How to monitor the condition of moving mechanical parts

Abstract

A condition monitoring method (1000) is disclosed that includes moving (1010) a mechanical part according to a cycle of a defined motion profile (PF), generating (1040) a first distribution (T1, p1) of measurements of forces (T) and / or measured motion parameters (p) causing acceleration when moving the mechanical part according to the defined motion profile, associating (1050) the first distribution to a first load response (L1), generating (1060) a second distribution (T2, p2) of associated registered values ​​(1070) of said forces and / or motion parameters measured subsequently when moving the mechanical part according to the cycle at a subsequent time, associating (1080) the second distribution to a second load response (L2), and determining (1100) an amount of variation in a load component of a mechanical load on the mechanical part based on a difference between the first load response and the second load response.

Description

[Technical field]

[0001] The present invention relates to a method, associated computer program product and condition monitoring system for monitoring the condition of moving mechanical parts, such as motors, pistons and other actuators employed in packaging or filling machines for liquid food processing applications. [Background technology]

[0002] Condition monitoring of machine parts in packaging or filling machines and related systems for producing sealed packaging containers for liquid or semi-liquid foods is important to configure optimal operating settings and ensure the desired performance over a period of time. Defects in the produced packaging containers can lead to suboptimal sterility performance. It is therefore desirable to develop efficient tools and procedures to identify defective operation of components in such systems that can lead to various types of defects in the produced packaging containers. The latest generation of filling machines or related equipment employed in the production of sealed packaging containers operates at very high speeds to further increase the throughput of the production line, making it difficult to accurately characterize all aspects of the performance of the packaging container production without interrupting the production line. This can lead to suboptimal performance and reduced throughput. The question is therefore how to introduce reliable control tools and control strategies with minimal impact on production while minimizing the resources required.

[0003] Moreover, the ever-decreasing tolerances of packaging and filling machines in high-throughput production lines increase the need to detect deviation trends in the behavior of machine components and make failure predictions. Current monitoring routines can compromise sterility performance before deviations are visible in the manufactured food containers. When combined with high-throughput production lines, the consequences are severe. There is also the further problem of how to pick up deviation trends from the large amount of data generated by such production lines and how to effectively identify the right maintenance actions to minimize the impact on production. Summary of the Invention [Problem to be solved by the invention]

[0004] The object of the present invention is to at least partially overcome one or more limitations of the prior art, in particular to provide an improved condition monitoring of machine parts in packaging or filling machines for liquid food processing applications, in particular to provide a method for reliable and timely detection of deviating operation or impending failures, in order to provide the operator with an effective tool for identifying and planning maintenance operations for the parts concerned without affecting production.

[0005] In a first aspect of the invention, there is provided a method for condition monitoring of a moving mechanical part in a packaging or filling machine for liquid food processing applications, comprising moving the mechanical part according to a cycle of a defined motion profile in response to a speed and position of a defined motion profile in order to accelerate the mechanical part to overcome a mechanical load, the mechanical load being achieved by the method comprising the sum of external mechanical forces on the mechanical part and contributed load components including inertia and / or moment of inertia, the method further comprising registering measurements of the forces causing said acceleration and / or measurements of motion parameters related to the movement of the mechanical part according to the defined motion profile. generating a first distribution of registered values; associating the first distribution with a first load response in a machine load model; and subsequently generating a second distribution of associated registered values ​​of the measured forces and / or the motion parameters when moving a machine part according to a cycle at a subsequent time point, wherein for said condition monitoring, determining a change in a load component of the machine load at the subsequent time point includes associating the second distribution with a second load response in the machine load model; and determining an amount of variation in a load component of the machine load based on a difference between the first load response and the second load response. The method includes assigning the load components as respective variable load parameters in a machine load model and generating a virtual load response output in response to the variable load parameters; generating a distribution of associated registration values ​​subsequently measured when the machine part is moved according to the cycle after the duration; correlating the distribution with a measured load response; determining a change in each variable load parameter in the machine load model that results in a minimized difference between the measured load response and the virtual load response output; and determining the maintenance action for the machine part after the duration of operation based on the change in the respective variable load parameters.

[0006] In a second aspect of the invention, there is provided a system for condition monitoring of a moving mechanical part in a packaging or filling machine in a liquid food processing application, comprising a processing unit configured to move the mechanical part according to cycles of a defined motion profile consisting of accelerating the mechanical part to overcome a mechanical load such that the speed and position of the defined motion profile is followed, the mechanical load comprising a sum of contributing load components consisting of external mechanical forces on the mechanical part and inertia and / or moment of inertia, the method further comprising measuring the measured forces causing said acceleration and / or the movement of the mechanical part according to the defined motion profile. the first distribution being associated with a first load response of a machine load model and subsequently generating a second distribution of associated registered values ​​of the measured force values ​​and / or operational parameters when moving the machine part according to the cycle at a subsequent time point; and determining a change in a load component of the machine load at the subsequent time point for the condition monitoring, the second distribution being associated with a second load response in the machine load model and determining an amount of variation in a load component of the machine load based on a difference between the first load response and the second load response. The processing unit is configured to determine a maintenance action for the mechanical load after the duration, including assigning the load components as respective variable load parameters in the mechanical load model, generating a virtual load response output in response to the variable load parameters, generating a distribution of associated registration values ​​subsequently measured when moving the mechanical load according to the cycle after the duration, relating the distribution to a measured load response, determining a change in each variable load parameter in the mechanical load model that results in a minimized difference between the measured load response and the virtual load response output, and determining the maintenance action based on the change in the respective variable load parameters.

