Condition monitoring of a cutting unit in a food packaging apparatus

Abstract

The state of a cutting unit (165a, 165b) in an apparatus for producing liquid food packages is monitored based on a time series of measurement values from a sensor (30), such as a pressure transducer, arranged to measure a cutting resistance of a cutting blade (166a, 166b) in the cutting unit when actuated to perform a cut to sever food-containing packages (106) from each other. The monitoring comprises generating a resistance time profile of the measurement values, detecting at least one predefined feature in the resistance time profile, determining a respective phase value of the predefined feature, and determining the state of the cutting unit (165a, 165b) from a set of input values comprising the respective phase value. The state can represent a degree of wear of the cutting blade (166a, 166b).

Description

Technical Field

[0001] This disclosure generally relates to food packaging machines for producing liquid food packaging, and more particularly to the condition monitoring of cutting units in food packaging machines. Background Technology

[0002] The industrial production and packaging of liquid foods are automated, involving advanced process control of food packaging machinery to achieve mass production. The safe and reliable operation of food packaging machinery is of paramount importance, as operational malfunctions and resulting production stoppages can have a profound impact on production costs and product quality. Early detection of operational malfunctions is crucial to avoiding performance degradation and damage to machinery or personal safety.

[0003] Food packaging machines typically include a cutting unit with one or more blades for cutting packaging material during the production of packages containing liquid foods. Over time, these blades wear down and need to be replaced. Typically, blades are replaced periodically based on runtime. However, this can lead to premature blade replacement, causing unnecessary production downtime, or premature replacement, potentially resulting in the rejection of a large number of packages due to insufficient quality.

[0004] Existing technology includes EP1666362, which proposes to measure the cutting resistance of a blade by means of a pressure sensor and to monitor the condition of the cutting blade by determining the pressure difference between the maximum resistance pressure during the cutting step and the constant resistance pressure after the maximum resistance pressure, and to compare the pressure difference with a reference value.

[0005] Similarly, WO2017 / 102864 proposes detecting the pressure in the hydraulic system used to actuate the cutting blade. When the pressure detected exceeds a predetermined threshold while the cutting blade is cutting through packaging material, it indicates that the cutting blade needs to be replaced.

[0006] While these proposed techniques can be used to detect the need for blade or cutting insert replacements in a well-controlled testing environment, they may lack sufficient reliability for installation in a real-world production environment. They also cannot identify other potential failure scenarios within the cutting unit. Summary of the Invention

[0007] One objective is to overcome, at least partially, one or more limitations of the existing technology.

[0008] One objective is to provide an alternative technology for monitoring the status of the cutting unit in a liquid food packaging machine.

[0009] A further objective is to provide a technology that achieves high reliability in production environments.

[0010] One or more of these objectives, and others that may become apparent from the following description, are achieved, at least in part, by a method for monitoring the state of a cutting blade, a computer-readable medium, a monitoring device, and an apparatus for producing liquid food packaging, as described in the independent claims, embodiments of which are defined by the dependent claims.

[0011] A first aspect of this disclosure is a method for monitoring the state of a cutting unit in an apparatus for producing liquid food packaging. The apparatus includes a cutting unit configured as a roll of packaging material in the form of a vertically sealed tube, the tube being filled with liquid food, forming a transverse seal within the tube, and the food-containing packaging being cut apart from each other by a cutting blade in the cutting unit. The method includes: obtaining a time series of measured values ​​from a sensor arranged to measure the cutting resistance of the cutting blade when actuated to cut the corresponding transverse seal; processing the time series of measured values ​​to generate a resistance-time curve; detecting at least one predefined feature in the resistance-time curve; determining a corresponding phase value of the at least one predefined feature within the resistance-time curve; and determining the state of the cutting unit based on a set of input values ​​including the corresponding phase value.

[0012] The first aspect is based on the following findings from extensive experimentation: the characteristic phase (“time”) in the resistance-time curve responds to the state of the cutting unit, including the degree of wear on the cutting blade. Therefore, the first aspect provides an alternative technique for monitoring the state of cutting units in a running packaging machine. The insight gained from using phase to determine the state of the cutting unit opens up the possibility of determining the state based on a combination of phase values ​​and further input values. This can be represented by other evaluation parameters besides phase, such as magnitude, time variation, or variability, to further improve the robustness of monitoring and achieve high reliability in production environments. The use of phase further opens up the possibility of detecting other failure conditions of the cutting unit, such as incorrect cutting performance or timing, or increased friction within the cutting unit.

[0013] A second aspect of this disclosure is a computer-readable medium containing computer instructions that, when executed by a processor, cause the processor to perform the method of the first aspect or any embodiment thereof.

[0014] A third aspect of this disclosure is a monitoring device. The monitoring device includes a signal interface for connection to a sensor arranged to measure and output a time series of measurements representing the cutting resistance of the cutting blade in an apparatus for producing liquid food packaging, as the cutting blade is actuated to cut a corresponding seal formed in a tube filled with liquid food, and logic configured to control the monitoring device to perform the method of the first aspect or any embodiment thereof.

