 is inserted into a wall 19 of the line 18 and fastened there, for example by pressing or gluing or the like. The deposits 12, which are deposited everywhere on the inner surface 14, will therefore also cover the end face of the optical fiber 24.

[0027] The excitation light 26, which exits the end 28 of the optical fiber 24, excites fluorescence in the deposits 12. The resulting fluorescent light 30, which couples into the end 28 of the optical fiber 24, is guided through the optical fiber 24 to the light detector 22 via a dichroic splitter mirror 32, which transmits the transmitted light 26 and reflects the fluorescent light 30. Other coupling and decoupling methods for the transmitted light or fluorescent light are possible. The aforementioned dichroic splitter mirror 32 is only one example.

[0028] The light detector 22 generates electrical reception signals depending on the incidence of light, which are fed to an evaluation unit 34.

[0029] The evaluation unit 34 comprises a computing unit 36 ​​and a memory 38, as well as an output 40. In the evaluation unit 32, the evaluation of the received signals takes place in order to ultimately have a statement about the presence or absence of the deposits 12 and, if applicable, about the thickness of the deposits 12. How one arrives at usable statements and how the evaluation unit must be designed for this purpose is a central element of this invention.

[0030] The evaluation unit 34 also serves to control the light transmitter 20. The device 10 according to the invention is preferably operated in a pulsed manner. Individual pulses of the excitation light 26 are repeatedly emitted, and the corresponding fluorescent light 30 is received synchronously. Pulsed operation saves energy and extends the service life. Since fouling is a slow process, a low update rate of the measurement signal is sufficient. To minimize the effect of photobleaching, the light transmitter is switched off between individual measurements. The individual measurement point can either be a measured value integrated over a defined time interval or the average of individual measurements pulsed in quick succession. Typical and reasonable values ​​are, for example, 1000 measurement points with a 2 µs pulse duration and a 10 µs cycle time, which results in a measurement time of 10 ms per measurement. Such a measurement is then repeated on a time scale of seconds.

[0031] Preferably, at least one or more additional sensors 42 are provided which can measure ambient conditions at the location of the deposits 12. These ambient conditions can include: a moisture content, a temperature, a turbidity, a color, a flow rate of the medium, a flow parameter such as the flow velocity of the medium, a pH value and / or a viscosity. In addition to the direct measurement of ambient conditions, these can also be transmitted from a system control system to the device 10 according to the invention. For example, in the form of information about which process step is currently taking place. The sensor does not need to know which step is involved exactly; it is sufficient if this is known in abstract form as a number. During the teaching process, it is then possible to observe how the sensor signal behaves during a process step X. It should be noted that the signal change will occur with a time delay.The sensor must take the corresponding time scale into account (e.g. process step X: Detergent y is pumped through the system. It takes time for the detergent to flow past the sensor installation location.).

[0032] The following explains the operation of the device according to the invention and thus also the design of the evaluation unit. Two cases must be distinguished in this explanation. One is the case where the device 10 serves to detect the deposits 12 with the aim of issuing a start signal for the rinsing process, and the other, reverse case, namely to detect when the deposits have been sufficiently removed during a rinsing process with the aim of issuing a stop signal for the rinsing process.

[0033] First, it's important to recognize that the fluorescent light produced depends on a variety of factors, referred to below as environmental parameters. Here are some examples: Every organic material, including the deposits 12, exhibits what's known as photobleaching. This is the bleaching of the material over time; if excitation light repeatedly falls on the material, the fluorescence becomes increasingly less intense over time.

[0034] Furthermore, the fluorescence activity of a material can depend on the moisture content of the material, the temperature of the material, the color of the material, the pH value of the material, the viscosity of the material, and other environmental parameters. These dependencies can vary greatly. Some dependencies are continuous, meaning the fluorescence changes continuously with the corresponding environmental parameter (all other conditions being equal). Other dependencies can be abrupt, such as when changing from one medium to another, e.g., rinsing fluid.

[0035] Other environmental parameters can be things like the size of the flow through the pipe, which can be represented by flow parameters such as the flow velocity.

[0036] One aim of the invention is to obtain reproducible statements about the presence and extent of the deposits 12 despite all these dependencies.

[0037] For this purpose, the evaluation unit 34 is designed to learn correction values ​​for at least one, preferably for several or all environmental parameters in a learning process.

