Method and device for monitoring the parallelism of two corrugating rollers which can be adjusted relative to each other

Abstract

The invention relates to a method for monitoring the parallelism of two corrugating rollers which can be adjusted relative to each other in a device for producing corrugated cardboard, having the following steps: – determining, by means of a sensor, at least one vibration frequency spectrum of two rotating corrugating rollers, which are designed to produce a corrugated cardboard laminated on at least one side, – determining at least one sideband level factor on the basis of the vibration frequency spectrum, – determining a need for an adjustment of the alignment of the corrugating rollers with respect to each other on the basis of the sideband level factor, and – outputting the need for an adjustment. The proposed solution also relates to a device for monitoring the parallelism of two corrugating rollers, to a machine for producing corrugated cardboard laminated on at least one side, and to a computer program product.

Description

[0001] Method and device for monitoring the parallelism of two corrugated rollers adjustable relative to each other

[0002] Description

[0003] The proposed solution relates to a method for monitoring the parallelism of two corrugating rollers adjustable relative to each other for corrugated board production, as well as a device with corresponding monitoring.

[0004] Corrugated rollers are used in corrugated board production to emboss the corrugated surface onto the board. These rollers have a corrugated surface for this purpose. Parallelism of the corrugations is crucial for a clean and precise formation of the corrugated surface.

[0005] To determine the parallelism of the profile ends of the corrugating rollers, carbon paper strips are often inserted into the rotating roller. The carbon paper strips are then placed side by side, and the symmetry of the two impressions is visually assessed. Such a check can only be performed when production is at a standstill. Furthermore, two people are required to feed the carbon paper strips cleanly. Therefore, this method of determining parallelism is time-consuming, interrupts production, and the assessment of parallelism is ultimately subjective and dependent on the individuals involved.

[0006] Therefore, the task is to improve the monitoring of parallel processes.

[0007] This problem is solved by a method according to claim 1 and a device according to claim 14.

[0008] The proposed method for monitoring the parallelism of two corrugating rollers adjustable relative to each other for corrugated board production therefore comprises at least:

[0009] - a sensory determination of at least one vibration frequency spectrum of the two rotating corrugated rollers,

[0010] - determining at least one sideband level factor based on the vibration frequency spectrum, determining the adjustment requirement for the alignment of the corrugated rollers relative to each other as a function of the sideband level factor, and outputting the adjustment requirement.

[0011] The sideband level factor (SBF), derived from the vibration frequency spectrum, allows for the assignment of a numerical value to the parallelism of the axes of both corrugating rollers. This makes parallelism quantifiable. Furthermore, deviations from parallelism can be detected using the SBF. Thus, the SBF enables an objective determination of adjustment requirements. The operator's experience is not a factor. In particular, the adjustment requirement can be determined and output by computer, allowing the machine operator to adjust the alignment of the corrugating rollers to optimize parallelism. Personnel costs and production downtime can therefore be reduced by the proposed method.

[0012] The SBF describes a measure of how pronounced at least one secondary band is above and / or below a tooth engagement frequency and / or a harmonic of the tooth engagement frequency.

[0013] For example, the SBF can be a ratio of the amplitude of at least one secondary band to the tooth engagement frequency, or to the harmonic of the tooth engagement frequency. The amplitude of the at least one secondary band can be a vibration amplitude at a tooth engagement frequency + / - the rotational frequency of the corrugating rollers, or at a harmonic of the tooth engagement frequency + / - the rotational frequency of the corrugating rollers.

[0014] As a further example, the SBF (Sustainable Frequency Band) can be the ratio of the area of ​​at least one secondary band above and / or below a tooth engagement frequency and / or a harmonic of the tooth engagement frequency to the width of the at least one secondary band. The secondary band here describes a frequency range that lies above or below the tooth engagement frequency of the corrugating rollers or a harmonic of the tooth engagement frequency. Depending on its position above or below the tooth engagement frequency or a harmonic of the tooth engagement frequency, the vibration frequency spectrum can be divided into upper and lower secondary bands. The tooth engagement frequency and the frequencies of the harmonics are not included in this calculation of the SBF. For example, the area of ​​the i-th upper secondary band A can be calculated as... u l p Calculate as the integral of the frequency-dependent amplitude a(f):

[0015] Here, f is the oscillation frequency, a(f) the oscillation amplitude in the frequency domain, F t is the frequency of the i-th harmonic oscillation of the tooth engagement frequency (i=0 tooth engagement frequency, i=1 first harmonic, etc.), E is a frequency difference of the secondary band to the frequency of the i-th harmonic oscillation, and f end is a cutoff frequency of the secondary band.

[0016] E intuitively describes the distance of the frequency that is closest in the frequency domain to the frequency of the i-th harmonic oscillation and that is still taken into account in the calculation. The quantity f end In contrast, it describes the most distant frequency still taken into account.

[0017] The value of E should be chosen with respect to a specific vibration frequency spectrum such that the signal of the i-th harmonic oscillation does not influence the calculation of the sideband area, but the sideband signal components are not unduly neglected. The value of f end The value of f should not be chosen too large in order to avoid disproportionately increasing the computation time. end However, it should not be chosen too small in order to correctly determine at least a large part of the area of ​​the secondary band. The secondary bands that occur are more pronounced in the spectral vicinity of the harmonic vibrations than in spectrally separated regions. Therefore, a calculation of the secondary band can also be possible if f end is chosen so that integration does not occur over the entire area of ​​at least one secondary band.

