Method and device for analyzing and / or monitoring state of chain stress

Abstract

The invention relates to a method for analyzing and / or monitoring the stress state of a chain, comprising the following steps: acquiring bulk acoustic signals from the operation of the chain, analyzing the acquired bulk acoustic signals, and determining the lubrication state and elongation of the chain. The invention also relates to a device for analyzing the running state of the chain, which comprises a body acoustic sensor and an evaluation unit.

Description

Methods and apparatus for analyzing and / or monitoring chain stress state

[0001] This invention relates to a method for analyzing and / or monitoring the stress state of a chain, the method comprising the steps of: acquiring bulk acoustic signals from chain operation, analyzing the acquired bulk acoustic signals, and determining the chain's lubrication state and elongation. The invention also relates to a chain operation state analysis device, which includes a bulk acoustic sensor and an evaluation unit. Prior Art

[0002] Chain drives are used for driving or transportation purposes in a variety of industrial applications, as well as in bicycles and motorcycles. A complete chain drive includes a chain, typically an endless loop, multiple sprockets for chain steering, and drive or transport elements connected to and driven by the chain or driven together with the chain. The chain wears during operation due to the wear of the moving parts within the links. Other factors, such as chain elongation, tension, bearing clearance, and bearing wear during operation, can also cause chain wear and ultimately lead to drive unit failure. Other factors affecting chain wear include the forces acting on the chain and the load, as well as external influences determined by the environment. Due to the complexity of these relationships, chain wear cannot be predicted, and therefore, potential malfunctions or even drive unit failures during operation cannot be predicted.

[0003] With the increasing number of fully automated machines and equipment, complex chain drive systems are being used more and more, as they are essential for modern factory automation. Due to the huge investment costs of such high levels of automation and global price pressures, it is imperative to minimize machine and equipment downtime to an absolute minimum and completely prevent unplanned downtime.

[0004] In addition to direct economic losses, such unplanned downtime can lead to indirect problems, such as disrupting the logistics chain and making it impossible to meet delivery deadlines, resulting in further economic losses. Since wear or elongation of the drive chain is unavoidable and cannot be predetermined, continuous monitoring of the chain drive is essential for timely inspection, adjustment of synchronization, and replacement of damaged chains.

[0005] Furthermore, it is known from existing technology that wear on the drive chain is determined by measuring the force, path, or angle of the chain tensioner, or by using two angle sensors on the drive and load wheels. However, chain tensioners are not needed everywhere, and angle sensors are not always available. Moreover, these sensors are affected by wear or chain elongation. Such methods must be precisely tuned for specific applications because, in these cases, measurements depend on the total chain length and the wear of the sprockets. Tuning is tedious and error-prone. Therefore, these methods cannot be universally applied.

[0006] Therefore, the objective of this invention is to provide a method for analyzing and / or monitoring the stress state of a chain, which can reliably and quickly identify the stress state. Another objective of this invention is to provide a chain running state analysis device for analyzing and / or monitoring the stress state of a chain, which can reliably and quickly identify the stress state. A further objective of this invention is to provide a software program that can reliably and quickly identify the stress state of a chain.

[0007] This task is solved by the method for analyzing and / or monitoring the stress state of a chain according to claim 1. Advantageous embodiments of the invention are set forth in the dependent claims.

[0008] The method of this invention is used to analyze and / or monitor the stress state of a chain and has three steps: The first step is to acquire bulk acoustic signals from the chain during operation. These bulk acoustic signals can be read, for example, by a microphone, acoustic sensor, accelerometer, or similar sensor.

[0009] The second step is to analyze the acquired bulk acoustic signals. For this purpose, the bulk acoustic signals can be converted from a time-based representation to a frequency-based representation using Fast Fourier Transform (FFT) and / or Discrete Fourier Transform (DFT). Optionally, the bulk acoustic signals can also be smoothed and squared.

[0010] The third step is to determine the chain's lubrication condition and elongation. This is achieved by utilizing the frequency distribution and sound pressure levels of the bulk acoustic signal, either before or after the analysis of the acquired signal. For example, one metric is the kurtosis (sharpening) of the frequency distribution. By determining the kurtosis and sound pressure levels of the bulk acoustic signal frequency distribution, the chain's lubrication condition and elongation can be determined.

[0011] Chains are subjected to stress during operation, naturally leading to wear. Because chains are critical components, they are often prematurely maintained or replaced as a preventative measure, resulting in unnecessary costs. Delayed maintenance or replacement of the chain carries the risk of chain breakage under load. Within a predictive maintenance framework, understanding the chain's stress state is important, such as the duration of operation and the load conditions under which it has been running. However, since the chain and its associated drive components (sprockets, steering elements, etc.) form a complex system, directly capturing the chain's stress state is meaningful.

[0012] Chain wear is mainly affected by the chain lubrication condition due to the large number of moving parts: the better the chain lubrication, the less wear it will experience.

[0013] One indicator of chain wear is its elongation. Even slight chain elongation can cause production errors in processes synchronized via chain drives, requiring manual readjustment.

[0014] The method of this invention enables continuous acquisition of chain stress state by determining lubrication status and elongation. When the chain stress state reaches or exceeds a set threshold, appropriate measures (chain maintenance, replacement) can be taken in a timely manner to prevent chain failure. This also saves costs, as the chain is only maintained or replaced when necessary.

[0015] In one improvement of the invention, a qualitative analysis of the chain's lubrication condition and / or elongation is performed. Quantitative analysis of the chain's lubrication condition and / or elongation is not necessary. For timely chain maintenance, it is only necessary to know whether the lubrication condition is sufficiently high and / or whether the chain elongation is sufficiently low. This saves costs.