[0015] A fourth aspect of this disclosure is an apparatus for producing liquid food packaging. The apparatus includes a cutting unit configured as a roll of packaging material in the form of a vertically sealed tube, the tube being filled with liquid food to form a transverse seal within the tube, and the transverse seal being cut by a cutting blade in the cutting unit to separate the food-containing packages from each other. The apparatus further includes: a sensor arranged to measure and output a timing value representing the cutting resistance of the cutting blade when actuated to cut the corresponding transverse seal; and a monitoring device of the third aspect or any embodiment thereof.

[0016] Other objects, embodiments, features and aspects of the subject matter of this disclosure will become apparent from the following detailed description and accompanying drawings. Attached Figure Description

[0017] The implementation scheme will now be described by way of example with reference to the attached diagram.

[0018] Figure 1A This is a side view of a roll-fed carton packaging machine based on an example. Figure 1B Show through in perspective view Figure 1A The material flow of the packaging machines, and Figure 1C yes Figure 1A A side view of an exemplary sealing mechanism in a packaging machine.

[0019] Figure 2A and Figure 2B These are exemplary resistance-time characteristic curves for sharp and worn cutting blades, respectively (by...). Figure 1C The pressure sensor in the sealing mechanism shown measures the pressure.

[0020] Figures 3A to 3C This is a block diagram of an exemplary monitoring device according to some implementation schemes.

[0021] Figure 4 This is a flowchart of an exemplary method for monitoring the condition of cutting blades in a packaging machine, according to some implementation schemes.

[0022] Figures 5A to 5C An example of the evaluation parameters for the resistance-time curve is shown.

[0023] Figure 6 This is a graph of data points extracted from the resistance-time curves of sharp and worn cutting blades (given by three exemplary evaluation parameters and divided into two clusters by a threshold surface). Detailed Implementation

[0024] The embodiments will now be described more fully below with reference to the accompanying drawings, which illustrate some, but not all, of the embodiments. In fact, the subject matter of this disclosure may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure may meet the requirements of applicable law.

[0025] Furthermore, it should be understood that, where possible, any advantages, features, functions, apparatus and / or operational aspects of any implementation described and / or contemplated herein may be included in any other implementation described and / or contemplated herein, and / or vice versa. Furthermore, where possible, any term expressed herein in the singular is also intended to include the plural form and / or vice versa, unless expressly stated otherwise. As used herein, “at least one” shall mean “one or more” and these phrases are intended to be interchangeable. Thus, the terms “a” and / or “an” shall mean “at least one” or “one or more”, even though the phrases “one or more” or “at least one” are used herein. As used herein, unless the context requires otherwise by express language or necessary implication, the word “comprising” or variations such as “containing” or “including” are used in an inclusive sense, specifying the presence of the stated features but not excluding the presence or addition of additional features in various implementations. As used herein, the term “and / or” includes any and all combinations of one or more of the listed associated elements. As used herein, the term “a group” of elements is intended to imply the provision of one or more elements.

[0026] As used in this article, “liquid food” means any food that is non-solid, semi-liquid, or pourable at room temperature, including beverages such as fruit juice, wine, beer, soda, as well as dairy products, sauces, oils, cream, custards, soups, pastes, etc., and solid foods in liquid form, such as beans, fruits, tomatoes, stews, etc.

[0027] As used herein, “packaging” means any packaging or container suitable for sealing and containing liquid food, including but not limited to containers formed of cardboard or packaging laminates (e.g., cellulose-based materials) and containers made of or containing plastic materials.

[0028] Similar reference markers refer to similar elements from beginning to end.

[0029] Figure 1AThis is a side view of a machine or apparatus 10 for packaging liquid food. Machine 10 is an example of a roll-feed carton-based packaging system. In the example shown, machine 10 includes a feeding section 11 for holding one or more rolls of carton-based sheet material; a micro-injection section 12 for applying molding opening devices to the sheet material; a bath section 13 for sterilizing the sheet material; a sterile chamber 14; a tube forming section 15 for forming the sheet material into tubes; a filling section 16 for filling, sealing, and cutting tubular material; and an discharge section 17 for outputting the package.

[0030] Figure 1B A general illustration shows a continuous packaging system that can be set up for liquid foods. Figure 1A The operating principle of machine 10 is as follows: Packaging material is conveyed to the factory where the packaging machine is installed in the form of sheet roll 100. Machine 10 unrolls the packaging material and feeds it into a bath 102, for example, containing hydrogen peroxide, to sterilize the packaging material. Alternatively, low-pressure electron beam (LVEB) technology can be used for sterilization. After sterilization, the packaging material is formed into a tube 104. More specifically, the longitudinal ends are continuously attached to each other in a process commonly referred to as longitudinal sealing. When the tube 104 has been formed, the machine fills it with liquid food. The machine forms package 106 from the food-containing tube 104 by performing a transverse seal at one end of the tube 104 and cutting it off when the seal is formed. The machine can perform different forming operations during and / or after the transverse seal to form package 106.