[0018] For example, f end be chosen so that the at F t + fend The existing oscillation amplitude just exceeds a predetermined value. That is, for all frequencies f in the interval [FI + E; F] t + f end The corresponding amplitude is greater than the predetermined threshold. This prevents a noise signal from being included in the integration without clipping relevant parts of the secondary band. Similarly, the area A l down The i-th lower subband is calculated as follows: down fi d a Fi - df.

[0019] This results in an area for the entire i-th supplementary volume with:

[0020] Sizes E and f end In principle, they can be different for each i-th harmonic oscillation.

[0021] To calculate the SBF, the area of ​​the i-th secondary band A^ is set in the ratio of the width of the secondary band in the frequency domain.

[0022] An exemplary measure of the width of a distribution is the standard deviation a. l . Thus, the i-th SBF can be calculated as follows, for example:

[0023] The calculation methods mentioned are exemplary equations in the case of a continuous frequency domain. It is equally conceivable and possible to consider and determine the SBF in a discrete frequency domain.

[0024] According to an exemplary embodiment of the proposed method, a sum of the first SBF values ​​can be calculated for i = 0, 1, 2 and used to determine the adjustment requirement. This can increase sensitivity to a non-parallel alignment of the corrugated rollers.

[0025] According to a further embodiment of the proposed method, the sum of the SBF (seamless surface area) of the first three secondary strips with i = 0, 1, 2 is determined. These secondary strips can be particularly sensitive to the parallelism of the corrugated rollers. In other words, the effect of pronounced secondary strips can be especially noticeable here. Therefore, calculating the sum of the SBF of the first three secondary strips saves computation time. In a further embodiment, the sum of the SBF of the first three upper secondary strips with i = 0, 1, 2 is also determined. This can save even more computation time.

[0026] According to a further supplementary or alternative embodiment of the proposed method, a vibration frequency spectrum is determined for a plurality of sections of one of the corrugating rollers that are subject to discrepancies. This makes it possible to determine the vibration frequency spectrum (VFS) for different locations on the corrugating roller. By comparing the VFS at different locations on the same corrugating roller, the vibration behavior of the roller at these different locations can be evaluated. Deviations in the VFS can indicate spatially inhomogeneous vibration behavior of the corrugating roller. Inhomogeneous vibration behavior can be associated with poor parallelism. Thus, based on a comparison of the VFS of different sections, it is possible to identify poor parallelism and therefore a need for alignment adjustment. This reduces the time required to determine the adjustment needs.

[0027] Alternatively or additionally, the need for adjustment can be detected if at least one SBF (Standard Load Factor) exceeds a predetermined threshold value. This can reduce the technical effort required to implement the proposed procedure.

[0028] With a further refinement of the proposed method, the adjustment requirement can be determined using logic to predict at least one SBF (Standard Field Boundary) depending on the orientation. This logic can therefore predict whether the SBF can be reduced by adjusting the orientation, and if so, in which direction and by how much the orientation needs to be adjusted. This can further improve the quality of parallelism monitoring.

[0029] According to a further supplementary or alternative embodiment of the proposed method, the logic can include determining the at least one SBF (Standardized Flow Rate) for at least two different orientations of the corrugating rollers relative to each other. The logic can thus provide that the relative orientation of the corrugating rollers is varied to determine the adjustment requirement. Based on the SBFs determined during the variation, the logic can determine the adjustment requirement. This can enable a targeted check to what extent the parallelism can be further improved from an initial orientation of the corrugating rollers, or whether an optimal parallel orientation already exists. According to a further supplementary or alternative embodiment of the proposed method, the logic can include determining a trend function that approximately describes the development of the at least one SBF as a function of the orientation.The actual development of at least one SBF (swimming angle) as a function of the orientation does not necessarily need to be known and / or recorded in order to determine the adjustment requirement. Rather, it is sufficient if the trend function near a current orientation describes the behavior of the SBF in such a way that it can be determined how the orientation must be changed to reduce the SBF. This can enable an efficient determination of the adjustment requirement.

[0030] In particular, the trend function reveals how the at least one SBF changes when one of the corrugated rollers is adjusted in a specific spatial direction. The gradient of the trend function indicates in which direction one of the corrugated rollers must be adjusted to reduce the SBF.

[0031] It is conceivable and possible, in principle, that the trend function describes the development of the SBF only for a limited range of orientations. For example, an arbitrarily complex functional relationship between the SBF and the orientation can be described locally by a first-order polynomial (a straight line). Likewise, a functional relationship can be described by a higher-order polynomial. In this case, the range in which the functional relationship can be approximately described by a polynomial of a specific order may depend on the order of the polynomial.

[0032] The functional relationship can also be described, for example, by a linear combination of polynomials of different orders (Taylor series). It may be true that the computational effort required to determine the trend function increases with increasing degree of the Taylor series. However, the range in which the trend function approximately describes the functional relationship between the at least one SBF and the orientation may also grow with higher order.