[0016] In another embodiment of the invention, maintenance reminders are derived and / or output from qualitative analysis. In another design of the invention, the maintenance reminders include information about the next lubrication and / or replacement of a portion or the entire chain. The reminders may also include warnings if the lubrication condition and / or elongation reach critical values. Such reminders can advantageously improve the plannability of maintenance and / or possible chain replacement, or reduce wear.

[0017] In another aspect of the invention, frequency analysis of the bulk acoustic signal is performed to determine the chain stress state. For this purpose, the frequency distribution of the bulk acoustic signal and its sound pressure level are utilized in particular. One metric is the kurtosis (sharpness) of the frequency distribution. By determining the kurtosis of the frequency distribution of the bulk acoustic signal and its sound pressure level, the chain lubrication state and elongation are determined.

[0018] In another aspect of this invention, the chain elongation is determined from a defined body acoustic signal frequency. Chain elongation results in a larger frequency distribution of the body acoustic signal frequency. The chain elongation is determined by determining the kurtosis of the frequency distribution of the body acoustic signal frequency.

[0019] In another design of the invention, the chain lubrication condition is determined from the shape and / or intensity of the determined bulk acoustic signal. Specifically, the frequency distribution area of ​​the bulk acoustic signal frequency and its sound pressure level are measures of the chain lubrication condition. The worse the chain lubrication condition, the higher the sound pressure level, and the larger the frequency distribution of the bulk acoustic signal frequency during chain operation.

[0020] In an advantageous design of the invention, training data is generated from the analyzed bulk acoustic data. In an improvement of the invention, the training data is used to train the KI / ML function. In another aspect of the invention, KI / ML data is generated from the training data. In an advantageous embodiment of the invention, the KI / ML function and / or KI / ML data are used to determine the stress state. Alternatively, complex algorithms or complex data analysis may be used.

[0021] The bulk acoustic signals of a chain in operation are typically very significant. Pattern recognition and / or machine learning methods, such as neural networks (models), can be advantageously used. For this purpose, bulk acoustic signals of unworn (e.g., initial state) and stressed chains are collected, associated with their respective machines and / or machine types, in relevant operating states and / or phases of motion.

[0022] Specific stress states can also be learned, which requires learning the bulk acoustic signals of the corresponding worn chains with known stress states. The stress state can be determined by variations in parameters such as chain elongation, clearance, chain plates, pins, bushings, rollers, etc. This can also be for individual links and / or chain components, as the weakest link can cause chain breakage.

[0023] In another design of the invention, the initial state of the chain is determined to ascertain the stress state. The initial state of the chain is particularly its new state. The monitored chain in its initial state exhibits optimal lubrication and is free from elongation due to plastic deformation.

[0024] In another design of the invention, the deviation between the actual state and the initial state of the chain is determined to ascertain the stress state. The deviation between the actual state and the initial state is a measure of the chain's stress state.

[0025] In another training exercise of this invention, KI / ML data analysis is used to determine the deviation between the actual state and the initial state of the chain. Alternatively, bulk acoustic signals can be used to learn the deviation. In this way, the determination of the deviation can be continuously improved.

[0026] In another embodiment of the invention, the method is placed in an initial state after the reset function is triggered. The method of the present invention can therefore be restarted at any time, particularly after lubrication, maintenance, or chain replacement, when the chain is in its initial state.

[0027] This task is also addressed by a chain running condition analysis device for analyzing and / or monitoring chain lubrication status and / or elongation. Advantageous embodiments of the invention are also set forth in the dependent claims.

[0028] The chain running status analysis device of the present invention is used to analyze and / or monitor the chain lubrication status and / or elongation, and includes a bulk acoustic sensor and an evaluation unit. The bulk acoustic sensor can collect bulk acoustic signals during chain operation. The evaluation unit is designed and adapted to analyze the chain elongation and lubrication status from the bulk acoustic signals collected by the bulk acoustic sensor.

[0029] In this invention, bulk acoustic signals are signals that exist in the form of mechanical vibrations within the acoustic frequency range. These include all signals generated by the vibration of an object. Pure acoustic signals are also bulk acoustic signals in the sense of this invention.

[0030] In this invention, a body acoustic sensor is any sensor capable of detecting body acoustic signals and / or signals generated by body acoustics. It also includes sensors that detect purely acoustic signals. Examples include microphones, sound sensors, body acoustic sensors, accelerometers such as piezoelectric accelerometers, or similar sensors.

[0031] In another embodiment of the invention, the evaluation unit has a first memory and / or storage area for storing the acquired bulk acoustic signals. Optionally, bulk acoustic signals of the initial state of the chain may also be stored.

[0032] In another design of the invention, the evaluation unit has a second memory and / or storage area for storing first KI / ML data, which is designed and adapted to analyze the lubrication status of the chain from the acquired bulk acoustic signals.

[0033] In another training exercise of the present invention, the evaluation unit has a third memory and / or storage area in which second KI / ML data is stored, the second KI / ML data being designed and adapted to analyze the chain elongation from the acquired stereoscopic signals.

[0034] In one improvement of the invention, the evaluation unit has a fourth memory and / or storage area, wherein the result data of the volume acoustic signal frequency analysis is stored.

[0035] In another design of the invention, the evaluation unit has a fifth memory and / or storage area, wherein first training data for training a first KI / ML function is stored, the first KI / ML function being designed and adapted to determine the lubrication state of the chain from the first KI / ML data.