[0038] A typical learning process of the invention comprises the steps: - Step a) Inserting the optical fiber 24 of the device 10 into a wall 19 of the line 18 carrying the medium 16, - Step b) Passing the medium 16 through the line 18, - Step c) Emitting the transmitted light 26 with constant power and guiding the transmitted light 26 through the optical fiber 24, - Step d) Detecting an environmental parameter, for example the temperature via the sensor 42, - Step e) Receiving the fluorescent light 30 coupled into the optical fiber 24. The light detector 22 will generate corresponding received signals, which are stored in the memory 38 together with the environmental parameter, i.e. the temperature during the measurement.

[0039] Steps c) to e) are now repeated for a preselected period of time, with the temperature changing. - Then, in step f), the received signals are evaluated together with the associated temperatures (as environmental parameters). For example, the intensity of the fluorescent light 30 is determined as a function of temperature.

[0040] However, since the goal is to obtain the fluorescence signal independent of temperature, in step g), correction values ​​are determined for the received signals at each temperature, which compensate for the changes in the fluorescence signal resulting from the temperature changes. This results in a signal that is independent of the ambient parameter 'temperature'.

[0041] In step h), the correction values ​​are stored together with the environmental parameters so that the corresponding correction values ​​are available for later measurements when the temperature is known.

[0042] Analogous procedures can be carried out for other environmental parameters.

[0043] Further learning can involve the medium itself. This means, for example, that the measurement signal is observed when various cleaning media (acids, alkalis, etc.) flow through the system. This can be done both with and without deposits. This allows for learning effects such as the swelling of deposits.

[0044] To obtain the correction values ​​for photobleaching, it is recommended in the previously described learning process after step b) to first emit the transmitted light 26 at constant power in a step b1) and initially only to observe the generated fluorescent light 30. As the deposits 12 grow, the fluorescence will increase. From a certain thickness of the deposits 12, the fluorescent light intensity will no longer increase, because then the deposits 12 have grown to such an extent that saturation occurs. Only then can one assume a constant fluorescent light if there were no photobleaching. However, since photobleaching does occur, the fluorescent light intensity will decrease over time (ambient parameters). This decrease is recorded with the learning process as described above, and corresponding correction values ​​are determined to compensate for this decrease.

[0045] The determination of correction values ​​can also be carried out using neural networks trained using machine learning.

[0046] Using the correction values ​​thus determined, the method according to the invention can now be implemented. The method comprises the following steps: In a step i), the device 10 is inserted into the wall 19 of the line 18 carrying the medium 16, which means that the optical fiber 24 is inserted into the wall 19 so that the deposits 12 can be deposited on the front end 28 of the optical fiber 24.

[0047] In a step ii), the medium 16 is passed through the line 18.

[0048] In a step iii), the transmitted light 26 is emitted and coupled into the optical fiber 24 and guided via this to the end 28 of the optical fiber 24 into the line 18.

[0049] In a step iv), the environmental parameter(s) is / are recorded, for example the time is measured or the temperature is determined or the like.

[0050] In a step v), the fluorescent light 30, which is coupled into the optical fiber 24, is guided to the light detector 22, received there, and corresponding reception signals are generated, which are guided to the evaluation unit 34.

[0051] In a step vi), the received signals are corrected in the evaluation unit 34 using the correction values ​​determined in the learning process.

[0052] Finally, in step (vii), the corrected received signals are evaluated to provide information as to whether and, if so, to what extent deposits 12 are present on the inner surface 14 of the line 18. Depending on this, a corresponding deposit signal is then output at the output 40, representing a measure of the deposits 12 in the line 18.

[0053] Preferably, a zero signal is calibrated in an additional step, whereby the zero signal corresponds to a clean line condition. For this purpose, a purging process is initiated before step ii), starting from a clean line. During the purging process, transmitted light 26 is emitted and corresponding fluorescent light 30 is recorded. Upon completion of the purging process, which continues until no more deposits remain, which must be determined by other means, the received signal then present is defined as the zero signal. Future purging processes can then be terminated prematurely, namely when the zero signal has been reached. This means considerable time savings and lower resource consumption.

