Determine the tightening category of the tightening operation performed by the tightening tool

Abstract

A method for determining a tightening category of a tightening operation performed by a tightening tool (10) is provided. The method comprises obtaining (S101) a time domain representation of a set of observed torque values ​​and angle values ​​for a fastener (20) that has been tightened by the tightening tool (10), obtaining (S102) a frequency domain representation of the set of observed torque values, associating (S103) a tightening category with each obtained time domain and frequency domain representation of the identified tightening operation, training (S104) a machine learning model with the obtained time domain and frequency domain representations and the tightening category and providing (S105) to the trained machine learning model further obtained time domain and frequency domain representations for the fastener that has been tightened by the tightening tool (10), wherein the trained machine learning model outputs an estimated tightening category for the provided set of observed torque values ​​and angle values.

Description

Technical Field

[0001] The invention relates to a method for determining a tightening class of a tightening operation performed by a tightening tool and to a control device for carrying out the method. Background Art

[0002] In the process of tightening a fastener such as a bolt or a screw using a tightening tool, various undesirable tightening results may occur to the tightened bolt or screw.

[0003] Analyzing the sensor data of a tightening tool provides valuable insights into tightening results. Analytical models can be developed to determine tightening results based on the sensor data of a tightening tool.

[0004] However, whether the tightening result is correct may be difficult to determine by a human operator or a machine, and it is more tedious and time-consuming. Summary of the invention

[0005] An object is to solve or at least alleviate this problem in the art and thereby provide an improved method of determining a tightening result of a tightening tool.

[0006] In a first aspect, the object is achieved by a method for determining a tightening category of a tightening operation performed by a tightening tool. The method comprises obtaining a time domain representation of a set of observed torque values ​​and angle values ​​for a fastener that has been tightened by the tightening tool, obtaining a frequency domain representation of a set of observed torque values ​​for the fastener that has been tightened by the tightening tool, associating a tightening category with each obtained time domain representation of the set of observed torque values ​​and angle values ​​and each obtained frequency domain representation of the set of observed torque values, the tightening category identifying the type of tightening operation that has been applied to the fastener, training a machine learning model with the obtained time domain representation of the set of observed torque values ​​and angle values, the obtained frequency domain representation of the set of observed torque values, and the tightening category associated with each set, and providing the trained machine learning model with further obtained time domain representations of the set of observed torque values ​​and angle values ​​for the fastener that has been tightened by the tightening tool and frequency domain representations of further sets of observed torque values, wherein the trained machine learning model outputs an estimated tightening category for the provided set of observed torque values ​​and angle values.

[0007] In a second aspect, the object is achieved by a control device configured to determine a tightening category of a tightening operation performed by a tightening tool, the control device comprising a processing unit and a memory, the memory comprising instructions executable by the processing unit, whereby the control device operates to obtain a time domain representation of a set of observed torque values ​​and angle values ​​for a fastener that has been tightened by the tightening tool, obtain a frequency domain representation of a set of observed torque values ​​for the fastener that has been tightened by the tightening tool, associate the tightening category with each acquired time domain representation of the set of observed torque values ​​and angle values ​​and each acquired frequency domain representation of the set of observed torque values In association, the tightening category identifies the type of tightening operation that has been applied to the fastener, a machine learning model is trained using a time domain representation of a set of acquired observed torque values ​​and angle values, a frequency domain representation of a set of acquired observed torque values, and a tightening category associated with each set, and the trained machine learning model is provided with a further acquired time domain representation of a set of observed torque values ​​and angle values ​​for the fastener that has been tightened by the tightening tool and a frequency domain representation of a further set of observed torque values, wherein the trained machine learning model outputs an estimated tightening category for the provided set of observed torque values ​​and angle values.

[0008] Therefore, in order to train the machine learning model, in addition to obtaining the time domain representation of the torque value and the angle value for the fastener that has been tightened by the tightening tool, the frequency domain representation of the torque value is also obtained. Advantageously, the frequency domain representation of the torque value helps to observe the characteristics of the torque data, which are usually not visible in the corresponding time domain representation.

[0009] In an embodiment, obtaining a frequency domain representation of a set of observed torque values ​​for fasteners that have been tightened by a tightening tool further includes obtaining a frequency domain representation of a set of observed angle values ​​for the fasteners that have been tightened; and associating further includes associating a tightening category with each obtained frequency domain representation of the set of observed angle values; training the machine learning model further includes training the machine learning model with the obtained frequency domain representation of the set of observed angle values; and wherein providing further includes providing the trained machine learning model with further obtained frequency domain representations of further sets of observed angle values ​​for fasteners that have been tightened by the tightening tool.