[0036] In one improvement of the invention, the evaluation unit has a sixth memory and / or storage area, wherein second training data for training a second KI / ML function is stored, the second KI / ML function being designed and adapted to determine the amount of chain elongation from the second KI / ML data.

[0037] In an advantageous design of the invention, training data is generated from the analyzed bulk acoustic data. In an improvement of the invention, the training data is used to train the KI / ML function. In another aspect of the invention, KI / ML data is generated from the training data. In an advantageous embodiment of the invention, the KI / ML function and / or the KI / ML data are used to determine the stress state.

[0038] The bulk acoustic signals of a chain in operation are typically very significant. Pattern recognition and / or machine learning methods, such as neural networks (models), can be advantageously used. For this purpose, bulk acoustic signals of unworn (e.g., initial state) and stressed chains are collected, associated with their respective machines and / or machine types, in relevant operating states and / or phases of motion.

[0039] Specific stress states can also be learned, which requires learning the bulk acoustic signals of the corresponding worn chains with known stress states. The stress state can be determined by variations in parameters such as chain elongation, clearance, chain plates, pins, bushings, rollers, etc. This can also be for individual links and / or chain components, as the weakest link can cause chain breakage.

[0040] In another aspect of the invention, the evaluation unit includes a processor designed and adapted to execute a software program of the method of the invention according to claims 1 to 16.

[0041] This task is also solved by a software program of the method of the present invention according to one or more of claims 1 to 16.

[0042] The software program has program code that can be stored on a machine-readable carrier or storage medium such as semiconductor memory, hard disk memory or optical memory, and is used to execute, implement and / or control method steps according to one of the above embodiments, particularly when the program product or program is running on a computer or device.

[0043] This task is also addressed by methods for analyzing and / or monitoring the chain lubrication condition. Advantageous embodiments of the invention are also set forth in the dependent claims.

[0044] The method of the present invention for analyzing and / or monitoring the lubrication status of a chain has three steps: the first step is to collect bulk acoustic signals from the chain during operation. The bulk acoustic signals can be read using a microphone.

[0045] The second step is to analyze the acquired bulk acoustic signals. For this, a Fast Fourier Transform (FFT) can be used to convert the bulk acoustic signals from a time-based representation to a frequency-based representation. The bulk acoustic signals can also be smoothed and squared.

[0046] The third step is to determine the chain's lubrication condition. This is achieved by utilizing the frequency distribution and sound pressure level of the bulk acoustic signal. One metric is the kurtosis of the frequency distribution. By determining the kurtosis and sound pressure level of the bulk acoustic signal's frequency distribution, the chain's lubrication condition can be determined.

[0047] Chains are subjected to stress during operation, naturally leading to wear. Because chains are critical components, they are often prematurely maintained or replaced as a preventative measure, resulting in unnecessary costs. Delayed maintenance or replacement of the chain carries the risk of chain breakage under load. Within a predictive maintenance framework, understanding the chain's stress state is important, such as the duration of operation and the load conditions under which it has been running. However, since the chain and its associated drive components (sprockets, steering elements, etc.) form a complex system, directly capturing the chain's stress state is meaningful.

[0048] Chain wear is mainly affected by the chain lubrication condition due to the large number of moving parts: the better the chain lubrication, the less wear it will experience.

[0049] The method of this invention enables continuous acquisition of the chain lubrication status by determining the lubrication condition. When the chain lubrication status reaches or exceeds a set threshold, appropriate measures (chain maintenance, replacement) can be taken in a timely manner to prevent chain failure. This also saves costs, as the chain is only maintained or replaced when necessary.

[0050] In one improvement of the invention, the lubrication condition of the chain is qualitatively analyzed. Quantitative analysis of the chain lubrication condition is not necessary. For timely chain maintenance, it is only necessary to know whether the lubrication condition is sufficiently high. This saves costs.

[0051] In another embodiment of the invention, maintenance reminders are derived and / or output from qualitative analysis. In another design of the invention, the maintenance reminders include information about the next lubrication and / or replacement of a portion or the entire chain. The reminders may also include warnings if lubrication conditions reach a critical level. These reminders can advantageously improve the plannability of maintenance and / or possible chain replacement, or reduce wear.

[0052] In another aspect of the invention, frequency analysis of the bulk acoustic signal is performed to determine the chain stress state. For this purpose, the frequency distribution of the bulk acoustic signal and its sound pressure level are utilized in particular. One metric is the kurtosis (sharpness) of the frequency distribution. By determining the kurtosis of the frequency distribution of the bulk acoustic signal and its sound pressure level, the chain lubrication state and elongation are determined.

[0053] In another aspect of this invention, the chain elongation is determined from a defined body acoustic signal frequency. Chain elongation results in a larger frequency distribution of the body acoustic signal frequency. The chain elongation is determined by determining the kurtosis of the frequency distribution of the body acoustic signal frequency.

[0054] In another design of the invention, the chain lubrication condition is determined from the shape and / or intensity of the determined bulk acoustic signal. Specifically, the frequency distribution area of ​​the bulk acoustic signal frequency and its sound pressure level are measures of the chain lubrication condition. The worse the chain lubrication condition, the higher the sound pressure level, and the larger the frequency distribution of the bulk acoustic signal frequency during chain operation.

[0055] In another embodiment of the present invention, the body sound data is analyzed by frequency analysis.

[0056] In an advantageous design of the invention, training data is generated from the analyzed bulk acoustic data. In an improvement of the invention, the training data is used to train the KI / ML function. In another aspect of the invention, KI / ML data is generated from the training data. In an advantageous embodiment of the invention, the KI / ML function and / or the KI / ML data are used to determine the lubrication state.