[0033] In one embodiment of the proposed method, the tendency function can be a regression line between at least three SBF measurements taken at different orientations. This can reduce the effort required to determine the adjustment requirement and conserve computing resources. The adjustment requirement can include information about the direction and magnitude by which the orientation of the corrugated rollers can be adjusted to optimize parallelism. In particular, the adjustment requirement can thus include a target orientation value to which the corrugated rollers should be adjusted to optimize parallelism.

[0034] The target value can be either a relative value, such as an angle between the corrugating rollers, or an absolute value, such as a position relative to a coordinate system decoupled from the corrugating rollers. With regard to a relative target value, knowledge of the absolute position of the corrugating rollers may be irrelevant. This can simplify the implementation.

[0035] For example, a direction can be derived from the sign of the gradient of the tendency function at the point of the current alignment, or initial alignment. The magnitude of this gradient can be proportional to its magnitude. By repeatedly determining the adjustment requirement, the alignment of the corrugated rollers can thus converge to an optimum. In particular, the alignment can converge to perfect parallelism.

[0036] In principle, it is also conceivable and possible that the adjustment requirement is implemented in the form of a feedforward control.

[0037] A control system generally refers to a technical process in which a specific process is continuously monitored and adjusted to achieve and maintain a desired target value. A control system therefore differs from a control system in that, within a control system, the process is not necessarily monitored and adjusted.

[0038] A feed-forward control system is a technical process that uses information about input disturbances or signals to proactively implement corrective measures to achieve the desired output. This process aims to minimize the impact of disturbances before they affect the system by applying predictive correction. In this case, the need for adjustment can be detected before the targeted parallelism deviates from an acceptable range.

[0039] This can be exemplified by implementing the logic with a known trend function. According to one embodiment of the proposed solution, a SBF (Side Field Base) is calculated in an initial orientation and assigned to a point on the known trend function. Based on this point on the known trend function, it can be determined in which direction and by what amount the orientation must be changed now or at a later time to adjust the orientation to the optimum of the trend function or to maintain it there in the future.

[0040] The tendency function can be suitable for taking into account the influence of time-changing parameters, such as wear, temperature, or product specifications. This can make it possible to predict how the orientation needs to be changed in the future to ensure parallel alignment despite these changes.

[0041] As an example, the minimum standard deviation (SBF) of an initial orientation can first be determined. Subsequently, it can be ascertained whether the orientation is already at its optimum or deviates from it by a predetermined threshold. If the measured SBF deviates from the optimum by more than the predetermined threshold, the known trend function can be used to determine how the orientation can be adjusted to achieve the optimum. This can eliminate the need to manually adjust the orientation to determine the required adjustment, thus avoiding any potential negative impact on production.

[0042] Similarly, the tendency function can be used to determine whether the already optimal orientation will deviate from the optimum by more than a predetermined threshold in the future due to time-changing parameters if the orientation is not adjusted accordingly.

[0043] According to a further supplementary or alternative embodiment of the proposed method, the logic can be trained using artificial intelligence (AI) that is trained with measured or simulated values ​​of at least one SBF (Standard Flow Function) for a multitude of different orientations of the corrugated rollers. Accordingly, the tendency function does not need to be known in advance or determined based on a variation of the orientation in order to determine the adjustment requirement.

[0044] In particular, even in the implementation of the logic with a logic controller (K), no variation of the orientation is necessary to determine the adjustment requirement. Instead, the logic controller can be trained with historical data to determine, based on an initial orientation, in which direction and by how much the orientation needs to be changed to optimize the orientation or reduce the SBF (System Failure Mode). This can further improve the monitoring of parallelism.

[0045] In one exemplary configuration, the AI ​​can be implemented using a neural network. This can further increase the effectiveness of the logic.

[0046] According to a further supplementary or alternative embodiment of the proposed method, after and / or during each adjustment of the corrugated roller orientation, at least one SBF (Standard Functional Factor) for the respective orientation can be determined and added to the training data along with the orientation. This can enable continuous adaptation of the AI ​​(Action Factor) to the device. In particular, this allows for retraining, thereby continuously adapting the AI ​​to aging and wear effects.

[0047] For example, the AI ​​can be retrained at regular intervals in response to an explicit command or when a predetermined change in the training data set is detected.

[0048] The predetermined change can be the addition of training data or a change in a predetermined relative or absolute number of training data points since the last training. For example, retraining can be performed whenever 1% of the training data has changed.

[0049] With a further refinement of the proposed method, the vibration frequency spectrum of at least one of the corrugated rollers can be determined by measuring a vibration within a specific time period. A vibration period corresponds to the displacement of a point as a function of time. Such a measurement can be implemented particularly cost-effectively.

[0050] According to a further supplementary or alternative embodiment of the proposed method, the vibration measured over time can be transformed into the frequency domain. This can enable a particularly effective and cost-efficient determination of a vibration frequency spectrum.

[0051] For example, a translation can be implemented using a Fourier transform.

[0052] According to a further supplementary or alternative embodiment of the proposed procedure, the output of the adjustment requirement can include displaying the adjustment requirement for a person. This can enable the person to optimize the alignment of the corrugated rollers in a targeted and reliable manner. Extensive experience or training of the person may therefore be unnecessary.

[0053] The indication of the adjustment requirement can be provided in the form of instructions. These instructions can include guidance on how to adjust the alignment of the corrugated roller.