[0010] In an embodiment, frequency domain representations of the observed torque values ​​and the observed angle values ​​are obtained by performing a Fast Fourier Transform (FFT) on the obtained time domain representations of the observed torque values ​​and the observed angle values.

[0011] In an embodiment, the machine learning is based on one or more of neural networks, random forest-based classification, and regression analysis.

[0012] In an embodiment, a warning of the estimated tightening category is provided, for example, to an operator of the tightening tool, the tightening tool itself, a monitoring room, or a remote cloud function.

[0013] In an embodiment, the tightening tool provides an audible and / or visual warning to an operator of the tool.

[0014] In an embodiment, the tightening category is selected from a group of tightening operations that have been applied to fasteners, the tightening operations representing breakaway, socket slip, stick-slip, thread locking, and high screw-in torque. Other categories are contemplated.

[0015] In a third aspect, there is provided a computer program comprising computer executable instructions which, when executed on a processing unit included in a tightening tool, cause the tightening tool to perform the steps of the method of the first aspect.

[0016] In a fourth aspect, there is provided a computer program product comprising a computer readable medium having the computer program of the third aspect embodied on the computer readable medium.

[0017] As will be appreciated, the control device may be operable to perform the method according to any one of the above described embodiments.Furthermore, in one aspect, a tightening tool comprising a control device is provided.

[0018] Generally, all terms used in the claims are interpreted according to their ordinary meaning in the technical field, unless otherwise explicitly defined herein. Unless otherwise explicitly stated, all references to "a / an / the element, device, component, means, step, etc." are openly interpreted as referring to at least one example of the element, device, component, means, step, etc. Unless explicitly stated otherwise, the steps of any method disclosed herein do not have to be performed in the exact order disclosed. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Aspects and embodiments will now be described by way of example with reference to the accompanying drawings, in which:

[0020] Figure 1 A tightening tool configured to apply torque to a fastener is shown, wherein tool embodiments may be implemented;

[0021] Figure 2a to Figure 2d shows different types of undesirable tightening operations that may be performed by a tool during a tightening procedure;

[0022] Figure 3 shows training of a machine learning model according to an embodiment;

[0023] Figure 4 Utilizing a trained ML model in an implementation scheme is shown;

[0024] Figure 5 A flow chart illustrating a method for determining a tightening category of a tightening operation performed by a tightening tool in an embodiment is shown;

[0025] Figure 6a shows the frequency domain representation of the torque during a stick-slip tightening operation;

[0026] Figure 6b shows a frequency domain representation of the torque during a normal tightening operation;

[0027] Figure 7 A Fast Fourier Transform (FFT) is shown being performed. DETAILED DESCRIPTION

[0028] Aspects of the present invention will now be described more fully hereinafter with reference to the accompanying drawings, in which certain embodiments of the invention are shown.

[0029] However, these aspects can be implemented in many different forms and should not be interpreted as limiting; on the contrary, these embodiments are provided by way of example so that the present invention will be thorough and complete, and the scope of all aspects of the present invention will be fully conveyed to those skilled in the art. Throughout the description, the same reference numerals refer to the same elements.

[0030] Figure 1 A tool in the form of a tightening tool 10 configured to apply torque to a fastener, such as a bolt 20 , is shown in which tool embodiments may be implemented.

[0031] The tightening tool 10 may be cordless or powered by a wire, and has a body 11 and a tool head 12. The tool head 12 has an output shaft having a socket 13 configured to be rotatably driven by a motor disposed inside the body 11 to apply torque to the bolt 20.

[0032] The tightening tool 10 may be arranged with a display device 14 , by means of which information relating to the operation of the tool 10 can be presented to an operator of the tool 10 , and an interface 23 , by means of which the operator can input data into the tool 10 .

[0033] The tightening tool 10 may further be arranged with communication capabilities in the form of a radio transmitter 15 to wirelessly transmit operating data (eg applied torque) to a remotely located device (eg a cloud server 30 ).