[0057] The bulk acoustic signals of chains in operation are typically very significant. Pattern recognition and / or machine learning methods, such as neural networks (models), are advantageously used. For this purpose, bulk acoustic signals are acquired from unworn (e.g., initial state) and stressed chains, respectively, in their associated machine and / or machine type, during relevant operating states and / or phases of motion. Specific lubrication conditions can also be learned; for this, the bulk acoustic signals of correspondingly worn chains with known lubrication conditions must be learned.

[0058] In another design of the invention, the initial state of the chain is determined to ascertain the lubrication condition. The initial state of the chain is specifically a new state of the chain. The monitored chain in its initial state exhibits optimal lubrication and no elongation.

[0059] In another design of the invention, the deviation between the actual state and the initial state of the chain is determined to determine the lubrication state. The deviation between the actual state and the initial state is a measure of the chain stress state.

[0060] In another training exercise of this invention, KI / ML data analysis is used to determine the deviation between the actual state and the initial state of the chain. Alternatively, bulk acoustic signals can be used to learn the deviation. In this way, the determination of the deviation can be continuously improved.

[0061] In another embodiment of the invention, the method is placed in an initial state after the reset function is triggered. The method of the present invention can therefore be restarted at any time, particularly after lubrication, maintenance, or chain replacement, when the chain is in its initial state.

[0062] This task is also addressed by a chain lubrication condition analysis device for analyzing and / or monitoring the chain lubrication condition. Advantageous embodiments of the invention are also set forth in the dependent claims.

[0063] The chain lubrication condition analysis device of the present invention is used to analyze and / or monitor the chain lubrication condition and / or elongation, and includes a bulk acoustic sensor and an evaluation unit. The bulk acoustic sensor can collect bulk acoustic signals from the chain during operation. The evaluation unit is designed and adapted to analyze the chain lubrication condition from the bulk acoustic signals collected by the bulk acoustic sensor.

[0064] In another embodiment of the invention, the evaluation unit has a first memory and / or storage area for storing the acquired bulk acoustic signals. Optionally, bulk acoustic signals of the initial state of the chain may also be stored.

[0065] In another design of the invention, the evaluation unit has a second memory and / or storage area for storing first KI / ML data, which is designed and adapted to analyze the lubrication status of the chain from the acquired bulk acoustic signals.

[0066] In another training exercise of the present invention, the evaluation unit has a third memory and / or storage area in which second KI / ML data is stored, the second KI / ML data being designed and adapted to analyze the chain elongation from the acquired stereoscopic signals.

[0067] In one improvement of the invention, the evaluation unit has a fourth memory and / or storage area, wherein the result data of the volume acoustic signal frequency analysis is stored.

[0068] In another design of the invention, the evaluation unit has a fifth memory and / or storage area, wherein first training data for training a first KI / ML function is stored, the first KI / ML function being designed and adapted to determine the lubrication state of the chain from the first KI / ML data.

[0069] In one improvement of the invention, the evaluation unit has a sixth memory and / or storage area, wherein second training data for training a second KI / ML function is stored, the second KI / ML function being designed and adapted to determine the amount of chain elongation from the second KI / ML data.

[0070] In an advantageous design of the invention, training data is generated from the analyzed bulk acoustic data. In an improvement of the invention, the training data is used to train the KI / ML function. In another aspect of the invention, KI / ML data is generated from the training data. In an advantageous embodiment of the invention, the KI / ML function and / or the KI / ML data are used to determine the stress state.

[0071] The bulk acoustic signals of a chain in operation are typically very significant. Pattern recognition and / or machine learning methods, such as neural networks (models), can be advantageously used. For this purpose, bulk acoustic signals of unworn (e.g., initial state) and stressed chains are collected, associated with their respective machines and / or machine types, in relevant operating states and / or phases of motion.

[0072] Specific stress states can also be learned, which requires learning the bulk acoustic signals of the corresponding worn chains with known stress states. The stress state can be determined by variations in parameters such as chain elongation, clearance, chain plates, pins, bushings, rollers, etc. This can also be for individual links and / or chain components, as the weakest link can cause chain breakage.

[0073] In another aspect of the invention, the evaluation unit includes a processor designed and adapted to perform software programs of the method of the invention.

[0074] This task is also solved by the software program of the method of the present invention.

[0075] The software program has program code that can be stored on a machine-readable carrier or storage medium such as semiconductor memory, hard disk memory or optical memory, and is used to execute, implement and / or control method steps according to one of the above embodiments, particularly when the program product or program is running on a computer or device.

[0076] This task is also addressed by methods for analyzing and / or monitoring chain elongation. Advantageous embodiments of the invention are also set forth in the dependent claims.

[0077] The method of the present invention for analyzing and / or monitoring chain elongation has three steps: the first step is to collect body acoustic signals from the chain during operation. The body acoustic signals can be read using a microphone.

[0078] The second step is to analyze the acquired bulk acoustic signals. For this, a Fast Fourier Transform (FFT) can be used to convert the bulk acoustic signals from a time-based representation to a frequency-based representation. The bulk acoustic signals can also be smoothed and squared.

[0079] The third step is to determine the chain elongation. This is done by utilizing the frequency distribution and sound pressure levels of the bulk acoustic signal. One metric is the kurtosis of the frequency distribution. By determining the kurtosis and sound pressure levels of the bulk acoustic signal frequency distribution, the chain elongation is determined.