[0031] The filling section 16 of machine 10 is, for example Figure 1C The diagram is shown in more detail below. To produce package 106 from tube 104, forming flaps 160a and 160b can be used in combination with sealing grippers 162a and 162b. Each sealing gripper 162a and 162b includes sealing devices 164a and 164b; and cutting units 165a and 165b with blades 166a and 166b (also referred to herein as "cutting blades") for separating the formed package from tube 104. Each combination of forming flaps 160a and 160b and sealing grippers 162a and 162b defines a corresponding mechanical unit 1a and 1b that moves with the tube. Figure 1CThe first and second stages of the packaging formation process are shown. In the first stage, forming flap 160a begins to shape the tube into the package shape, and the tube is filled with liquid food, for example, via a pipe (not shown) extending into the tube. Also in the first stage, sealing jaws 162a are operated to form a lateral seal using sealing device 164a. In the second stage, forming flap 160b is held in place to form the package shape. Also in the second stage, sealing jaws 162b are operated to form a lateral seal using sealing device 164b. When the lateral seal is complete, a cutter 166b is operated to cut off the lower part of the tube 104, where both ends are closed by the lateral seal.

[0032] In the following text, reference will be made to Figures 1A to 1C The exemplary machine 10 described herein illustrates an implementation of a technique for monitoring the status of a cutting unit in a packaging machine. In some implementations, the purpose of monitoring is to identify and signal the need to replace the corresponding blade when it is no longer considered sufficiently sharp. In these implementations, the status of the cutting unit thus indicates the condition of the blade (e.g., in terms of its wear). In other implementations, the status of the cutting unit may indicate whether the seal has been properly cut by the cutting unit, whether the cutting unit has performed a cut at the appropriate time, or whether the blade has experienced increased mechanical friction on its path of travel. In yet another implementation, the status may generally indicate whether the cutting unit is operating normally.

[0033] Monitoring is performed on sensors 30 associated with the corresponding blades 166a and 166b. Figure 1C The sensor 30 operates by measuring the sensor signal (“measurement signal”). The sensor 30 is arranged to measure the cutting resistance applied to the blades 166a and 166b when the blades 166a and 166b are operated to cut one of the transverse seals. In some embodiments, the sensor 30 is directly or indirectly attached to the blades 166a and 166b and includes an accelerometer, vibration sensor, torque sensor, strain gauge, or thin-film force sensor. In the embodiments presented in more detail below, the sensor 30 is a pressure sensor, also referred to as a pressure transducer, arranged to measure the hydraulic pressure in a hydraulic circuit 130 operated by the machine 10 to actuate the corresponding blades 166a and 166b to perform a cutting operation through the transverse seal. It is believed that using such a pressure sensor 30 facilitates robust monitoring.

[0034] Figure 2A The example demonstrates the... Figure 1C The time curve 301 of the cutting resistance of a sharp cutting blade is measured by one of the pressure sensors 30. Figure 2BThe corresponding time curves 301 for the wear of the cutting blade are also shown as examples. These representative time curves demonstrate that the increase in cutting blade wear has a profound impact on time curve 301, especially during the middle portion (IP) of time curve 301. Figures 2A to 2B The intermediate portion of IP, roughly represented by the figure, exhibits a sharp increase in pressure (cutting resistance) up to a first (positive) peak 301A, followed by a pressure drop up to a second intermediate (negative) peak 301B, where the pressure rises sharply again. The first peak 301A and the second peak 301B have been found to correspond to the penetration of the blades 166a and 166b into the transverse seal. The first peak 301A occurs precisely when the blade enters the seal (i.e., when cutting the seal begins). The second peak 301B is believed to be caused by the return of elastic energy that occurs after the blade has completely penetrated the seal (i.e., when the seal is cut through).

[0035] Figures 2A to 2B The most significant change in the time curve is the increase in the magnitude of IP in the middle section. This is likely at least partly due to the need to increase cutting force as the blade's sharpness decreases.

[0036] After a detailed analysis of numerous time curves of the packaging machine 10 operating under different settings and conditions, it has been found that the timing of various features (e.g., peaks 301A, 301B) in the intermediate IP portion also exhibits statistically significant variations. Time is also referred to as "phase" in this document and is the occurrence time of the corresponding feature relative to a known reference time point, such as the trigger signal generated by the hydraulic system for operating the corresponding blades 166a, 166b. Figures 2A to 2B The symbol is schematically indicated as RT. After extensive experimentation, the applicant has found that the phase of one or more features in the middle portion IP of time curve 301 is indeed useful for robust and reliable determination of the state of cutting units 165a, 165b. For example, it has been found that the phase of the features in curve 301 increases with increasing wear of the cutting blade. This may be at least partly attributed to the increased travel distance of the cutting blade with wear and / or the longer time required for a duller cutting blade to cut into and pass through the seal. Furthermore, it should be understood that the phase of the features in curve 301, including but not limited to peaks 301A, 301B, will change with the time of the cutting operation, and therefore this phase can be used to determine whether the cutting unit performed the cut at the appropriate time. Similarly, it should be understood that the phase of the features in curve 301, including but not limited to peaks 301A, 301B, will change with the amount of mechanical friction experienced by the blade along its travel path. It should also be recognized that incomplete cutting of the seal will appear as a phase transition of one or more features in curve 301. Regardless of the type of state to be determined, improved robustness of monitoring can be achieved by analyzing the phase in conjunction with one or more other evaluation parameters, which will be further described below.