[0054] It is also conceivable and possible that, in response to the detection of a need for adjustment, a signal is sent to remote monitoring personnel. According to the proposed method, a device set up for this purpose can therefore specifically inform individuals when, within the framework of the proposed method, it is detected that the parallelism of the corrugated rollers needs to be readjusted.

[0055] According to a further supplementary or alternative embodiment of the proposed method, the alignment of the corrugated rollers can be automatically adjusted as required by an output device equipped with an electronically controlled motor or actuator (power-operated adjustment). Accordingly, the output of the adjustment requirement can include sending an adjustment signal for power-operated adjustment of the alignment. Thus, the adjustment requirement can be determined fully automatically and computer-aided, and the alignment of the corrugated rollers can be controlled accordingly. This can further reduce personnel costs and enable fully automatic operation of a machine configured for the proposed method.

[0056] According to one embodiment of the proposed method, the deviation of at least one SBF (Standard Field Observation) from a target value can be determined to ascertain the adjustment requirement. In other words, using at least one SBF and an associated target value, a needs assessment can be performed to determine whether the alignment requires adjustment, i.e., whether an adjustment is necessary. This can enable a particularly fast, reliable, and automated needs assessment. In particular, fully automated alignment control is possible.

[0057] For example, the target value can be a predetermined value. This can further simplify the implementation of the demand check. Alternatively, the target value can be learned during the operation of a device set up for the proposed procedure, or adjusted from an initial value. Accordingly, an optimum of at least one demand factor (DF) can be identified during device operation and continuously or regularly stored as the target value. This allows for adjustments to be made to the optimum of at least one DF that may have changed due to aging, maintenance, or environmental conditions. For example, this can prevent a change in the optimum caused by replacing individual components in the device, which could lead to an incorrect demand check.

[0058] According to a further embodiment of the proposed method, a target alignment value can be determined using the logic for predicting at least one SBF (Single Footprint). The alignment can then be automatically adjusted according to this target value. This allows for targeted control of the corrugated roller alignment in response to a detected need for adjustment.

[0059] For example, the logic can predict a target orientation suitable for shifting at least one SBF (Standard Building Function) from its current value towards the target value. This can enable continuous automatic optimization of the orientation, even if a corresponding orientation cannot be directly derived from the SBF's target value or if such derivation involves considerable computational effort. Thus, even with complex relationships between the at least one SBF and the orientation, automated monitoring and / or control of the orientation can be implemented in a resource-efficient manner.

[0060] As a further example, the logic can predict a target orientation that realizes the setpoint of at least one SBF. In other words, the at least one SBF for the predicted orientation can correspond to the setpoint of at least one SBF, except for a prediction error. This can enable a rapid correction of the orientation. In particular, a control loop can be implemented in this way that quickly and precisely corrects any impermissible orientation.

[0061] According to a further embodiment of the proposed method, the target alignment value can be determined in response to the deviation of at least one SBF exceeding a predetermined value. In other words, a needs check can first be performed to verify whether the at least one SBF deviates from a target value by a predetermined amount. If the check is negative, the method can be terminated, or the method can be repeated to implement continuous monitoring. If the check is positive, a target alignment value can be determined using the logic for predicting the at least one SBF. This target value can be output to an electronically controlled output device, which is configured to adjust the alignment of the corrugated rollers according to the target alignment value using an external force, i.e., an actuator.The method can therefore regulate the alignment of the corrugated rollers.

[0062] The aforementioned problem is solved by a device for manufacturing corrugated board that monitors the parallelism of two corrugating rollers that are adjustable relative to each other. The proposed device comprises at least the following:

[0063] - two corrugating rollers adjustable relative to each other, designed for the production of corrugated cardboard,

[0064] - at least one sensor for measuring the vibration of at least one of the corrugated rollers during normal operation,

[0065] - at least one output device for outputting an adjustment requirement with regard to the alignment of the corrugated rollers, and

[0066] - a control unit connected to the output device and the at least one sensor, which is configured to determine a vibration frequency spectrum of the at least one corrugated roller based on the measured vibration, to determine at least one sideband level factor of the vibration frequency spectrum, to determine an adjustment requirement depending on the sideband level factor, and to output the adjustment requirement via the output device.

[0067] The sideband level factor (SBF), derived from the vibration frequency spectrum, allows for the assignment of a numerical value to the parallelism of the axes of both corrugating rollers. This makes parallelism quantifiable. Furthermore, deviations from parallelism can be detected using the SBF. Thus, the SBF enables an objective determination of adjustment requirements. The operator's experience is not a factor. In particular, the adjustment requirement can be determined and displayed by computer, allowing a machine operator to adjust the alignment of the corrugating rollers to optimize parallelism. Personnel costs and production downtime can therefore be reduced by the proposed device.According to a further supplementary or alternative embodiment of the proposed device, the device can comprise a plurality of sensors, each configured to measure vibration on a section of one of the corrugated rollers. Each of the plurality of sensors can be coupled to the control unit and configured to transmit the measured vibration data to the control unit. The control unit can use the received vibration data to determine a vibration frequency spectrum for each of the sections. The control unit can further be configured to compare the plurality of vibration frequency spectra with one another and, based on the comparison of the vibration frequency spectra, to determine an adjustment requirement.In particular, the need for adjustment can be identified when a deviation of the majority of the vibration frequency spectra for different sections of the same corrugated roller exceeds a predetermined value.