[0007] In a third aspect of the invention, this is achieved by a computer program product comprising instructions which, when said program is executed by a computer, cause the computer to carry out the steps of the method according to the first aspect.

[0008] Further embodiments of the invention are defined in the dependent claims, in which features relating to the first aspect may also be implemented in the second and subsequent aspects and vice versa.

[0009] By determining first and second load responses of the moving machine part and determining a variation in a load component contributed by a mechanical load on the machine part based on a difference between the first and second load responses, an accurate and reliable classification of the condition of the machine part is provided. In this manner, condition monitoring of the moving machine part is facilitated, and deviations in operation or impending failures can be reliably and timely detected with minimal data analysis required by an operator.

[0010] Further objects, features, aspects and advantages of the present invention will become apparent from the following detailed description and drawings. [Means for solving the problem]

[0011] Embodiments of the present invention will now be described, by way of example only, with reference to the accompanying drawings, in which: FIG. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 is a schematic diagram of a condition monitoring system for moving mechanical parts of a packaging or filling machine in a liquid food processing application. [Diagram 2] FIG. 1 illustrates a motion profile defined by set speeds of a moving mechanical part. [Figure 3a] FIG. 1 shows an example of a first distribution of measured forces (T1) causing an acceleration of a mechanical part according to a defined motion profile and a measured motion parameter (p1) related to the motion of the mechanical part according to the defined motion profile. [Figure 3b]A figure showing an example of a second distribution of a measured force (T2) causing an acceleration of a machine part according to a defined motion profile and a measured motion parameter (p2) related to the motion of the machine part according to the defined motion profile. [Figure 4a] This corresponds to the illustration in FIG. 3a. [Figure 4b] FIG. 13 shows a further example of a second distribution of measured forces (T2) and motion parameters (p2) associated with the motion of a machine part according to a defined motion profile. [Figure 5a] This corresponds to the illustration in FIG. 3a. [Figure 5b] FIG. 13 shows a further example of a second distribution of measured forces (T2) and motion parameters (p2) associated with the motion of a machine part according to a defined motion profile. [Figure 6a] This corresponds to the illustration in FIG. 3a. [Figure 6b] FIG. 13 shows a further example of a second distribution of measured forces (T2) and motion parameters (p2) associated with the motion of a machine part according to a defined motion profile. [Figure 7a] This corresponds to the illustration in FIG. 3a. [Figure 7b] FIG. 13 shows a further example of a second distribution of measured forces (T2) and motion parameters (p2) associated with the motion of a machine part according to a defined motion profile. [Figure 8a] 1 is a flow chart of a method for condition monitoring of moving mechanical parts of a packing or filling machine in a liquid food processing application. [Figure 8b] 13 is another flow chart of a method for condition monitoring of moving mechanical parts of a packing or filling machine in a liquid food processing application. [Figure 8c] 13 is another flow chart of a method for condition monitoring of moving mechanical parts of a packing or filling machine in a liquid food processing application. [Figure 8d] FIG. 1 is a schematic diagram of a condition monitoring system for moving mechanical parts of a packaging or filling machine in a liquid food processing application. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Hereinafter, the present invention will be described in more detail with reference to the accompanying drawings, in which some, but not all, embodiments of the present invention are shown. The present invention may be embodied in many different forms and should not be construed as being limited to the embodiments set forth herein.

[0014] Figure 8a shows a flow chart of a method 1000 for condition monitoring of moving mechanical parts (not shown) in a packing or filling machine 300 for liquid food processing applications. The order in which the steps of the method 1000 are described and illustrated should not be construed as limiting, as steps may be performed in various orders. Figure 1 is a schematic diagram of a system 200 configured to perform the method 1000 described below. The moving mechanical parts may comprise servo motors, linear actuators, or any actuating parts that drive the motion axes of the packing or filling machine 300.

[0015] The method 1000 includes a step 1010 of moving a mechanical part according to a cycle of a defined motion profile (PF). FIG. 2 shows an example of a defined motion profile (PF) over time (t), where a set speed (v1) is a trapezoidal shape that accelerates to a defined value and then decelerates before the motion is reversed. In one embodiment, the moving mechanical part is a servo motor driving a belt in two directions. The method 1000 includes a step 1020 of accelerating the mechanical part to overcome a mechanical load to follow the speed (v) and position (speed v1 in the example of FIG. 2) of the defined motion profile (PF). The value (v1) in FIG. 2 represents the amplitude of the speed (v1) as a function of time (t). Thus, the servo motor needs to apply a torque to overcome the mechanical load and accelerate the belt according to the defined motion profile (PF). The mechanical load includes the sum of contributing load components (C), further referred to as C1, C2, C3 in the following examples, but generally referred to as load component (C) for simplicity in this disclosure. The load components (C) comprise the external mechanical forces acting on the machine components and the inertia and / or moment of inertia exhibited by the machine components. The external mechanical loads may comprise different types of friction as described below, or other external forces applied to the machine components. The moment of inertia is associated with rotating masses, i.e. rotating machine parts such as servo motors, while inertia is generally associated with other motions, e.g. linear actuators.