[0037] In the following text, reference will be made to Figures 3 to 4. Figure 6 An exemplary implementation for detecting the wear condition of a cutting blade is presented.

[0038] Figure 3A An exemplary monitoring device 20 is shown, comprising a first signal interface 21 configured to connect via wired or wireless means to a sensor 30 for measuring cutting resistance. In the illustrated embodiment, the monitoring device 20 is implemented on a software-controlled computing device comprising a processor 22 and computer memory 23. The processor 22 may include, for example, one or more of the following: a CPU (“Central Processing Unit”), a DSP (“Digital Signal Processor”), a microprocessor, a microcontroller, a GPU (“Graphics Processing Unit”), an ASIC (“Application-Specific Integrated Circuit”), a combination of discrete analog and / or digital components, or some other programmable logic device, such as an FPGA (“Field-Programmable Gate Array”). A control program containing computer instructions may be stored in memory 23 and executed by processor 22 to perform any monitoring method as described or implied herein. The control program may be provided to the monitoring device 20 on a computer-readable medium, which may be a tangible (non-transitory) product (e.g., magnetic media, optical disc, read-only memory, flash memory, etc.) or a propagating signal. The monitoring device 20 further includes a second signal interface 24 which may provide output data to module 18 of machine 10 via wired or wireless means. The output data allows module 18 to display monitoring results to the operator of machine 10, generate indication signals to warn the operator of current or early fault conditions of the cutting unit in machine 10, store monitoring results in a data log, and so on. Module 18 may also include control equipment for machine 10 and be operable to command machine 10 to stop when an emergency repair or maintenance need for the cutting unit is detected. In some embodiments, monitoring device 20 is a physical device located at or near machine 10. In other embodiments, monitoring device 20 is implemented by a remote server, such as through cloud computing.

[0039] An exemplary method for monitoring the condition of knives or cutting blades in a packaging machine will refer to Figure 4 The flowchart description in the document, and can be derived from... Figure 1A-1C Machine 10 related Figure 2A The monitoring equipment 20 in the middle is executed.

[0040] exist Figure 4 In the example, step 401 obtains the time series of the measured values ​​from sensor 30.

[0041] Step 402 processes the time series of the measured values ​​to generate a resistance-time curve, for example... Figures 2A to 2BAs illustrated in the example. The resistance-time curve represents the force applied to the cutting blade when it is actuated to cut through the seal (also referred to herein as the "cutting cycle"). Step 402 may, for example, include data sampling, A / D conversion, baseline correction, and reference time (see [reference]). Figures 2A to 2B This involves one or more of the following: time synchronization, averaging, etc., related to RT (Reference Time). Time synchronization can be performed to correlate corresponding measurements with a time point relative to a reference time point. Averaging can include time alignment and average measurements over multiple consecutive cut cycles, thus forming an averaged time curve. Averaging can improve the robustness of monitoring.

[0042] Step 403 processes the resistance-time curve generated in step 402 to detect or identify at least one predefined feature. See below for reference. Figures 5A to 5C Further examples of predefined features are presented.

[0043] Step 404A determines the phase value of each predefined feature within the resistance-time curve.

[0044] Step 405 determines the condition of the cutting blade based on a set of input values ​​including a phase value. In some embodiments, step 405 may include comparing the input values ​​with corresponding thresholds to determine the condition. The condition may be determined in at least two different condition names, such as "acceptable" and "unacceptable". Additional condition specifications may indicate that replacement is imminent but not immediately required. In some embodiments, step 405 may generate condition values ​​that indicate the quality of the cutting blade, for example, on a continuous or discrete scale (e.g., from 1 to 10).

[0045] Figures 5A to 5C It is a diagram of the resistance-time curve 301, and indicates an example of a predefined feature and the corresponding evaluation parameters that may be included in the input value group in step 405. Figure 5A This represents the first peak 301A and its phase value PH1, and the second peak 301B and its phase value PH2. Furthermore, the phase value PH3 is given by the time of the maximum time derivative of curve 301 or the intermediate portion IP (see [reference]). Figures 2A to 2B ).exist Figure 5A In the lower part, signal 301' shows the absolute value of the time derivative as a function of time along curve 301. Figure 5AThis also includes the phase value PH4, which is given by the positive peak value 301C following the maximum time derivative. The positive peak value 301C was also found to be related to the condition of the cutting blade. It is currently believed that the undulating structure following peak value 301C is caused by the "hammering effect" in the hydraulic system when the cutting blade penetrates the seal. The hammering effect, also known as hydraulic shock, is a pressure fluctuation or wave caused when a moving fluid is forced to stop or suddenly changes direction. Therefore, the positive peak value 301C is influenced by the negative peak value 301B and may therefore also be affected by the condition of the cutting blade. Figure 5C Another example of the phase value PH5 is shown, which is given by the time point of the maximum magnitude value within curve 301.