[0068] According to a further supplementary or alternative embodiment of the proposed device, the device can include a remote computer configured to determine the adjustment requirement. For example, the control unit can be configured to determine the at least one SBF (Selective Footprint) and send the SBF with its corresponding orientation to the remote computer. The remote computer can receive the data pair and be configured to determine an adjustment requirement. The computer can send the determined adjustment requirement to the control unit. The control unit can be configured to receive the adjustment requirement and send it to the output device.

[0069] It is also conceivable and possible that the control unit sends the vibration measured for at least one orientation and / or the vibration frequency spectrum determined from the vibration for at least one orientation to the remote computer. This can reduce the necessary computing capacity at the location of the control unit.

[0070] It is also conceivable and possible that the remote computer is equipped with memory on which the control function for determining the adjustment requirement and / or a pre-known tendency function is stored. The remote computer and / or the control unit can be configured to receive and send data to each other via either a wireless or a wired connection. Furthermore, the remote computer can be configured to receive data from multiple control units. For example, it is conceivable and possible that the remote computer receives SBF alignment data pairs from different control units, each of which has acquired this data from different pairs of corrugated rollers.

[0071] The remote computer can use the data pairs from the various control units to train a computer. This can improve the size of the data set and thus the quality of the specific adjustment requirement.

[0072] In principle, it is conceivable and possible for the proposed device to include one or more vibration sensors on each or both of the corrugated rollers. As described above, a local difference in the SBF (Standard Flow Rate) can be determined using multiple sensors on one of the corrugated rollers. It is also conceivable and possible to use multiple sensors for redundant measurement to avoid an erroneous adjustment requirement in the event of a sensor malfunction.

[0073] The at least one sensor can be configured as an accelerometer mounted on one of the corrugated rollers. This can enable a cost-effective implementation of the vibration measurement. It is also conceivable and possible for the at least one sensor to be configured as an interferometer.

[0074] To display the adjustment requirement for a person, the device can be equipped with a display device, in particular a screen. The display device can be coupled to the control unit and configured to receive and display the adjustment requirement and / or an instruction from the control unit. It is also conceivable that the display device is a mobile, wireless device. For example, the device can be configured to transmit the determined adjustment requirement via radio link to a person who is not necessarily present on site.

[0075] It is also conceivable that the adjustment requirement could be sent to a mobile device, such as a mobile phone, belonging to a designated person. This could improve remote monitoring of the proposed device.

[0076] In a further embodiment of the proposed method, the proposed device can be equipped with an electronically controlled adjustment device configured to adjust the alignment of the corrugated rollers. The adjustment device can be coupled to the control unit and configured to send an adjustment signal to the device. This signal can include information regarding the adjustment requirement, enabling the alignment of the corrugated rollers to be adjusted electronically according to this requirement. This allows for fully automatic control. Therefore, the control unit can be configured to regulate the alignment.

[0077] According to one conceivable and possible embodiment of the proposed device, the at least one control unit can be configured to determine the adjustment requirement by detecting a deviation of the at least one SBF (Selective Beam Position) from a setpoint. In other words, using the at least one SBF and an associated setpoint, a needs check can be performed to determine whether the alignment needs to be adjusted, i.e., whether an adjustment is required. This can enable a particularly fast, reliable, and automated needs check.

[0078] According to a further embodiment of the proposed device, the at least one control unit can be configured to determine a target alignment value based on the deviation of the at least one SBF (Standard Field of Operation) using logic for predicting the at least one SBF. This can enable targeted alignment control in response to a detected need for adjustment.

[0079] According to a further embodiment of the proposed device, the at least one control unit can be configured to determine the target value of the alignment in response to the deviation of the at least one SBF exceeding a predetermined value.

[0080] In particular, the proposed device can be part of a single-facer or a double-facer.

[0081] According to a further embodiment of the proposed device, the device can be set up to carry out the proposed procedure.

[0082] Furthermore, the aforementioned problem is also solved by a computer program product that includes machine-readable instructions which, upon execution, cause a control unit of the proposed device to carry out the proposed method. The proposed statements regarding the advantages and possible configurations of the proposed method apply analogously to the proposed device and the proposed computer program product.

[0083] The attached figures illustrate possible implementation variants of the proposed solution.

[0084] This shows:

[0085] Figure 1 shows a schematic representation of a first embodiment of the proposed device for corrugated board production with monitoring of corrugating roll parallelism,

[0086] Figure 2 shows a schematic representation of another

[0087] Implementation variant of the proposed device, which is designed for the automatic adjustment of the corrugated roller parallelism,

[0088] Figure 3 shows a schematic representation of a variant embodiment of the proposed device with a remote computer.

[0089] Figure 4 shows a schematic vibration frequency spectrum of the

[0090] Corrugated rollers,

[0091] Figure 5 shows a vibration frequency spectrum of the corrugated rollers based on measurement data.

[0092] Figure 6 shows a schematic representation of a tendency function of at least one sideband level factor for different orientations, and

[0093] Figures 7 to 11 show flowcharts of different embodiments of the proposed method.