[0016] The method further includes registering measurements of the force (T) that causes the aforementioned acceleration according to the defined motion profile (PF) and / or measurements 1030 of the motion parameters (p) associated with the motion of the mechanical component according to the defined motion profile (PF). In the case of a rotating mechanical component such as a servo motor, the force (T) should be interpreted as the torque required to drive the mechanical load according to the defined motion profile (PF). Similarly, in the example where the mechanical component is a linear actuator, for example an actuator such as a hydraulic piston, the force (T) should be interpreted as the force required to be applied by such a linear actuator to drive the mechanical load according to the defined motion profile (PF). In both cases, the force (T) is measured by a sensor (not shown) and the measurements are recorded. Figure 3a shows an example of a first distribution of the torque (T1) applied by the mechanical component such that the set speed (v1) of the defined motion profile (PF) follows the actual speed (v2). The value (T1) in Figure 3a represents the torque amplitude (T1) as a function of time (t).

[0017] Alternatively or additionally, a motion parameter (P), e.g. a position related to the motion of the machine part according to the defined motion profile (PF), may be measured and registered. Said position may in some examples correspond to a position error (p1) as illustrated in FIG. 3a, i.e. the difference between a position set according to the defined motion profile (PF) and a resulting / actual position. The value (p1) in FIG. 3a represents the amplitude of the position error (p1) as a function of time (t). It is contemplated that various other measurable motion characteristics coupled to the motion of the current machine part, such as displacement, torque, or other force, velocity, or acceleration values ​​describing the measured motion during the defined motion profile (PF), may be determined for the purposes of implementing the method 1000.

[0018] The method 1000 comprises a step 1040 of generating a first distribution (T1, p1) of the registered values, for example as shown in the above-mentioned FIG. 3a. The method 1000 comprises a step 1050 of relating the first distribution (T1, p1) to a first load response (L1) in a mechanical load model (VM). Thus, in the packaging or filling machine 300, each machine component driving a motion axis will have an associated load response that is time-dependent, searched for in response to a motion defined by a motion profile (PF). The mechanical load model (VM) is thus continuously fed with the respective load responses for the different motion axes of the packaging or filling machine 300, and at defined points in time of the machine's operation, a virtual representation of said machine is constructed.

[0019] The method 1000 comprises a step 1060 of generating a second distribution (T2, p2) of the associated registered values ​​1070 of said forces (T) and / or motion parameters (p) measured subsequently at a subsequent time point when moving the machine component according to a cycle defined by the motion profile (PF). The subsequent time point may be several hours after the operation of the packaging or filling machine 300, for example several weeks, months or years after the operation on the production line. Figure 3b is an example of a second distribution (T2, p2) obtained at such a subsequent time point. The method 1000 comprises a step 1080 of relating the second distribution (T2, p2) to a second load response (L2) in the mechanical load model (VM) of the current motion axis of the packaging or filling machine 300.

[0020] The method 1000 comprises a step 1090 of determining the amount of change in the load component (C) of the mechanical load at a subsequent time. Determining the change in the load component (C) comprises a step 1100 of determining the amount of variation of the load components, such as the load components C1, C2, C3 of the mechanical load, based on the difference between the first load response (L1) and the second load response (L2), for condition monitoring. Determining the first and second load responses (L1, L2) of the moving mechanical part and determining the amount of variation of the contributing load component (C) of the mechanical load on the mechanical part based on the difference between the first load response (L1) and the second load response (L2) allows an accurate and reliable classification of the condition of the mechanical part. Thus, the variation between the first and second load responses (L1, L2) is utilized to determine the change in the load component (C). For example, it is determined whether the load component causing the variation is related to the type of external force of the mechanical part, such as friction or impulse on the mechanical part, and / or the change in the inertia exhibited by the mechanical part. The selected maintenance actions may be associated with the combination of load components (C) identified as causing the variations in the load response (L1, L2). In this way, easy condition monitoring of moving machine parts is provided for reliable and timely detection of deviating behavior or impending failures with minimal data analysis required by the operator.

[0021] Conventional solutions that generally focus on detecting whether a certain measurable characteristic is within a defined threshold level for condition monitoring are limited in terms of resolving changing trends at the early stages of deviations. For example, a servo motor may experience position delays when its peak torque is approached or exceeded. Even small deviations when such delays occur may be detrimental in a high throughput production line and may already compromise sterility performance, as described above. In other situations, a servo motor may require cooling for operation close to or beyond its effective torque limit, resulting in downtime of the production line. The method 1000 provides for capturing occurrences of detrimental deviations coupled to motion axes in a packaging or filling machine 300, and in particular, determining which load component (C) is responsible for the deviation based on the detected difference between the first and second load responses (L1, L2), as described above. In this way, specific maintenance work related to the identified load component (C) can be easily performed by an operator.

[0022] To determine the relationship between the first load response (L1) and the second load response (L2), various statistical measures may be used, such as determining the mean or variance, or the relationship between the trends of the distributions (T1, p1, T2, p2). A statistically significant relationship is identified when the deviations resulting from such a comparison are within defined statistical limits or criteria. In this way, by comparing a set of such distributions from load responses (L1, L2) obtained at different times, the variation of the machine part over time may be identified.