[0080] Chains are subjected to stress during operation, naturally leading to wear. Because chains are critical components, they are often prematurely maintained or replaced as a preventative measure, resulting in unnecessary costs. Delayed maintenance or replacement of the chain carries the risk of chain breakage under load. Within a predictive maintenance framework, understanding the chain's stress state is important, such as the duration of operation and the load conditions under which it has been running. However, since the chain and its associated drive components (sprockets, steering elements, etc.) form a complex system, directly capturing the chain's stress state is meaningful.

[0081] One indicator of chain wear is its elongation. Even slight chain elongation can cause production errors in processes synchronized via chain drives, requiring manual readjustment.

[0082] The method of this invention enables continuous acquisition of chain elongation by determining the elongation amount. When the chain elongation reaches or exceeds a set threshold, appropriate measures (chain maintenance, replacement) can be taken in a timely manner to prevent chain failure. This also saves costs, as the chain is only maintained or replaced when necessary.

[0083] In one improvement of the invention, the chain elongation is qualitatively analyzed. Quantitative analysis of the chain elongation is not necessary. For timely chain maintenance, it is only necessary to know whether the chain elongation is sufficiently small. This saves costs.

[0084] In another embodiment of the invention, maintenance reminders are derived and / or output from qualitative analysis. In another design of the invention, the maintenance reminders include information about the next lubrication and / or replacement of a portion or the entire chain. The reminders may also include warnings if elongation reaches a critical value. These reminders can advantageously improve the plannability of maintenance and / or possible chain replacement, or reduce wear.

[0085] In another aspect of the invention, frequency analysis of the bulk acoustic signal is performed to determine the chain stress state. For this purpose, the frequency distribution of the bulk acoustic signal and its sound pressure level are utilized in particular. One metric is the kurtosis (sharpness) of the frequency distribution. By determining the kurtosis of the frequency distribution of the bulk acoustic signal and its sound pressure level, the chain lubrication state and elongation are determined.

[0086] In another aspect of this invention, the chain elongation is determined from a defined body acoustic signal frequency. Chain elongation results in a larger frequency distribution of the body acoustic signal frequency. The chain elongation is determined by determining the kurtosis of the frequency distribution of the body acoustic signal frequency.

[0087] In another design of the invention, the chain lubrication condition is determined from the shape and / or intensity of the determined bulk acoustic signal. Specifically, the frequency distribution area of ​​the bulk acoustic signal frequency and its sound pressure level are measures of the chain lubrication condition. The worse the chain lubrication condition, the higher the sound pressure level, and the larger the frequency distribution of the bulk acoustic signal frequency during chain operation.

[0088] In an advantageous design of the invention, training data is generated from the analyzed body acoustic data. In an improvement of the invention, the training data is used to train the KI / ML function. In another aspect of the invention, KI / ML data is generated from the training data. In an advantageous embodiment of the invention, the KI / ML function and / or the KI / ML data are used to determine elongation.

[0089] The bulk acoustic signals of a chain in operation are typically very significant. Pattern recognition and / or machine learning methods, such as neural networks (models), are advantageously utilized. For this purpose, bulk acoustic signals from unworn (e.g., initial state) and stressed chains are collected, correlated with their respective associated machines and / or machine types, in relevant operating states and / or phases of motion.

[0090] Specific elongation can also be learned, which requires learning the bulk acoustic signals of the corresponding worn chains with known stress states. The stress state can be determined by variations in parameters such as chain elongation, clearance, chain plates, pins, bushings, rollers, etc. This can also be for individual links and / or chain components, as the weakest link can cause chain breakage.

[0091] In another design of the invention, the initial state of the chain is determined to ascertain the amount of elongation. The initial state of the chain is specifically its new state. The monitored chain in its initial state exhibits optimal lubrication and no elongation.

[0092] In another design of the invention, the deviation between the actual state and the initial state of the chain is determined to determine the amount of elongation. The deviation between the actual state and the initial state is a measure of the chain elongation.

[0093] In another training exercise of this invention, KI / ML data analysis is used to determine the deviation between the actual state and the initial state of the chain. Alternatively, bulk acoustic signals can be used to learn the deviation. In this way, the determination of the deviation can be continuously improved.

[0094] In another embodiment of the invention, the method is placed in an initial state after the reset function is triggered. The method of the present invention can therefore be restarted at any time, particularly after lubrication, maintenance, or chain replacement, when the chain is in its initial state.

[0095] This task is also addressed by a chain elongation monitoring device for analyzing and / or monitoring chain elongation. Advantageous embodiments of the invention are also set forth in the dependent claims.

[0096] The chain elongation monitoring device of the present invention is used to analyze and / or monitor chain elongation, and includes a bulk acoustic sensor and an evaluation unit. The bulk acoustic sensor can collect bulk acoustic signals from the chain during operation. The evaluation unit is designed and adapted to analyze the chain elongation from the bulk acoustic signals collected by the bulk acoustic sensor.

[0097] In another embodiment of the invention, the evaluation unit has a first memory and / or storage area for storing the acquired bulk acoustic signals. Optionally, bulk acoustic signals of the initial state of the chain may also be stored.

[0098] In another design of the invention, the evaluation unit has a second memory and / or storage area for storing first KI / ML data, which is designed and adapted to analyze the lubrication status of the chain from the acquired bulk acoustic signals.

[0099] In another training exercise of the present invention, the evaluation unit has a third memory and / or storage area in which second KI / ML data is stored, the second KI / ML data being designed and adapted to analyze the chain elongation from the acquired stereoscopic signals.

[0100] In one improvement of the invention, the evaluation unit has a fourth memory and / or storage area, wherein the result data of the volume acoustic signal frequency analysis is stored.