[0046] It should be understood that, depending on the representation of the resistance time curve, step 402 can detect predefined features as, for example, local peaks, minimums or maximums, or minimum or maximum time derivatives.

[0047] If curve 301 is highly structured (e.g., as...) Figures 5A to 5C As shown), step 403, which involves locally detecting one or more peaks 301A-301C within curve 301 (e.g., within the middle portion IP), may be challenging. In some embodiments, such peaks can be detected by first processing curve 301' to detect the maximum time derivative and then processing curve 301' to detect peaks associated with the time point of the maximum time derivative. Peaks 301A and 301B are typically found to occur before the maximum time derivative, and peak 301C occurs after the maximum time derivative. In some embodiments, step 403 can therefore search for one or more peaks 301A-301C within a time window associated with the time of the maximum time derivative.

[0048] like Figure 4 As shown in the dashed box, the method may include one or more further steps 404B-404E for determining additional input values ​​to be included in a set of input values, along with phase values, for use in step 405. In some embodiments, the additional input values ​​may include one or more magnitude values, one or more variation values, one or more internal variability values, one or more inter-variability values, or any combination thereof.

[0049] In some embodiments, method 400 may include step 404B determining one or more magnitude values ​​of curve 301. The corresponding magnitude values ​​represent the amount of cutting resistance given by curve 301. As mentioned above, at least for some characteristics of curve 301, the magnitude may increase with increasing cutting blade wear. In some embodiments, step 404B determines the magnitude of curve 301 at a selected time point relative to a predefined characteristic detected in step 403. For example, as... Figure 5AAs shown, the magnitude value M1 represents the cutting resistance at peak 301A, the magnitude value M2 represents the cutting resistance at peak 301B, the magnitude value M3 represents the cutting resistance at the time point of the maximum time derivative, and the magnitude value M4 represents the cutting resistance at peak 301C. The corresponding magnitude values ​​M1 to M4 can be given by individual amplitude values ​​(cutting resistance) in curve 301 or by the average or median of amplitude values ​​within small time windows around their respective characteristics. Further examples of the magnitude values ​​are shown in... Figure 5C The diagram shows magnitudes M5 and M6 representing the average and median amplitude values ​​of curve 301, respectively, and magnitudes Q1 and Q3 representing the first and third quartiles of the amplitude values ​​of curve 301, respectively. In a variation, the intermediate portion ( Figures 2A to 2B The amplitude value within the IP (in the calculation) is any one of the magnitude values ​​M5, M6, Q1, and Q3. Another example is shown... Figure 5C In the figure, the magnitude value MAX represents the maximum amplitude value in curve 301. Other magnitude values ​​not shown include the sum of positive or negative values ​​within curve 301 or a subset thereof (e.g., IP), impulse factors, or crest factors. The impulse factor can represent the magnitude of the peak divided by the variability in curve 301 (or IP) (e.g., given as RMS). The crest factor can represent the maximum amplitude divided by the variability in curve 301 (or IP). Many alternatives are conceivable. It can be noted that the magnitude values ​​do not need to be determined in the time domain, but can be determined in the frequency domain. For example, one or more magnitude values ​​can be given by the amplitude representing at least one basis function of curve 301, such as harmonic frequencies obtained by Fourier analysis of curve 301, wavelets obtained by wavelet analysis, or intrinsic mode functions (IMFs) obtained by empirical mode decomposition (EMD) of curve 301.

[0050] In some implementations, method 400 may include step 404C of determining one or more variation values ​​of curve 301. The corresponding variation values ​​represent curve 301 (or IP, Figures 2A to 2B The magnitude of the time variation within the selected time point. In some implementations, step 404B determines the time derivative in the curve of the predefined feature detected in step 403 relative to the selected time point. For example, the change value could represent the magnitude of the maximum time derivative, such as... Figure 5A As shown in C1, or the magnitude of the slope on either side of any peak 301A to 301C. Another example of the variation value represents the sum of the time derivatives within curve 301 or a subset thereof (e.g., IP).

[0051] In some embodiments, method 400 may include step 404D, which determines at least one internal variation value, representing variability within curve 301 or a subset thereof (e.g., IP). Variability can be represented by any conventional variability measure, including but not limited to RMS (root mean square), variance, standard deviation, or any variation thereof. In at least some embodiments, the internal variation value may be associated with the condition of the cutting blade. For example, it may increase with increasing wear. Figure 5B In this context, an example of internal variability is designated as V1 and represents the RMS of the amplitude value in curve 301.

[0052] In some embodiments, method 400 may include step 404E, which determines at least one inter-variance value representing the variability among multiple curves 301 generated for different cutting cycles during operation of the packaging machine 10. Variability can be represented by any conventional variability measure, including but not limited to RMS, variance, standard deviation, or any variation thereof. In some embodiments, the inter-variance value may represent the variability of phase values, magnitude values, change values, or internal variation values. Typically, to improve processing efficiency, step 404E may calculate the inter-variance value of the evaluation parameter values ​​previously generated by the method, for example, if implemented in method 400, through step 404A, or alternatively any of steps 404B-404D. In a variation, step 404E may calculate the internal variation value by processing the time series of curves 301.