[0034] The steps of the method performed by the tool 10 to be described below are actually performed by the control device 21, which includes a processing unit 16 implemented in the form of one or more microprocessors, which are arranged to execute a computer program 17 downloaded to a storage medium 18 associated with the microprocessor (e.g., a random access memory (RAM), a flash memory or a hard drive). The processing unit 16 is arranged to cause the tool 10 to perform a method according to an embodiment when a suitable computer program 17 including computer executable instructions is downloaded to the storage medium 18 and executed by the processing unit 16. The storage medium 18 can also be a computer program product including the computer program 17. Alternatively, the computer program 17 can be transferred to the storage medium 18 via a suitable computer program product (e.g., a digital versatile disc (DVD) or a memory stick). As a further alternative, the computer program 17 can be downloaded to the storage medium 18 via a network. Alternatively, the processing unit 16 can be implemented in the form of a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), a complex programmable logic device (CPLD), etc.

[0035] As will be appreciated, the control device 21 may be arranged inside the tightening tool 10 or connected to the tool 10 , for example as a control device 21 attached to the outside of the body 11 of the tool 10 .

[0036] In addition, the tool 10 includes a first built-in sensor 19 capable of measuring the torque applied to the bolt 20 and a second built-in sensor 22 capable of measuring the rotation angle of the drive shaft of the tightening tool 10. The torque sensor 19 and the angle sensor 22 communicate with the processing unit 16 so that the processing unit 16 can obtain the measured torque value and angle value. Alternatively, the tightening tool 10 may include only an angle sensor (without a torque sensor), wherein the torque value can be calculated based on the sensed rotation angle value.

[0037] Now, when an operator uses the tightening tool 10 to tighten a fastener such as a bolt 20, it is important for the tightened bolt to perform the tightening operation correctly to maintain its tightening durability. Otherwise, there is a risk of tightening deterioration (which may cause the bolt 20 to unscrew in the worst case). Therefore, it is crucial to tighten the bolt 20 correctly, and if it is not tightened correctly, it is desirable to obtain a corresponding indication so that the operator can use the tool 10 to correctly retighten the bolt 20.

[0038] One such tightening operation performed by the tool 10 (and its human operator) is known as a breakaway. During a breakaway, the torque value suddenly drops to zero before the tightening procedure being performed is completed.

[0039] For example, if the torque value suddenly drops to zero before tightening is complete, disengagement between the output shaft and socket 13 and the bolt 20 may have occurred. This may be caused by operator behavior, worn tool parts and / or fasteners, incorrect tool programming, etc.

[0040] Figure 2a to Figure 2d A number of undesirable tightening operations are shown being performed by the tool 10. The combination of torque values ​​and angle values ​​of a tightening procedure being performed is sometimes referred to as a trajectory.

[0041] Figure 2a A specific type of tightening operation, commonly referred to as socket slip, is shown, which may occur when torque values ​​increase and decrease at periodic angular intervals.

[0042] Figure 2b Another specific type of tightening operation, commonly referred to as stick-slip, is shown, which may occur when torque values ​​fluctuate at a high frequency.

[0043] Figure 2c Another specific type of tightening operation is shown; the so-called thread locking, which may occur when the applied torque reaches the target torque immediately after the rundown phase. This is usually caused by a sudden locking of the threads of the tightened bolt 20, for example because the bolt 20 enters the thread at an angle or if the size is incorrect.

[0044] at last, Figure 2d A further specific type of tightening operation is shown, which is known as high screw-in torque, which may occur when the applied torque reaches the target torque during the screw-in phase. This is usually caused by higher than normal friction in the threads of the bolt 20, for example because the bolt 20 has an incorrect size, the wrong thread coating or the bolt 20 tilts when entering the thread.

[0045] As understood, these are examples of incorrect tightening operations performed by the tool 10 and should preferably be avoided. However, if they still occur, it is desirable to notify the operator or some supervisory function thereof.

[0046] Therefore, in an embodiment, a trained machine learning (ML) model is used to determine whether an incorrect tightening operation has been performed based on the acquired torque values ​​and angle values, and if so, to determine which specific type of tightening operation has been performed, i.e., to determine whether it is one of the above-mentioned "disengagement", "socket slippage", "stick-slip", "thread locking" and "high screw-in torque" (or any other identified tightening operation).

[0047] Figure 3 shows the training of the ML model according to the embodiment, and Figure 4Utilizing a trained ML model in an implementation is shown. Further references Figure 5 , which shows a flowchart illustrating a method for determining a tightening category of a tightening operation performed by a tightening tool 10 according to an embodiment.