[0054] In addition to the external environmental parameters mentioned above, other parameters can also be taken into account, for which correction values ​​can also be found. Such parameters can be substance parameters that reflect the substance content of the medium 16. These parameters can be, for example, a protein content, a fat content, or the like. QUOTES CONTAINED IN THE DESCRIPTION

[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions. Cited patent literature

[0000] DE 102016109819 B4

[0002]

Claims

[1] Device for detecting deposits (12) on a surface which is an inner surface (14) of a container (18) containing a medium (16), in particular a line (18) through which a medium flows, with - a light transmitter (20), - a light detector (22) for generating reception signals as a function of the incidence of light on the light detector (22), - at least one optical fiber (24) which guides light from and / or to the sensor, the end (28) of which faces away from the sensor forming the surface on which the deposits (12) are deposited, - a control and evaluation unit (34) for controlling the light transmitter (20) and evaluating the received signals, - wherein the transmitted light (26) emitted by the light transmitter (20) generates fluorescence in the deposits (12) and the fluorescent light (30) is detectable by the light detector (22), characterized by , - that the evaluation unit (34) is designed to learn correction values ​​for at least one environmental parameter, which represents an environmental condition for the deposits (12), in a learning process and - that the evaluation unit (34) is designed to correct the received signals with the correction values ​​and - that the evaluation unit (34) is further designed to generate a deposit signal representing the presence of deposits (12) from the corrected received signals and to output it at an output (40). [2] Device according to claim 1, characterized by that the correction values ​​depend on the environmental parameters. [3] Device according to one of the preceding claims, characterized by that the deposition signal is a measure of the thickness of the deposit. [4] Device according to one of the preceding claims, characterized bythat a first environmental parameter is time and first correction values ​​take into account a decrease in the received signal over time due to photobleaching. [5] Device according to one of the preceding claims, characterized by that further environmental parameters indicate a moisture content, a temperature, a turbidity, a color, a flow rate, a flow parameter such as flow velocity, a pH value and / or a viscosity and corresponding correction values ​​take into account a change in the received signal with a change in the respective environmental condition. [6] Device according to one of the preceding claims, characterized by that a substance parameter represents a substance content of the product, such as protein content, fat content or the like, and corresponding correction values ​​take into account a change in the received signal with a change in the substance parameter. [7] Device according to one of the preceding claims, characterized bythat one or more sensors are provided to measure the environmental parameter(s). [8] A teaching method for teaching correction values ​​for a sensor according to one of the preceding claims, comprising the steps: - a) Inserting the device into a wall of the pipe carrying the medium, - b) Passing the medium through the pipe, - c) Emitting the transmitted light with constant power and guiding the transmitted light through the optical fiber, - d) recording the environmental parameter(s), - e) receiving the fluorescent light coupled into the optical fiber and generating and storing the received signals in a memory together with the environmental parameter(s), - Repeat steps c) to e) for a preselected period of time, with changing environmental parameters, - f) evaluating the received signals and / or evaluating the course of the received signals with the changing environmental parameters and - g) determining correction values ​​for the received signals for the associated environmental parameter(s), - h) Saving the correction values ​​together with the environmental parameters. [9] Learning method according to claim 8, characterized by that after step b) first in a step - b1) transmitted light is emitted with constant power and the transmitted light is guided through the optical fiber and the fluorescent light coupled into the optical fiber is received and corresponding received signals are generated and only when the received signal no longer increases, - then follow steps c) to f). [10] Learning method according to claim 8 or 9, characterized bythat the correction values ​​are determined using a neural network that has previously been trained in a variety of learning procedures for the various parameters using machine learning methods. [11] Method for operating a device (10) according to one of the preceding claims 1 to 7, for detecting deposits (12) on a surface which is an inner surface (14) of a container (18) containing a medium (16), in particular a line (18) through which a medium flows, comprising the steps: - i) inserting the device (10) into a wall (19) of the container (18), - ii) passing the medium (16) through the container (18), - iii) emitting the transmitted light (26) and guiding the transmitted light (26) through an optical fiber (24), - iv) recording the environmental parameter(s), - v) receiving the fluorescent light (30) coupled into the optical fiber (24) and generating received signals, - vi) evaluating the received signals by correcting them with the correction values ​​determined in the learning process for the environmental parameter(s) detected, - (vii) evaluating the corrected received signals and outputting a deposit signal representing the presence of deposits (12) in the container (18). [12] Method according to claim 11, characterized by that before step ii) a zero signal is calibrated, whereby the zero signal corresponds to a clean line condition and for calibration, starting from a clean container, a rinsing process is started and during this process received signals are recorded and at the end of the rinsing process the then existing received signal is defined as the zero signal.

Images