[0053] Figure 3B It includes those configured to execute Figure 4 A block diagram of an exemplary monitoring device 20 illustrating the logic of method 400. In the example shown, the logic comprises a set of modules or units 201, 202. The corresponding modules 201, 202 can be implemented using hardware or a combination of software and hardware. For example, regarding... Figure 3A As described, the monitoring device 20 can be implemented on a software-controlled computing device. Figure 3B In this process, preprocessing module 201 is configured to perform steps 401-402. Module 201 is configured to operate on signal 300, which is generated by sensor 30 (…). Figure 1C , 3A The module generates and includes a time series of measured values, and outputs a resistance-time curve 301. The analysis module 202 is configured to perform steps 403, 404A, and 405, and optionally any one of steps 404B-404E. The module 202 is configured to process the curve 301 or the time series of curve 301 generated by the module 201 to determine the condition of the cutting blade and output a corresponding index 302, which thus indicates the current condition of the cutting blade.

[0054] Figure 3C This is a block diagram of an exemplary analysis module 202. Analysis module 202 includes calculation submodules 211-215. (The last part, "through interaction with...", appears to be an unrelated fragment and is left untranslated.) Figure 4 By analogy with method 400, submodules 212-215 are optional. Submodule 211 is configured to perform steps 403 and 404A and generate one or more phase values ​​for curve 301. Submodule 212 is configured to perform steps 403 and 404B and generate one or more magnitude values ​​for curve 301. Submodule 213 is configured to perform steps 403 and 404C and generate one or more variation values ​​for curve 301. It can be noted that step 403 can differ between submodules 211-213 by detecting different predefined features. Such data can be shared between submodules and / or generated by a dedicated submodule (not shown) when two or more submodules 211-213 operate on the same predefined feature. Submodule 214 is configured to perform step 404D and generate one or more internal variation values ​​for curve 301. Submodule 215 is configured to perform step 404E to generate one or more intermediate variation values. Submodule 215 can operate on the evaluation parameter values ​​generated over time by at least one of the submodules 211-214. Analysis module 202 further includes evaluation module 216, which is configured to perform step 405 and operate on the values ​​generated by submodules 211-215 to generate and output condition index 302.

[0055] In some implementations, the evaluation submodule 216 includes a rule-based algorithm for evaluating a set of input values. This rule-based algorithm can evaluate corresponding input values ​​associated with a given threshold. If input values ​​for multiple evaluation parameters are provided to submodule 216, for example... Figure 3C As shown, the input values ​​span a multidimensional parameter space. Furthermore, the corresponding threshold can be viewed as defining one or more multidimensional surfaces that separate subspaces associated with different cutting blade conditions. Figure 6 In this diagram, each data point is given by three input values ​​extracted from the corresponding curve. Each data point is associated with a known condition of the cutting blade: "acceptable" or "unacceptable." In the parameter space shown, data points associated with acceptable conditions form clusters in the first subspace 601, and data points associated with unacceptable conditions form clusters in the second subspace 602. As shown, subspaces 601 and 602 are separated by a three-dimensional surface 603. Figure 6 The example in [the example] shows that the condition of the cutting blade can be determined by setting a threshold to define surface 603. Figure 6 (For illustrative purposes only) The input values ​​are given by the evaluation parameters A6, Q3 and PH4, where Q3 and PH4 are as defined above, and A6 is the amplitude of the sixth harmonic frequency of curve 301.

[0056] In some implementations, the evaluation submodule 216 includes a machine learning-based model trained to determine the condition of the cutting blade based on a feature vector containing multiple input values. Any suitable machine learning-based model known in the art can be used, including but not limited to neural networks such as artificial neural networks (ANNs) or convolutional neural networks (CNNs), ensemble learning methods such as random forests, support vector machines (SVMs), or any combination thereof. However, it can be noted that the input values ​​may be preprocessed by the analysis module 212 before being input into the machine learning-based model, for example, through normalization or scaling preprocessing, as is well known in the art.

[0057] While exemplary embodiments have been described with reference to determining the wear condition of the cutting blade, as those skilled in the art will understand, the foregoing teachings can be readily modified to determine another type of state of the cutting unit including the cutting blade, such as any other state of the cutting unit mentioned or implied above. For example, modifications to the exemplary embodiments may include modified selection of predefined features to be detected in the resistance-time curve (see step 403), modified set of input values ​​to be analyzed to determine the state (step 405), or modified analysis of the set of input values, such as using a modified threshold, modified training of a machine learning-based model, etc.

[0058] The following section lists some of the aspects and implementation plans disclosed above.