[0023] Determining the amount of variation of the load component (C) may comprise a step 1101 of determining a change in derivative between the first distribution (T1, p1) and the second distribution (T2, p2) of the respective first and second load responses (L1, L2). Referring to the example of FIG. 4a, the load response (L1) associated with the defined motion profile (PF) is divided into three time periods (t1, t2, t3), corresponding to periods of acceleration, constant speed and deceleration, respectively. The change in the derivative of the torque (T2) in the second load response (L2) during the acceleration period (t1) is determined compared to the corresponding period in the first load response (L1). The derivative increases during t1 for the torque T2 compared to the torque T1 in the first load response (L1). Similarly, the derivative of T2 increases during the period of deceleration t3 compared to T1. In this way, the variation of the derivative of the distribution in the first and second load responses (L1, L2) may be utilized to identify the amount of variation of the load component (C) that contributes to the total mechanical load on the current axis of motion. The example in Figures 4a-b shows the load component (C) depending on the acceleration.

[0024] The method 1000 may comprise a step 1102 of determining the change in derivative as the change in viscous friction contributing as an external mechanical force to the mechanical load, and a step 1103 of relating the load component (C) to the viscous friction. Thus, the viscous friction can be considered to be proportional to the speed and acting "against" the direction of movement. Viscous friction is particularly large in worm gears and greased sliding surfaces. When the grease dries, the viscous friction increases. The increase in the derivative of T2 between t1 and t3 of the second load response L2 compared to T1 of the first load response L1 is therefore associated with an increase in viscous friction in the example of Fig. 4a-b.

[0025] The method 1000 may comprise a step 1103' of determining the load component (C) as the first load component C1 in the mechanical load model (VM). Thus, the viscous friction may be associated with the first load component C1 in the mechanical load model (VM). Thus, the load responses (L1, L2) stored in the mechanical load model (VM) for different motion axes of the packaging or filling machine 300 may be compared to determine the change in the derivatives (T1, p1, T2, p2) of the respective distributions during periods of acceleration and deceleration in the motion defined by the motion profile (PF). Thus, the associated load component, for example the first load component C1, may be determined from the load responses (L1, L2) of the mechanical load model (VM) as the change in viscous friction for the current motion axis.

[0026] As will be further described below with reference to FIG. 1, after determining the effect of the first load component C1, e.g., viscous friction, on the load response (L1, L2), a simulated load response, hereinafter referred to as a virtual load response output (VL), can be generated using a mechanical load model (VM) with the first load component C1 as a variable parameter (VC1) and a defined motion profile (PF) as an input. The simulated load response may be compared to a measured load response (ML) having a distribution of values ​​measured at a subsequent time point for the current motion axis (e.g., T2 in FIG. 4b), also with the defined motion profile (PF) as an input. The difference between the measured load response (ML) and the simulated load response (VL) can be minimized by adjusting the first load component C1 as a variable parameter to characterize the effect of the first load component C1, e.g., viscous friction, on the mechanical load at the subsequent time point. A maintenance routine can be identified that depends on the amount of adjustment of VC1 in the mechanical load model (VM) when minimizing the difference between the measured load response (ML) and the simulated load response (VL). The adjustment of VC1 may be determined relative to a predetermined reference within a chance load model (VM) that represents a desired operating state of the machine components.

[0027] Returning to the example of Figures 4a-b, it should be appreciated that the difference in torques T1, T2, between different load responses (L1, L2) is the first load component C1, and a similar comparison can be made between position errors p1, p2, to determine the variation of the first load component C1. For example, the derivative of p2 in Figure 4b increases slightly between t1 and t3 compared to p1 in Figure 4a.

[0028] The step of determining the variation of the load component (C) may comprise a step 1104 of determining the variation of the maximum values ​​(max1, max2) and / or minimum values ​​(min1, min2) in the first and second distributions (T1, p1, T2, p2) of the respective first and second load responses (L1, L2). That is, the first distribution (T1, p1) is compared with the second distribution (T2, p2) and said variation is determined. Figure 3a-b shows an example where the maximum value (max2) of the torque curve (T2) in the second load response (L2) is increased with respect to the maximum value (max1) of the torque curve (T1) in the first load response (L1). Furthermore, the minimum value (min2) of T2 is changed with respect to the minimum value (min1) of T1 in the first load response (L1). That is, the absolute value of min2 is decreased. Therefore, the variation in the maximum and / or minimum values ​​of the distribution in the first and second load responses (L1, L2) may be utilized to identify the amount of variation in the load component (C) that contributes to the total load on the machine component for the current axis of motion.

[0029] The example of Figures 3a-b shows that when the speed of the defined motion profile (PF) is greater than zero, the torque curve T2 is generally shifted in the positive direction compared to T1. This change is accompanied by an offset (o y ) in the positive direction. A similar shift and offset of T2 (o y) can be realized for min1, min2 as well as for the intermediate section where the speed is constant. The method 1000 calculates the offset (o) between the first and second load responses (L1, L2) for their respective maximum (max1, max2) and minimum (min1, min2) values. y ) as the change in static friction that contributes as an external mechanical force to the mechanical load. The method 1000 may further comprise a step 1106 of relating the load component (C) to the static friction. The static friction can be considered to be constant in value and acting "against" the direction of movement. Thus, in the example of Figures 3a-b, the offset of T2 (o y ) is associated with increased static friction.