[0101] In another design of the invention, the evaluation unit has a fifth memory and / or storage area, wherein first training data for training a first KI / ML function is stored, the first KI / ML function being designed and adapted to determine the lubrication state of the chain from the first KI / ML data.

[0102] In one improvement of the invention, the evaluation unit has a sixth memory and / or storage area, wherein second training data for training a second KI / ML function is stored, the second KI / ML function being designed and adapted to determine the amount of chain elongation from the second KI / ML data.

[0103] In an advantageous design of the invention, training data is generated from the analyzed bulk acoustic data. In an improvement of the invention, the training data is used to train the KI / ML function. In another aspect of the invention, KI / ML data is generated from the training data. In an advantageous embodiment of the invention, the KI / ML function and / or the KI / ML data are used to determine the stress state.

[0104] The bulk acoustic signals of a chain in operation are typically very significant. Pattern recognition and / or machine learning methods, such as neural networks (models), can be advantageously used. For this purpose, bulk acoustic signals of unworn (e.g., initial state) and stressed chains are collected, associated with their respective machines and / or machine types, in relevant operating states and / or phases of motion.

[0105] Specific stress states can also be learned, which requires learning the bulk acoustic signals of the corresponding worn chains with known stress states. The stress state can be determined by variations in parameters such as chain elongation, clearance, chain plates, pins, bushings, rollers, etc. This can also be for individual links and / or chain components, as the weakest link can cause chain breakage.

[0106] In another aspect of the invention, the evaluation unit includes a processor designed and adapted to perform software programs of the method of the invention.

[0107] This task is also solved by the software program of the method of the present invention.

[0108] Embodiments of the method for determining the chain running state and the chain running state analysis device of the present invention are shown in the accompanying drawings in a simplified schematic manner, and are described in detail in the following description.

[0109] It is shown that:

[0110] Figure 1: Chain operation status analysis device of the present invention, including a memory.

[0111] Figure 2: Chain operation status analysis device of the present invention, with six memory units.

[0112] Figure 3: Spectrum of the acquired bulk acoustic signal used to determine elongation

[0113] Figure 4: Spectrum of the acquired bulk acoustic signal used to determine lubrication status

[0114] Figure 5a: Spectrum of the acquired bulk acoustic signal used to determine elongation, with the chain in its initial state.

[0115] Figure 5b: Spectrum of the acquired bulk acoustic signal used to determine the elongation, where the chain elongation is sufficiently small.

[0116] Figure 5c: Spectrum of the acquired bulk acoustic signal used to determine the elongation, where the chain elongation is no longer sufficiently small.

[0117] Figure 6a: Spectrum of the collected bulk acoustic signal used to determine the lubrication state, with the chain in its initial state.

[0118] Figure 6b: Spectrum of the acquired bulk acoustic signal used to determine the lubrication condition; the chain lubrication condition is sufficient.

[0119] Figure 6c: Spectrum of the acquired bulk acoustic signal used to determine the lubrication status; the chain lubrication status is no longer sufficient.

[0120] Figure 1 schematically illustrates the structure of the chain running status analysis device of the present invention. The chain running status analysis device has a bulk acoustic sensor S, which is designed as a piezoelectric vibration sensor. The bulk acoustic sensor S is arranged such that the monitored chain moves relative to the bulk acoustic sensor S during operation. The bulk acoustic sensor S is preferably fixedly arranged near a sprocket, steering wheel, or steering roller (less than 1 meter). The bulk acoustic sensor S is connected to an evaluation unit C via an interface. The evaluation unit C has a processor connected to a memory containing software programs for analyzing the chain running status. The evaluation unit C is preferably designed as a mobile device (laptop, tablet, smartphone, etc.) and can therefore be placed at a distance from the bulk acoustic sensor S. Alternatively, the evaluation unit can be part of a network or integrated into the chain running status analysis device. The evaluation unit C is connected to a memory M via another interface.

[0121] To perform chain stress state analysis, a bulk acoustic sensor S collects bulk acoustic signals from the running chain. The chain is in motion at this time. The bulk acoustic sensor S continuously collects bulk acoustic signals. The collected signals are converted into digital data on the evaluation unit C and analyzed through frequency analysis. The data from the collected chain bulk acoustic signals are used for qualitative analysis of lubrication status and elongation.

[0122] To this end, the acquired bulk acoustic signals are compared with those of the same monitored chain and / or similar chains in their initial state. Alternatively, training data can be used for this purpose. The bulk acoustic signals of the monitored chain in its initial state are acquired by placing the monitored chain in the new state on the equipment on which the monitored chain operates. The operating parameters (speed, force, direction of travel, etc.) of the monitored chain in its initial state are also consistent with and / or within a comparable or similar range to the operating parameters of the chain in operation. The monitored chain in its new state is particularly well-lubricated and has no elongation. The bulk acoustic signals of the monitored chain in its initial state are stored in memory M. If the analysis of the monitored chain's bulk acoustic signals shows that the lubrication condition reaches or exceeds the threshold stored in memory M, the evaluation unit C outputs a maintenance prompt, specifically a chain lubrication prompt. If the analysis of the monitored chain's bulk acoustic signals shows that the elongation reaches or exceeds the threshold also stored in memory M, the evaluation unit C outputs a prompt to replace part or the entire chain.