[0059] Project 1. A method for monitoring the state of a cutting unit in an apparatus for producing liquid food packaging, the apparatus comprising a cutting unit and configured as a roll of packaging material in the form of a vertically sealed tube, the tube being filled with liquid food, a transverse seal being formed in the tube, and the transverse seal being cut by a cutting blade in the cutting unit to cut the food-containing packaging to one another, the method comprising: obtaining (401) a time series of measurements from a sensor arranged to measure the cutting resistance of the cutting blade when actuated to cut the corresponding transverse seal; processing (402) the time series of measurements to generate a resistance-time curve; detecting (403) at least one predefined feature in the resistance-time curve; determining (404A) a corresponding phase value of the at least one predefined feature in the resistance-time curve; and determining (405) the state of the cutting unit based on a set of input values ​​including the corresponding phase value.

[0060] Project 2. According to the method of Project 1, the time series of measurements are obtained to represent the hydraulic pressure in the hydraulic circuit used to actuate the cutting blade to cut the transverse seal.

[0061] Item 3. According to the method of Item 1 or 2, at least one of the predefined features includes one or more of the following: peaks (301A, 301B, 301C) in the resistance time curve (301), the minimum or maximum value of the resistance time curve (301), or the minimum or maximum time derivative in the resistance time curve (301).

[0062] Project 4. According to the method of Project 3, wherein the peaks (301A, 301B) correspond to one of the following: entering the cutting blade (166a, 166b) of the corresponding transverse seal, or penetrating the cutting blade (166a, 166b) of the corresponding transverse seal.

[0063] Item 5. According to the method of Item 2 or 3, wherein the detection (403) includes processing the resistance time curve (301) to detect the time point of the maximum time derivative in the resistance time curve (301), and processing the resistance time curve (301) to detect one or more of the following: a negative peak (301B) before the time point, a first positive peak (301A) before the time point, or a negative peak (301B) or a second positive peak (301C) after the time point.

[0064] Project 6. The method according to any of the preceding projects further includes determining at least one order of magnitude of the (404B) resistance-time curve (301), wherein the at least one order of magnitude is included in the set of input values.

[0065] Item 7. The method according to Item 6, wherein the at least one magnitude value comprises the magnitude of the resistance-time curve at a selected time point relative to at least one predefined characteristic.

[0066] Item 8. According to the method of Item 6 or 7, wherein at least one order of magnitude value includes one or more of the following: the amplitude of at least one predefined feature, the average or median of at least one subset of the resistance time curve (301), the sum of positive or negative values ​​in at least one subset of the resistance time curve (301), the impulse factor, the peak factor, and the amplitude of at least one basis function representing the resistance time curve (301).

[0067] Project 9. The method according to any of the preceding projects further includes determining (404C) at least one change value representing the magnitude of time change within the resistance time curve (301), wherein the at least one change value is included in the set of input values.

[0068] Item 10. According to the method of Item 9, wherein at least one variation value comprises one or more of the following: the time derivative of a resistance time curve (301) at a selected time point relative to at least one predefined feature, or the sum of the time derivatives of at least one subset of the resistance time curve (301).

[0069] Project 11. The method according to any of the preceding projects further includes determining (404D) at least one internal variation value representing the variability of at least one subset of the resistance time curve (301), wherein the at least one internal variation value is included in the set of input values.

[0070] Item 12. According to the method of Item 11, at least one internal variation value contains one or more of the following: standard deviation, variance, or RMS.

[0071] Project 13. The method according to any of the preceding projects further includes determining (404E) at least one intermediate variation value representing the variability among the plurality of resistance time curves (301), wherein the at least one intermediate variation value is included in the set of input values.

[0072] Item 14. According to the method of Item 13, at least one of the inter-variance values ​​represents one or more of the following variability among the multiple resistance time curves (301): corresponding phase value, at least one order of magnitude value, at least one change value or at least one internal variation value.

[0073] Item 15. According to the method of Item 13 or 14, at least one of the variable values ​​therein contains one or more of the following: standard deviation, variance, or RMS.

[0074] Item 16. The method according to any of the preceding items, wherein the processing (402) comprises: processing the time series of the measured values ​​to generate a resistance-time curve (301) to represent the cutting resistance as a function of time.

[0075] Item 17. According to the method of any of the preceding items, wherein the determination (405) state comprises: comparing the set of input values ​​with the corresponding set of thresholds.

[0076] Item 18. According to the method of any one of Items 1-16, wherein the determination (405) state comprises: providing the set of input values ​​to the trained machine learning model (216).

[0077] Item 19. A computer-readable medium containing computer instructions that, when executed by a processor (22), cause the processor (22) to perform any of the methods of items 1-18.

[0078] Item 20. A monitoring device comprising a signal interface (21) for connection to a sensor (30) arranged to measure and output a time series of measured values ​​representing the cutting resistance of a cutting blade (166a, 166b) in an apparatus (10) for producing a liquid food package (106) when a cutting blade (166a, 166b) is actuated to cut a corresponding seal formed in a tube (104) filled with liquid food; and logic (201, 202) configured to control the monitoring device to perform the method of any one of items 1-18.