[0030] The method 1000 may comprise a step 1106' of determining the load component (C) as a second load component C2 in the mechanical load model (VM). Thus, the static friction may be associated with the second load component C2 in the mechanical load model (VM). The load responses (L1, L2) stored in the mechanical load model (VM) for different motion axes of the packaging or filling machine are determined by determining the offset (o) between the respective distributions (T1, p1, T2, p2) during the period when the speed is greater than zero in the motion defined by the motion profile (PF). y ) are compared to identify the relevant load component, i.e., the second load component C2, may then be determined from the load response (L1, L2) of the Mechanical Load Model (VM) as the change in static friction for the current motion axis.

[0031] The example of Fig. 5a-b shows the corresponding range (d y1 ) compared to the range (d y2 ) is shown. The method 1000 calculates the change in inertia and / or moment of inertia contributing to the mechanical load by calculating the range (d y1, d y2 ) as described above. As discussed above, the moment of inertia is related to the rotating mass. Method 1000 may further comprise step 1108 of relating the load component (C) to the inertia and / or moment of inertia. An increase in the moment of inertia requires more torque to accelerate the associated mass to a set speed and more torque to stop the mass from moving. Thus, the range of increase d of the second load response L2 may be determined by the step 1106. y2 is associated with an increase in the moment of inertia. An increase in mass can be seen, for example, when replacement parts made of a different material are used, such as steel instead of aluminum.

[0032] The method 1000 may comprise a step 1108' of determining a third load component C3 in the mechanical load model (VM) as the load component (C). Thus, an inertia, or moment of inertia, may be associated with the third load component C3 in the mechanical load model (VM). Thus, the load responses (L1, L2) stored in the mechanical load model (VM) for the different motion axes of the packaging or filling machine 300 are determined as ranges (d1, p1, T2, p2) between the respective distributions (T1, p1, T2, p2) during the period when the speed is greater than zero in the motion defined by the motion profile (PF). y1 , d y2 ) may be compared to identify variations in the load response (L1, L2) of the mechanical load model (VM). Thus, a relevant load component, e.g., the third load component C3, may be determined from the load response (L1, L2) of the mechanical load model (VM) as a change in inertia or moment of inertia of the current motion axis.

[0033] Figure 6a-b shows two external load transients (F t ) is shown in the second load response L2. For example, the moving machine part may undergo two collision events with an obstacle which causes an instantaneous increase in the external load. Figures 7a-b show another example where the effect on the load response (T2) of the sum of the above mentioned load components C1, C2, C3 is shown.

[0034] The method 1000 may comprise a step 1109 of determining the amount of variation of the first, second and third load components (C1, C2, C3) in the mechanical load model (VM) based on the aforementioned difference between the first load response (L1) and the second load response (L2). The first load response (L1) may be determined for a known reference state for each relevant motion axis of the packaging or filling machine 300, which may represent a desired, e.g. healthy, operating state. The defined motion profile (PF) may be adjusted to a specific machine component for each motion axis. The second load response (L2) is determined at a subsequent time point for each of the defined motion profiles (PF) respectively associated with the motion axis. The variation of the first, second and third load components (C1, C2, C3) may be determined from the first and second load responses (L1, L2) as described above. Then, depending on the variation of the load components (C1, C2, C3), a maintenance action may be triggered.

[0035] The method 1000 may comprise a step 1110 of assigning the load components (C1, C2, C3) as respective variable load parameters (VC1, VC2, VC3) in a mechanical load model (VM) to generate a virtual load response output (VL) depending on the variable load parameters (VC1, VC2, VC3). For example, the method 1000 may comprise a step 1110 of assigning the first, second and third load components (C1, C2, C3) as respective variable load parameters (VC1, VC2, VC3) in the mechanical load model (VM) to generate a virtual load response output (VL) depending on the variable load parameters (VC1, VC2, VC3). The mechanical load model (VM) represents the mechanical load of a particular moving machine component, and the components (C1, C2, C3) of the mechanical load may be controlled by the variable load parameters (VC1, VC2, VC3). The defined motion profile (PF) may be input to the mechanical load model (VM) as shown diagrammatically in FIG. 1. The defined motion profile (PF) defines a set input value, e.g., a set speed (v1). A force (T), e.g., torque (T1, T2), required to drive the mechanical load in the mechanical load model according to the set input value is calculated and is referred to here as the virtual load response output (VL). The mechanical load in the mechanical load model is represented by load parameters (VC1, VC2, VC3) in the above example. The mechanical load model (VM) may communicate with a control algorithm, such as a PID regulator, which takes as input the set value of the defined motion profile (PF), e.g., the set speed (v1), together with the actual speed (v2) and actual position of the virtual load response output (VL). The PID outputs the torque (T1, T2) and the position error (p1, p2), and the determined torque (T1, T2) is fed back as an input to the mechanical load model (VM).