[0123] Another embodiment of the chain operation status analysis device of the present invention is shown in Figure 2. This chain operation status analysis device has a bulk acoustic sensor S, which is connected to an evaluation unit C via an interface. The evaluation unit C is connected to six memory devices M1, M2, M3, M4, M5, and M6 via an interface. The memory devices M1, M2, M3, M4, M5, and M6 are different from each other and arranged independently. The memory devices M1, M2, M3, M4, M5, and M6 can also be different storage areas on mass storage and / or cloud storage.

[0124] The first memory M1 stores the bulk acoustic signals of the monitored chain during operation, acquired by the bulk acoustic sensor S. Optionally, it also stores the bulk acoustic signals of the monitored chain in its initial state, for comparing the bulk acoustic signals during operation and in the initial state. The second memory M2 stores first KI / ML data. The first KI / ML data is designed and adapted to analyze the chain's lubrication state from the acquired bulk acoustic signals. The third memory M3 stores second KI / ML data. The second KI / ML data is designed and adapted to analyze the chain's elongation from the acquired bulk acoustic signals. The fourth memory M4 stores the result data of the bulk acoustic signal frequency analysis. The fifth memory M5 stores first training data for training the first KI / ML function. The first KI / ML function is designed and adapted to determine the chain's lubrication state from the first KI / ML data. The sixth memory M6 stores second training data for training the second KI / ML function. The second KI / ML function is designed and adapted to determine the chain's elongation from the second KI / ML data.

[0125] The bulk acoustic signals of a running chain are typically very significant. Pattern recognition and / or machine learning methods, such as neural networks (models), are advantageously employed. For this purpose, bulk acoustic signals from unworn (e.g., initial state) and stressed chains are collected, associated with their respective machines and / or machine types, in relevant operating states and / or motion phases, stored in a fifth memory M5 and a sixth memory M6, and the model is trained. The first and second training data are continuously supplemented and updated by collecting bulk acoustic signals from the running chain, and the first and second KI / ML functions qualitatively determine the chain's lubrication status and elongation.

[0126] Specific stress states can also be learned, which requires learning the bulk acoustic signals of the corresponding worn chains with known stress states. The stress state can be determined by variations in parameters such as chain elongation, clearance, chain plates, pins, bushings, rollers, etc. This can also be for individual links and / or chain components, as the weakest link can cause chain breakage.

[0127] Figures 3 and 4 exemplarily illustrate the spectrum of the acquired bulk acoustic signal within a relevant frequency range for analyzing the chain's lubrication status (Figure 3) and elongation (Figure 4). To analyze the bulk acoustic signal acquired by the bulk acoustic sensor S, the time-based acoustic signal is converted to a frequency range using a Fast Fourier Transform (FFT) to obtain a frequency signal. The frequency signal may optionally be filtered. The evaluation unit C has suitable software for this purpose, which runs on a processor. In another step, the filtered frequency signal is averaged to obtain an average frequency signal. For example, a squared average calculation is performed for this purpose.

[0128] Evaluation unit C determines the kurtosis using the average frequency signal and compares the determined kurtosis with a kurtosis threshold stored in evaluation unit C. If the determined kurtosis exceeds the kurtosis threshold, evaluation unit C outputs a chain maintenance prompt.

[0129] Figure 5 shows an embodiment of the bulk acoustic signal spectrum used to determine the elongation of the monitored chain. The threshold S of the bulk acoustic signal frequency is marked. f Less than the threshold S f The frequency range indicates the extent to which the elongation of the monitored chain reaches a point where the chain or a section of the chain must be replaced.

[0130] The stereophonic signal spectrum of the monitored chain in its initial state (Figure 5a) shows a steep peak distribution of the stereophonic signal frequency. The links of the monitored chain have uniform small elongations.

[0131] The spectrum of the bulk acoustic signal of the monitored chain under stress (Figure 5b) shows the normal peak distribution of the bulk acoustic signal frequency. The bulk acoustic signal frequency reaches the threshold S. f However, it is not lower than that. The links of the monitored chain have an elongation distribution, but it has not reached the point where the monitored chain must be maintained or replaced.

[0132] The bulk acoustic signal spectrum of the monitored chain under excessive stress (Figure 5c) shows a flat peak distribution of the bulk acoustic signal frequency. The chain links exhibit such an elongation distribution that the bulk acoustic signal frequencies of some links are below the threshold S. f Evaluation unit C outputs a prompt to replace part or the entire chain.

[0133] Figure 6 shows an embodiment of a bulk acoustic signal spectrum used to determine the lubrication status of a monitored chain. The threshold ranges S for the bulk acoustic signal frequency and sound pressure level are marked. k Threshold range S k It indicates the range of frequency and sound pressure, within which the lubrication of the monitored chain is sufficient for the operation of the monitored chain.

[0134] The stereophonic signal spectrum of the monitored chain in its initial state (Figure 6a) shows a steep peak distribution of the stereophonic signal frequency and sound pressure. The links of the monitored chain have a uniformly high lubrication state.

[0135] The spectrum of the bulk acoustic signal of the monitored chain under stress (Figure 6b) shows the normal peak distribution of the frequency and sound pressure of the bulk acoustic signal. The frequency and sound pressure of the bulk acoustic signal reach the threshold range S. k However, it did not exceed that level. The links of the monitored chain showed a lubrication status distribution, but not to the point where maintenance or lubrication of the monitored chain was necessary.

[0136] The bulk acoustic signal spectrum of the monitored chain under excessive stress (Figure 6c) shows a flat peak distribution of bulk acoustic signal frequency and sound pressure level. The chain links exhibit such a lubrication distribution that the bulk acoustic signal frequency and sound pressure level of some links exceed the threshold range S. k Evaluation unit C outputs a prompt to replace part or the entire chain.