[0094] Figure 1 shows a schematic representation of the proposed device for producing corrugated board with monitoring of the parallelism of two corrugating rollers 100, 102 that are adjustable relative to each other. The device comprises a first corrugating roller 100, which is pressed against a second corrugating roller 102 with a contact pressure. The corrugating rollers 100, 102 can be adjusted in their relative orientation. The two corrugating rollers 100, 102 are configured to produce corrugated board. A sensor 300 is arranged on the first corrugating roller 100, 102 for measuring a vibration of the first corrugating roller 100 during the intended operation of the corrugating rollers 100, 102. The sensor 300 is coupled to a control unit 400, which is configured to receive the data measured by the at least one sensor 300 and to determine a vibration frequency spectrum based on the measured data.Furthermore, the control unit 400 is configured to calculate at least one sideband level factor (SBF) of the vibration frequency spectrum and, based on the SBF, to determine an adjustment requirement using logic for predicting the SBF depending on the orientation. The adjustment requirement describes whether, in which direction, and by what value the orientation should be adjusted to reduce vibration of the first corrugated roller 100.

[0095] Furthermore, an output device 500 is coupled to the control unit 400. The control unit 400 is configured to send the specified adjustment requirement to the output device 500. The output device 500 is configured to output the adjustment requirement.

[0096] In further embodiments that differ from the embodiment shown in Figure 1, in principle any other form of a pressure device 200 is conceivable which is suitable for introducing a pressure into one of the corrugated rollers 100, 102.

[0097] In principle, the device can also have a different number of sensors 300 for measuring the vibration than the one sensor 300 shown here. The at least one sensor 300 can be arranged on each of the corrugated rollers 100, 102. In the case of multiple sensors 300, these can be distributed across both corrugated rollers 100, 102 or arranged on one corrugated roller 100, 102.

[0098] Figure 2 shows an embodiment with an electronically controllable alignment of the second corrugated roller 102. For this purpose, the output device 502 is designed with an electronically adjustable setting device for adjusting the alignment of the corrugated rollers. This can be, for example, an electronically adjustable bearing of the second corrugated roller 102. Accordingly, in response to the determination of an adjustment requirement, an adjustment signal can be sent to the setting device 502 to adjust the alignment of the second corrugated roller 102 relative to the first corrugated roller 100 according to the adjustment requirement. The output of the adjustment requirement thus includes the adjustment of the alignment. The embodiment shown in Figure 2 is therefore designed for automatic, computer-aided monitoring and control or regulation of the alignment.

[0099] In principle, it is conceivable and possible that the proposed device includes a remote computer that performs individual parts of the proposed procedure.

[0100] Figure 3 shows a possible embodiment with a remote computer 402 coupled to the control unit 400. The coupling encompasses any type of connection that enables data exchange between the control unit 400 and the remote computer 402. In the embodiment shown in Figure 3, the control unit 400 is configured to determine the vibration frequency spectrum via the sensor 300 and send it to the remote computer 402. The remote computer 402 is configured to receive the vibration frequency spectrum for one or more orientations and, using artificial intelligence and / or a pre-known trend function and / or a trend function determined from the SBF (System Based Function), to determine the adjustment requirement. The adjustment requirement is sent back to the control unit 400. The control unit 400 then sends the adjustment requirement to the output device 502.

[0101] Figure 4 shows a schematic vibration frequency spectrum 600 as a function of the vibration amplitude a versus the frequency f. Vibration amplitudes a for various frequency ranges are included. At a tooth engagement frequency 602 of the corrugating rollers 100, 102, as well as at multiples of the tooth engagement frequency 603 (harmonic overtone), the vibration frequency spectrum 600 exhibits amplitudes a of large magnitude. Above both the tooth engagement frequency 602 and the harmonic overtone 603, broader frequency ranges with lower vibration amplitudes 604, 605, 606, 607 (secondary bands) are found. The secondary bands 604, 605, 606, 607 can be divided into upper secondary bands 604, 605 and lower secondary bands 606, 607. The lower secondary bands 606 and 607 can be found below the tooth engagement frequency 602 and the harmonic overtone 603.Accordingly, the upper secondary bands 604 and 605 are found above the tooth engagement frequency 602 and the harmonic overtone 603. To determine at least one secondary band, the proposed device can be configured to use only a portion of the vibration frequency spectrum 600, for example, the upper secondary bands 604 and 605, or exactly one upper secondary band 604 or 605.

[0102] Figure 5 shows a vibration frequency spectrum 600 based on a measurement, corresponding to the frequency distribution shown schematically in Figure 4.

[0103] Based on the vibration frequency spectrum 600, at least one SBF (Standard Frequency Beam) is calculated. If the orientation of the corrugated rollers 100, 102 is varied, the vibration behavior of the corrugated rollers 100, 102 changes. In particular, the secondary belts 604, 605, 606, 607 also change within the vibration frequency spectrum 600. For example, the amplitudes a and the width of the secondary belts can increase as the corrugated rollers 100, 102 vibrate more strongly. This change leads to a change in at least one SBF.