[0079] Item 21. An apparatus for producing liquid food packaging (106), the apparatus comprising cutting units (165a, 165b) configured as a roll of packaging material in the form of a vertically sealed tube (104), the tube (104) being filled with liquid food, a transverse seal being formed in the tube (104), and the transverse seal being cut by cutting blades (166a, 166b) in the cutting units (165a, 165b) to cut the food-containing packaging (106) apart from each other, the apparatus further comprising: a sensor (30) arranged to measure and output a time series of measured values ​​representing the cutting resistance of the cutting blades (166a, 166b) when actuated to cut the respective transverse seal; and a monitoring device (20) according to Item 20.

Claims

1. A method of monitoring the state of a cutting unit in an apparatus for producing liquid food packages, the apparatus comprising the cutting unit and being configured to vertically seal a web of packaging material in the form of tubes, fill the tubes with liquid food, form transverse seals in the tubes, and cut the transverse seals by means of a cutting blade in the cutting unit to separate the food-containing packages from one another, wherein, The method comprises: obtaining a time series of measurement values from a sensor arranged to measure a cutting resistance of the cutting blade when actuated to cut a respective transversal seal; processing the time series of measurement values to generate a resistance time profile; detecting at least one predefined feature in the resistance time profile; determining a respective phase value of the at least one predefined feature within the resistance time profile; determining the state of the cutting unit from a set of input values comprising the respective phase value; processing the resistance time profile to determine a condition of the cutting blade; outputting a respective indicator indicative of a respective indicator of a current condition of the cutting blade.

2. The method of claim 1, wherein the time series of measurement values is obtained to represent a hydraulic pressure in a hydraulic circuit for actuating the cutting blade to cut the transversal seal.

3. The method of claim 1 or 2, wherein the at least one predefined feature comprises one or more of: a peak value (301A, 301B, 301C) in the resistance time profile (301), a minimum or maximum value of the resistance time profile (301), or a minimum or maximum time derivative in the resistance time profile (301).

4. The method of claim 3, wherein the peak value (301A, 301B) corresponds to one of: the cutting blade (166a, 166b) entering the respective transversal seal, or the cutting blade (166a, 166b) penetrating the respective transversal seal.

5. The method of claim 3, wherein the detecting comprises processing the resistance time profile (301) to detect a time point of the maximum time derivative in the resistance time profile (301), and processing the resistance time profile (301) to detect one or more of: a negative peak value (301B) before the time point, a first positive peak value (301A) before the time point, and a negative peak value (301B) or a second positive peak value (301C) after the time point.

6. The method of any of the preceding claims 1, further comprising determining at least one magnitude value of the resistance time profile (301), wherein the at least one magnitude value is included in the set of input values.

7. The method of claim 6, wherein the at least one magnitude value comprises a magnitude of the resistance time profile at a selected time point relative to the at least one predefined feature.

8. The method of any of the preceding claims 1, further comprising determining at least one change value representing a time-varying magnitude within the resistance time profile (301), wherein the at least one change value is included in the set of input values.

9. The method of claim 8, wherein the at least one change value comprises one or more of: a time derivative in the resistance time profile (301) at a selected time point relative to the at least one predefined feature, or a sum of time derivatives of at least one subset of the resistance time profile (301).

10. The method according to the preceding claim 1, further comprising determining at least one intra-variability value representing variability in at least one subset of the resistance time curves (301), wherein the at least one intra-variability value is included in the set of input values.

11. The method according to the preceding claim 1 or 2 or 4, further comprising determining at least one inter-variability value representing variability between a plurality of resistance time curves (301), wherein the at least one inter-variability value is included in the set of input values.

12. The method according to claim 11, wherein the at least one inter-variability value represents variability between a plurality of resistance time curves (301) in one or more of: the respective phase values; at least one magnitude value of the resistance time curves (301), wherein the at least one magnitude value is included in the set of input values; at least one variation value representing a time-varying magnitude within the resistance time curves (301), wherein the at least one variation value is included in the set of input values; or at least one intra-variability value representing variability in at least one subset of the resistance time curves (301), wherein the at least one intra-variability value is included in the set of input values.

13. A computer readable medium comprising computer instructions which, when executed by a processor (22), cause the processor (22) to perform the method according to any one of claims 1-12.

14. A monitoring device comprising a signal interface (21) for connection to a sensor (30) arranged to measure and output a time series of measurement values representing a cutting resistance of a cutting blade (166a, 166b) in a device (10) for producing packages (106) of liquid food, when the cutting blade (166a, 166b) is actuated to cut a respective seal formed in a tube (104) filled with liquid food; and logic (201, 202) configured to control the monitoring device to perform the method according to any one of claims 1-12.

15. A device for producing packages (106) of liquid food, the device comprising cutting units (165a, 165b) and being configured to vertically seal a web of packaging material in the form of a tube (104), fill the tube (104) with liquid food, form transverse seals in the tube (104), and cut the transverse seals with cutting blades (166a, 166b) in the cutting units (165a, 165b) to separate food-containing packages (106) from each other, the device further comprising: a sensor (30) arranged to measure and output a time series of measurement values representing a cutting resistance of the cutting blades (166a, 166b) when actuated to cut a respective transverse seal; and a monitoring device (20) according to claim 14.

Images