[0036] The virtual load response output (VL) may be controlled by the load parameters (VC1, VC2, VC3). Thus, the virtual load response output (VL) may be adjusted to model a particular motion axis in the packaging or filling machine 300 by controlling the load parameters (VC1, VC2, VC3). That is, the load parameters (VC1, VC2, VC3) may be optimized to generate a virtual load response output (VL) optimized for a measured load response (ML) in the packaging or filling machine 300 having the same defined motion profile (PF) as input, as shown diagrammatically in FIG. 1, by executing the defined motion profile (PF) as input to the mechanical load model (VM). The optimization may consist of minimizing the difference between the virtual load response output (VL) and the measured load response (ML) of the mechanical components of the current motion axis, such as a measured torque curve.

[0037] A mechanical load model (VM) may thus be established as an image of the packaging or filling machine 300 at a defined point in time, with the mechanical load on each axis of motion having a representation associated with it by the load parameters (VC1, VC2, VC3) determined as described above in relation to Figures 1 to 7. That is to say, the different load components (C1, C2, C3) in the mechanical load model (VM) and the associated load parameters (VC1, VC2, VC3) may be determined from the characteristics of the load response (L1, L2).

[0038] Considering the determined influence of the load parameters (VC1, VC2, VC3) on the virtual load response output (VL), the load parameters (VC1, VC2, VC3) may be varied to optimize the virtual load response output (VL) for a measured load response (ML) at a later time, e.g., after several weeks, months or years of operation of the moving machine components. Depending on how much the different load parameters (VC1, VC2, VC3) need to be varied to minimize the difference (e.g., T2 in FIG. 3b) between the generated virtual load response output (VL) and the subsequently measured load response (ML), a maintenance routine may be determined.

[0039] Thus, the method 1000 may comprise a step 1200 of determining a maintenance action for the machine component after a duration of operation. The method 1000 may comprise a step 1210 of generating a distribution (e.g., T2, p2) of associated registration values ​​(T, p) measured subsequently when the machine component is moved according to a defined motion profile (PF) after said duration. The method 1000 may comprise a step 1220 of relating the distribution to a measured load response (ML). The method 1000 may comprise a step 1230 of determining a change in the respective load parameters (VC1, VC2, VC3) in the machine load model (VM) that results in a minimized difference between the measured load response (ML) and the virtual load response output (VL). The method 1000 may comprise a step 1240 of determining a maintenance action based on the change in the respective load parameters (VC1, VC2, VC3).

[0040] In one example, with reference to the discussion of Figures 4a-b, the method 1000 comprises a step 1231 of assigning a first load component (C1) as a first variable load parameter (VC1) in a mechanical load model (VM), and a step 1231' of modifying the first variable load parameter (VC1) associated with the first load component (C1) to minimize a difference in derivatives between the virtual load response output (VL) and the measured load response (ML), and a maintenance action for controlling viscous friction is determined.

[0041] In a further example, with reference to the discussion of FIGS. 3a-b, the method 1000 may comprise a step 1232 of assigning a second load component (C2) as a second variable load parameter (VC2) in the mechanical load model (VM) and a step 1232′ of modifying the second variable load parameter (VC2) associated with the second load component (C2) to account for the difference between the virtual load response output (VL) and the measured load response (ML) by an offset (o) between the respective maximum and minimum values ​​(min1, max1), (min2, max2). y ) to determine the maintenance effort required to control static friction.

[0042] In a further example, and in relation to the discussion of Figures 5a-b, the method 1000 may comprise a step 1233 of assigning a third load component (C3) as a third variable load parameter (VC3) of the mechanical load model (VM) and a step 1233' of modifying the third variable load parameter (VC3) associated with the third load component (C3) to determine a difference between the virtual load response output (VL) and the measured load response (ML) within a range (dL) between respective maximum and minimum values ​​(min1, max1), (min2, max2). y1 , d y2 ) with respect to the magnitude of the inertia and / or moment of inertia.

[0043] The load components (C1, C2, C3) identified as contributing to the change in the load response, e.g., the change in T2 over time, can thus be the basis for selecting a maintenance routine. The virtual load response output (VL) can be automatically optimized for the measured load response (ML) at any point in time by iteratively adjusting the respective load parameters (VC1, VC2, VC3) such that the difference between the resulting virtual load response output (VL) and the measured load response (ML) is minimized. Thus, the correct maintenance routine can be automatically determined by the method 1000 based on the determined adjustments of the load parameters (VC1, VC2, VC3). The proposed maintenance routine can be directly communicated to the operator or technician.

[0044] It is conceivable that for each motion axis of the packaging or filling machine 300, multiple moving machine components are represented in the mechanical load model (VM). Thus, the measured load response (ML) may comprise multiple respective load responses for the motion axis at a particular time. The method 1000 and the associated system 200 provide for continuous and / or automatic evaluation of multiple respective load responses against a corresponding virtual load response output (VL) from the mechanical load model (VM) that may be established as a previous reference state of the motion axis. The evaluation may be performed as a batch process across multiple motion axes, obtaining the respective load responses sequentially or in parallel. Based on the determined adjustments of the load parameters (VC1, VC2, VC3) as described above, deviations may be detected early and the operator may be directly notified of proposed maintenance actions. Multiple moving machine components corresponding to each motion axis may be grouped into an assembly in the mechanical load model (VM) with multiple load parameter sets (VC1, VC2, VC3). Based on how the load parameters (VC1, VC2, VC3) are adjusted as a whole, a maintenance action may be determined, for example to determine a maintenance service for an entire collection of machine components.