[0104] Figure 6 shows three values ​​marked with an "x" for the at least one SBF (System for Operational Function) at three different orientations x1, x2, x3 in an SBF orientation diagram. The orientation x can represent any quantity suitable for describing a relative position. In particular, this can include angles and distances or adjustment states of actuators. For values ​​of the orientation variable x shown further to the right in the diagram, the SBF values ​​shown in Figure 6 exhibit a decreasing trend. A decreasing SBF value is associated with a reduced vibration of the corrugated rollers 100, 102. To determine the adjustment requirement, a regression line F_x1 is determined for the three specified values ​​x1, x2, x3 of the at least one SBF. Within the range of the measured values, the regression line F_x1 approximates the behavior of a functional relationship F_x2 between the SBF and the orientation, which may be unknown in practice.The regression line is therefore a trend function F_x1 in the sense of the proposed solution. From the slope, or gradient, of the regression line F_x1, it can be seen that an optimum of at least one SBF with respect to the orientation x can be found for values ​​of x greater than x3.

[0105] The adjustment requirement derived from the tendency function F_x1 thus includes at least the specification that the orientation x must be adjusted by increasing x relative to x3. Alternatively, the adjustment requirement can also include a specific value by which the orientation variable x must be increased. This amount can depend, particularly linearly, on the slope of the tendency function F_x1. By repeatedly determining the tendency function F_x1 after adjustment according to a previously determined adjustment requirement, the device converges in the direction of the optimal orientation x4.

[0106] Figures 7 to 11 show exemplary sequences of the proposed procedure.

[0107] The embodiment of the proposed method shown in Figure 7 comprises determining the vibration frequency spectrum 600, determining the at least one SBF, and determining an adjustment requirement with regard to the orientation x. The adjustment requirement is then output.

[0108] In further alternative embodiments of the proposed method, determining the at least one SBF can involve varying the orientations x. Based on this variation, it is possible to determine a trend function F_x1, which approximates a potentially unknown functional relationship F_x2 between the SBF and the orientation x.

[0109] Figure 8 shows a possible embodiment of the method in which the orientation x is varied and a trend function F_x1 is determined. In particular, the orientation x is adjusted such that at least one SBF (single-point fluctuation) is determined for three different orientations x. Determining the trend function F_x1 involves calculating a regression line. From the regression line F_x1, it is derived how the orientation x must be varied to reduce the SBF, or the vibration of the corrugated rollers 100, 102.

[0110] In principle, the trend function F_x1 can include not only a regression line, but any polynomials.

[0111] Figure 9 shows another possible sequence of the proposed procedure. Here, a vibration frequency spectrum is determined for a plurality of sections of one of the corrugating rollers that are subject to discrepancies. Subsequently, the SBF (Standard Frequency Range) for the different sections on the corrugating roller and any deviations are determined. Based on the deviation, an adjustment requirement is identified.

[0112] In alternative or supplementary configurations, it is also conceivable and possible that the logic for predicting the at least one SBF depending on the orientation x is trained with an artificial intelligence that is trained with measured and / or simulated values ​​of the at least one SBF for a large number of different orientations x.

[0113] Furthermore, it is conceivable and possible that the output of the adjustment requirement includes an externally driven adjustment of the orientation x. Thus, the entire proposed procedure can be executed using a computer. In particular, control of the orientation x can be implemented in this way.

[0114] Figure 10 shows a possible embodiment of the method as a computer-aided process. After the process is started, a vibration frequency spectrum 600 of at least one of the corrugated rollers 100, 102 is first determined by sensors. Based on the vibration frequency spectrum 600, the at least one SBF (Surface Flow Unit) is determined. As explained above for determining the adjustment requirement, a check is then performed to determine whether an adjustment is required. If no adjustment requirement is detected, the vibration frequency spectrum 600 is determined again. This enables continuous monitoring of the alignment x. However, if an adjustment requirement is detected, the alignment x can be adjusted fully automatically and in a computer-aided manner via the output device 502, 504, which is equipped with an electronically controlled bearing, according to the determined adjustment requirement.

[0115] Figure 11 shows a further embodiment of the proposed method as an automated control procedure. In contrast to the embodiment shown in Figure 9, here, after determining the at least one sideband level factor (SBF), a deviation of the at least one sideband level factor (SBF) from a setpoint is determined. Subsequently, it is checked whether the deviation exceeds a predetermined value. If the check is negative—that is, if the predetermined value is not exceeded—the procedure is repeated, meaning another vibration frequency spectrum (600 Hz) is determined. If the check is positive, a target value for the orientation x is determined using the logic for predicting the at least one sideband level factor (SBF).This target value of the orientation x is output to an electronically controlled output device 502, 504, in order to adjust the orientation x according to the target value by external force, i.e., via an actuator. The process is then repeated. Thus, this implementation variant provides continuous monitoring and control of the orientation.

[0116] The proposed solution is not limited to the specific embodiments discussed here. Rather, the proposed solution encompasses any combination of features from the discussed embodiments, provided that these can be combined in a feasible manner by those skilled in the art.