[0137] List of reference numerals

[0138] C Assessment Unit

[0139] Memory types M, M1, M2, M3, M4, M5, and M6

[0140] S-type acoustic sensor

[0141] S f Frequency threshold

[0142] S k Kurtosis threshold

Claims

1. A method for analyzing and / or monitoring the stress state of a chain, comprising the following steps: • acquiring bulk acoustic signals from chain operation, • analyzing the acquired bulk acoustic signals, and • determining the lubrication state and elongation of the chain.

2. The method for analyzing and / or monitoring the stress state of a chain according to claim 1, characterized in that, Qualitative analysis was performed on the lubrication status and / or elongation of the chain.

3. The method for analyzing and / or monitoring the stress state of a chain according to claim 2, characterized in that, The qualitative analysis yields and / or outputs maintenance tips.

4. The method for analyzing and / or monitoring the stress state of a chain according to claim 3, characterized in that, Maintenance tips include information about the next lubrication and / or replacement of parts and / or the entire chain.

5. A method for analyzing and / or monitoring the stress state of a chain according to one or more of the preceding claims, characterized in that, To determine the stress state of the chain, frequency analysis was performed on the bulk acoustic signal.

6. The method for analyzing and / or monitoring the stress state of a chain according to claim 5, characterized in that, The chain elongation is determined from the defined frequency of the body acoustic signal.

7. The method for analyzing and / or monitoring the stress state of a chain according to claim 5 or 6, characterized in that, The lubrication status of the chain is determined from the shape and / or intensity of the determined bulk acoustic signal.

8. A method for analyzing and / or monitoring the stress state of a chain according to one or more of the preceding claims, characterized in that, Analysis of body sound data using frequency analysis.

9. A method for analyzing and / or monitoring the stress state of a chain according to one or more of the preceding claims, characterized in that, Training data is generated from the analyzed body sound data.

10. The method for analyzing and / or monitoring the stress state of a chain according to claim 9, characterized in that, The training data is used to train the KI / ML functions.

11. The method for analyzing and / or monitoring the stress state of a chain according to claim 10, characterized in that, Generate KI / ML data from training data.

12. The method for analyzing and / or monitoring the stress state of a chain according to any one or more of claims 9 to 11, characterized in that, KI / ML functionality and / or KI / ML data are used to determine the stress state.

13. A method for analyzing and / or monitoring the stress state of a chain according to one or more of the preceding claims, characterized in that, The initial state of the chain is determined in order to determine the stress state.

14. The method for analyzing and / or monitoring the stress state of a chain according to claim 13, characterized in that, To determine the stress state, the deviation between the actual state and the initial state of the chain is determined.

15. The method for analyzing and / or monitoring the stress state of a chain according to claim 14, characterized in that, KI / ML data analysis was used to analyze the deviation between the actual state and the initial state of the chain.

16. A method for analyzing and / or monitoring the stress state of a chain according to one or more of the preceding claims, characterized in that, After the reset function is triggered, the method is placed in its initial state.

17. A chain running condition analysis device for analyzing and / or monitoring chain lubrication condition and / or elongation, comprising: • a bulk acoustic sensor (S), • an evaluation unit (C) wherein the evaluation unit (C) is designed and adapted to analyze the chain elongation and lubrication condition from the bulk acoustic signals acquired by the bulk acoustic sensor (S).

18. The chain running condition analysis device according to claim 17 for analyzing and / or monitoring chain lubrication status and / or elongation, characterized in that, The evaluation unit (C) has a first memory (M1) and / or storage area in which the acquired stereo signals are stored.

19. The chain running condition analysis device according to claim 17 or 18 for analyzing and / or monitoring chain lubrication status and / or elongation, characterized in that, The evaluation unit (C) has a second memory (M2) and / or storage area in which first KI / ML data is stored, the first KI / ML data being designed and suitable for analyzing the lubrication status of the chain from the acquired bulk acoustic signals.

20. A chain running condition analysis device for analyzing and / or monitoring chain lubrication status and / or elongation according to any one or more of claims 17 to 19, characterized in that, The evaluation unit (C) has a third memory (M3) and / or storage area in which second KI / ML data is stored, which is designed and suitable for analyzing the chain elongation from the acquired bulk acoustic signals.

21. A chain running condition analysis device for analyzing and / or monitoring chain lubrication status and / or elongation according to one or more of claims 17 to 20, characterized in that, The evaluation unit (C) has a fourth memory (M4) and / or storage area in which the result data of the volume acoustic signal frequency analysis is stored.

22. A chain running condition analysis device for analyzing and / or monitoring chain lubrication status and / or elongation according to any one or more of claims 17 to 21, characterized in that, The evaluation unit (C) has a fifth memory (M5) and / or storage area where first training data for training the first KI / ML function is stored, the first KI / ML function being designed and adapted to determine the lubrication state of the chain from the first KI / ML data.

23. A chain running condition analysis device for analyzing and / or monitoring chain lubrication status and / or elongation according to any one or more of claims 17 to 22, characterized in that, The evaluation unit (C) has a sixth memory (M6) and / or storage area where second training data for training a second KI / ML function is stored, the second KI / ML function being designed and adapted to determine the amount of chain elongation from the second KI / ML data.

24. A chain running condition analysis device for analyzing and / or monitoring chain lubrication status and / or elongation according to any one or more of claims 17 to 23, characterized in that, The evaluation unit (C) includes a processor designed and adapted to execute a software program of the method of the present invention according to claims 1 to 16.

25. A software program for performing the method of the present invention according to any one or more of claims 1 to 16.

Images