[0045] In this manner, method 1000 provides reliable and timely detection of deviant operation or impending failure to give operators an effective tool to identify and plan maintenance activities for related components without impacting production.

[0046] There is also provided a system 200 for condition monitoring of moving machine parts of a packaging or filling machine 300 in liquid food processing applications. The system 200 comprises a processing unit 201, shown diagrammatically in combination with 8d in FIG.

[0047] The processing unit 201 is configured to move the mechanical part 1010 according to a cycle of a defined motion profile (MP) comprising a step 1020 of accelerating the mechanical part to overcome a mechanical load, the mechanical load comprising a sum of contributing load components (C) consisting of external mechanical forces acting on the mechanical part and the inertia and / or moment of inertia exhibited by the mechanical part. The processing unit 201 performs a step 1030 of registering values ​​of measured forces (T) causing said acceleration and / or measured motion parameters (p) related to the motion of the mechanical part according to the defined motion profile, a step 1040 of generating a first distribution (T1, p1) of the registered values, a step 1050 of associating the first distribution with a first load response (L1) in a mechanical load model (VM), and a step 1051 of associating the first distribution with a first load response (L1) in a mechanical load model (VM) with the forces (T) and / or motion parameters subsequently measured when moving the mechanical part according to the cycle at a subsequent time point. The method includes a step 1060 of generating a second distribution (T2, p2) of associated registered values ​​1070 of a parameter (p), a step 1080 of associating the second distribution with a second load response (L2) in the machine load model, and a step 1090 of determining a change in a load component (C) of the machine load at the subsequent time, the step 1100 comprising determining an amount of variation in the load component (C) of the machine load based on a difference between the first load response (L1) and the second load response (L2) for condition monitoring.

[0048] In this manner, system 200 provides the advantageous benefits discussed above in connection with Figures 1-7. System 200 provides accurate and reliable classification of the condition of machine components. In this manner, condition monitoring of moving machine components is facilitated, providing reliable and timely detection of deviant operation or impending failure with minimal data analysis required by an operator.

[0049] The processing unit 201 may be configured to assign the load components (C1, C2, C3) as respective variable load parameters (VC1, VC2, VC3) in a machine load model (VM) and generate a virtual load response output (VL) depending on the variable load parameters (VC1, VC2, VC3). The processing unit 201 may comprise a step 1200 for determining a maintenance action for the machine component after a duration including a step 1210 for generating a distribution (e.g., T2, p2) of associated registration values ​​(T, p) measured subsequently when moving the machine component according to a cycle, i.e., a defined motion profile (PF), after said duration. The processing unit 201 may comprise a step 1220 for relating the distribution to a measured load response (ML) and a step 1230 for determining a change in the respective variable load parameters (VC1, VC2, VC3) in the machine load model (VM) that results in a minimized difference between the measured load response (ML) and the virtual load response output (VL). The processing unit 201 may be configured to determine a maintenance action 1240 based on the changes in the respective variable load parameters (VC1, VC2, VC3).

[0050] A computer program product is provided comprising instructions that, when the program is executed by a computer, cause the computer to perform the steps of the method 1000 described above in relation to Figures 1-7.

[0051] From the foregoing description, while various embodiments of the present invention have been described and illustrated, the invention is not limited thereto and may be embodied in other ways within the scope of the subject matter defined in the following claims.

Claims

1. A method (1000) for condition monitoring of moving mechanical parts in a packaging or filling machine (300) for liquid food processing applications, comprising: A step (1010) of moving a mechanical part according to a cycle of a defined motion profile (PF); accelerating (1020) the mechanical part to overcome a mechanical load according to the velocity (v) and position of the defined motion profile, the mechanical load being determined by contributing load components (C １ , C ２ , C ３ ) including the sum of the load components include an external mechanical force acting on the mechanical component and an inertia and / or a moment of inertia exhibited by the mechanical component; The method further comprises: - registering (1030) measurements of the force (T) causing said acceleration and / or measurements of a motion parameter (p) related to the motion of said machine part according to said defined motion profile; The first distribution of registered values ​​(T １ , p １ ) (1040); In a mechanical load model (VM), the first distribution is converted into a first load response (L １ ) (1050); At a subsequent time point, a second distribution (T) of the forces and / or associated registered values ​​(1070) of the motion parameters measured when moving the machine part according to the cycle is calculated. ２ , p ２ ) (1060); In the mechanical load model, the second distribution is referred to as a second load response (L ２ ) (1080); determining (1090) a change in a load component of the mechanical load at said subsequent time point; determining (1100) a variation amount of a load component of the mechanical load based on a difference between the first load response and the second load response for the condition monitoring; The load components are then applied to respective variable load parameters (VC) of a mechanical load model to generate a virtual load response output (VL) depending on the variable load parameters. １ , V.C. ２ , V.C. ３ ) as the step (1110); determining (1200) a maintenance operation for said machine part after a duration of operation; generating (1210) a distribution of associated registration values ​​subsequently measured as the machine part moves according to the cycle after said duration; Correlating (1220) the measured load response (ML) with the distribution; Each variable load parameter (VC) in the mechanical load model that results in a minimized difference between the measured load response (ML) and the virtual load response output (VL) １ , V.C. ２ , V.C. ３ ) (1230), and Each variable load parameter (VC １ , V.C. ２ , V.C. ３ determining (1240) said maintenance action based on a change in A method for providing the above.