[0117] Reference symbol list

[0118] 100, 102 Grooved roller

[0119] 200 pressure device

[0120] 202 Tension roller

[0121] 204 Deflection roller

[0122] 206 Pressure band

[0123] 300 Sensor

[0124] 400 control unit

[0125] 402 remote computer

[0126] 500, 502, 504 Output device

[0127] 600 vibration frequency spectrum

[0128] 602 Tooth engagement frequency

[0129] 603 first harmonic of the tooth engagement frequency

[0130] 604 upper accessory band of tooth engagement frequency

[0131] 605 upper secondary band of the first harmonic

[0132] 606 lower accessory band of tooth engagement frequency

[0133] 607 lower secondary band of the first harmonic f frequency a amplitude

[0134] SBF sideband level factor x, x1, x2, x3, x4 alignment

[0135] F, F1, F2, F3, F4 Sideband level factor F_x1, trend function F_x2 functional relationship

Claims

Claims 1. Method for monitoring the parallelism of two corrugating rollers (100, 102) adjustable relative to each other in a device for corrugated board production, comprising: - Sensory determination of at least one vibration frequency spectrum (600) of two rotating corrugating rollers (100, 102) designed for the production of corrugated board laminated on at least one side, - Determine at least one sideband level factor (SBF) based on the vibration frequency spectrum (600), - Determining the adjustment requirement of an alignment (x, x1 , x2, x3, x4) of the Corrugated rollers (100, 102) relative to each other depending on the Sideband level factor (SBF), and - Outputting the adjustment requirement.

2. Method according to claim 1, characterized in that the adjustment requirement is determined using a logic for predicting the at least one sideband level factor (SBF) as a function of the parallelism.

3. Method according to claim 2, characterized in that the logic comprises determining a trend function (F_x1) that approximately describes a development of the at least one sideband level factor (SBF) as a function of the orientation (x, x1 , x2, x3, x4).

4. Method according to claim 3, characterized in that the trend function (F_x1) is formed with a regression line between at least three sideband level factors (SBF) measured at different orientations (x, x1 , x2, x3, x4).

5. Method according to one of the preceding claims, characterized in that the logic is designed with a Cl which is trained with measured and / or simulated values ​​of the at least one sideband level factor (SBF) at a plurality of different orientations (x, x1 , x2, x3, x4) of the corrugated rollers (100, 102 ).

6. Method according to claim 5, characterized in that after and / or during each adjustment of the orientation (x, x1 , x2, x3, x4) of the corrugated rollers (100, 102) the at least one sideband level factor (SBF) for the respective orientation (x, x1 , x2, x3, x4) is determined and added to the training data together with the orientation (x, x1 , x2, x3, x4).

7. Method according to one of the preceding claims, characterized in that, to determine the vibration frequency spectrum (600), a vibration of at least one of the corrugated rollers (100, 102) is measured during the period.

8. Method according to claim 7, characterized in that the oscillation measured in the period is transformed into the frequency domain.

9. Method according to one of the preceding claims, characterized in that the output includes displaying the adjustment requirement for a person.

10. Method according to one of the preceding claims, characterized in that the output includes sending an adjustment signal for externally actuated adjustment of the alignment (x, x1 , x2, x3, x4) according to the order requirements.

11. Method according to one of the preceding claims, characterized in that a deviation of the at least one sideband level factor (SBF) from a target value is determined for determining the adjustment requirement.

12. Method of one of the preceding claims, insofar as it relates back to claim 2, characterized in that a target value of the alignment (x, x1 , x2, x3, x4) is determined using the logic for predicting the at least one sideband level factor (SBF).

13. Method according to claims 11 and 12, characterized in that the target value of the alignment (x, x1 , x2, x3, x4) is determined in response to the deviation of the at least one sideband level factor (SBF) exceeding a predetermined value.

14. Device for monitoring the parallelism of two corrugated rollers (100, 102) adjustable relative to each other, comprising: - two corrugating rollers (100, 102) adjustable relative to each other, which are set up for the production of corrugated board laminated on at least one side, - at least one sensor (300) for measuring a vibration of at least one of the corrugated rollers (100, 102) during intended operation, - a control unit (400) connected to at least one sensor (300), and - at least one output device (500, 502, 504) connected to the control unit (400) for outputting an adjustment requirement with regard to an orientation (x, x1 , x2, x3, x4) of the corrugated rollers (100, 102), wherein the control unit (400) is configured to determine a vibration frequency spectrum (600) of the at least one corrugated roller (100, 102) based on the measured vibration, to determine at least one sideband level factor (SBF) of the vibration frequency spectrum (600), to determine an adjustment requirement depending on the sideband level factor (SBF) and to output the adjustment requirement via the output device (500, 502, 504).

15. Device according to claim 14, characterized in that the control unit (400) is configured to determine a deviation of the at least one sideband level factor (SBF) from a setpoint value for determining the adjustment requirement.

16. Device according to claim 14 or 15, characterized in that the control unit (400) is configured with logic for predicting the at least one sideband level factor (SBF) as a function of the parallelism of the corrugated rollers to determine a target value of the alignment (x, x1 , x2, x3, x4).

17. Device according to claim 16, insofar as it relates back to claim 15, characterized in that the control unit (400) is configured to determine the target value of the alignment (x, x1 , x2, x3, x4) in response to the deviation of the at least one sideband level factor (SBF) exceeding a predetermined value.

18. Device according to one of claims 14 to 17, characterized in that the output device (500, 502, 504) is configured for graphical output of the adjustment requirement and / or for externally actuated adjustment of the alignment (x, x1 , x2, x3, x4).

19. Machine for the production of corrugated board laminated on at least one side, comprising a device according to one of claims 14 to 18.

20. Computer program product comprising machine-readable instructions which the control unit (400) of a device according to any one of claims 14 to 18 in the To initiate the execution of a method according to any one of claims 1 to 13.

Images