2. The step of determining the amount of fluctuation of the load component includes: determining (1101) a change in derivative between first and second distributions of respective first and second load responses; The method of claim 1.

3. determining (1102) a change in said derivative as a change in viscous friction contributing as an external mechanical force to said mechanical load; and relating the load component to the viscous friction (1103). The method of claim 2.

4. In the mechanical load model, the first load component (C １ ) determining the load components (1103′) as The method of claim 3.

5. The step of determining the amount of fluctuation of the load component includes: The maximum values ​​(max １ , max ２ ) and / or minimum value (min １ , min ２ ) determining a change in (1104) 10. The method of claim 1 .

6. The offset (o) between the first load response and the second load response with respect to their respective maximum and minimum values ​​as a change in static friction contributed as an external mechanical force to the mechanical load. ｙ ) (1105); and correlating the load component with static friction (1106). The method of claim 5.

7. The second load component of the mechanical load model (C ２ ) determining the load components (1106′) as The method of claim 6.

8. The change in inertia and / or moment of inertia contributing to the mechanical load is calculated by calculating the difference in the magnitude of the range between the minimum and maximum values ​​of the first and second load responses (d ｙ１ , d ｙ２ ) (1107); and associating the load components with inertia and / or moments of inertia (1108). The method of claim 5.

9. The third load component of the mechanical load model (C ３ determining the load components (1108′) as The method of claim 8.

10. determining (1109) variations of the first, second and third load components in the mechanical load model based on the difference between the first load response and the second load response. The method of claim 4.

11. and generating the virtual load response output (VL) in response to the variable load parameters, the respective variable load parameters (V １ , V.C. ２ , V.C. ３ ) assigning the first, second and third load components as The method of claim 10.

12. The first variable load parameter (VC) of the mechanical load model １ assigning the first load component as (1231); The first variable load parameter (VC) associated with the first load component is adjusted to minimize a difference between a derivative of a virtual load response output and a measured load response such that a maintenance action for controlling viscous friction is determined. １ ) changing (1231′), The method of claim 1.

13. A second variable load parameter (VC) of the mechanical load model ２ assigning (1232) a second load component as The second variable load parameter (VC) associated with the second load component is adjusted to minimize a difference between a virtual load response output and a measured load response with respect to an offset between their respective maximum and minimum values ​​such that a maintenance action for controlling stiction is determined. ２ ) modifying (1232') The method of claim 1.

14. The third variable load parameter (VC) of the mechanical load model ３ assigning a third load component as (1233); The third variable load parameter (VC) associated with the third load component is adjusted to minimize a difference between a virtual load response output and a measured load response with respect to the magnitude of a range between the respective maximum and minimum values, so that maintenance actions for controlling inertia and / or moment of inertia are determined. ３ ) modifying (1233'), The method of claim 1.

15. A computer program comprising instructions that cause a computer to carry out the steps of the method of claim 1 when the program is executed by a computer.

16. A system (200) for condition monitoring of moving mechanical parts of a packaging or filling machine (300) in a liquid food processing application, comprising: a processing unit (201); The processing unit A step (1010) of moving a mechanical part according to a cycle of a defined motion profile (PF); accelerating (1020) the mechanical part to overcome a mechanical load according to the velocity (v) and position of the defined motion profile, the mechanical load being determined by contributing load components (C １ , C ２ , C ３ ) including the sum of the load components include an external mechanical force acting on the mechanical component and an inertia and / or a moment of inertia exhibited by the mechanical component; The method further comprises: - registering (1030) measurements of the force (T) causing said acceleration and / or measurements of a motion parameter (p) related to the motion of said machine part according to said defined motion profile; The first distribution of registered values ​​(T １ , p １ ) (1040); In a mechanical load model (VM), the first distribution is converted into a first load response (L １ ) (1050); At a subsequent time point, a second distribution (T) of the forces and / or associated registered values ​​(1070) of the motion parameters measured when moving the machine part according to the cycle is calculated. ２ , p ２ ) (1060); In the mechanical load model, the second distribution is referred to as a second load response (L ２ ) (1080); determining (1090) a change in a load component of the mechanical load at said subsequent time point; determining (1100) a variation amount of a load component of the mechanical load based on a difference between the first load response and the second load response for the condition monitoring; The load components are then applied to respective variable load parameters (VC) of a mechanical load model to generate a virtual load response output (VL) depending on the variable load parameters. １ , V.C. ２ , V.C. ３ ) as the step (1110); determining (1200) a maintenance operation for said machine part after a duration of operation; generating (1210) a distribution of associated registration values ​​subsequently measured as the machine part moves according to the cycle after said duration; Correlating (1220) the measured load response (ML) with the distribution; Each variable load parameter (VC) in the mechanical load model that results in a minimized difference between the measured load response (ML) and the virtual load response output (VL) １ , V.C. ２ , V.C. ３ ) (1230), and Each variable load parameter (VC １ , V.C. ２ , V.C. ３ determining (1240) said maintenance action based on a change in Equipped with system.

Images