[0048] Therefore, during the training phase, in step S101 , an observed torque value T t for a bolt 20 that has been tightened by the tightening tool 10 is acquired. TD and angle value A TD Therefore, these are the time domain (TD) representations of the previous references. Figure 2a to Figure 2d The graph shows the regular time domain torque values ​​and angle values.

[0049] For example, each set may include, for example, one hundred torque values ​​and corresponding angle values ​​acquired by the processing unit 16 from the torque sensor 19 and the angle sensor 22 when the tightening tool 10 performs a tightening procedure.

[0050] Any acquired torque and angle values ​​may be stored locally in the memory 18 and / or wirelessly transmitted via the radio transmitter 15 to a remotely located cloud server 30 .

[0051] Furthermore, in step S102, a frequency domain (FD) representation T of a set of torque values ​​for the fastener that has been tightened by the tightening tool 10 is obtained. FD Advantageously, the frequency domain representation of the set of torque values ​​T FD It is helpful to observe the characteristics of torque data, which are usually represented in the time domain corresponding to a set of torque values ​​T TD It is not visible in.

[0052] Compared to the frequency domain representation showing the torque during a normal tightening operation Figure 6b , Figure 6a The frequency domain representation of the torque during a stick-slip tightening operation is shown by two torque peaks. The time domain representation of stick-slip was previously described in Figure 2b Shown in.

[0053] In step S103, a tightening category TC is associated with each set of torque values ​​and angle values ​​obtained in steps S102 and S103. Thus, the tightening category identifies the type of tightening operation that has been applied to the fastener. As previously exemplified, the tightening category may include "breakaway", "socket slip", "stick-slip", "thread locking", and "high screw-in torque", etc., or any other identified tightening operation.

[0054] In an embodiment, the operator of the tool 10 identifies which particular tightening category a set of measurement values ​​will be associated with based on the set of torque values ​​and angle values ​​obtained. For example, if the observed torque value T obtained is TDand angle value A TD The time domain representation has the form presented to the operator via the display device 14. Figure 2b , the operator concludes that the associated tightening category should be “stick-slip” and inputs information identifying the tightening category into the tool 10 via the interface 23 .

[0055] Accordingly, the operator can conclude that the torque value T FD The frequency domain representation of Figure 6a , which emphasizes that the acquired set of observations should indeed be classified as “stick-slip”.

[0056] In step S104 , the identified tightening category together with the acquired time domain and frequency domain representations are provided to the ML model for training.

[0057] This is typically repeated for a large number of observed time domain representations of torque values ​​and angle values ​​and corresponding frequency domain representations of the observed torque values, and thus the ML model will be effectively trained in step S104 to associate a tightening class with each provided set of torque values ​​and angle values.

[0058] Therefore, if Figure 4 As shown, when the tool 10 subsequently performs a tightening procedure, in step S105, the trained ML model is provided with a time domain representation (T TD , A TD ), and a frequency domain representation (T ) of a further set of observed torque values ​​for the fastener that has been tightened by the tool 10 in step S105. FD ).

[0059] As a result, for a further provided set of observed torque values ​​and angle values, the trained ML model will advantageously output an estimated tightening class, for example via the display device 14, so that the operator can determine whether to re-execute the tightening procedure.

[0060] It is also advantageous to also consider the frequency domain representation T of the set of observed torque values FD , the ML model can obtain more information, which provides a more robust and accurate classification compared to considering only the time domain representation of the set of observed torque and angle values.

[0061] In an embodiment, methods for training ML models include neural networks, random forest-based classification, regression analysis, etc. One advantage of neural networks is that they can learn to extract both temporal and spatial patterns from a data set.

[0062] In the implementation scheme, Figure 7As shown, by performing the so-called Fast Fourier Transform (FFT) on the time domain representation of the observed torque value, that is, T FD =FFT(T TD ) to obtain the frequency domain representation of the observed torque value.

[0063] Training the ML model and subsequently determining the tightening category for the torque values ​​and angle values ​​observed during the tightening procedure may be performed locally by the processing unit 16, or by the cloud server 30. The determined tightening category may be immediately communicated to the operator of the tool 10, for example, via the display device 14. Thus, the operator may unscrew the bolt 20 and perform a new tightening procedure to obtain a proper tightening.

[0064] In a further embodiment, in addition to obtaining a frequency domain representation of a set of observed torque values ​​for a fastener that has been tightened by the tightening tool 10 in step S102, a frequency domain representation A of a set of observed angle values ​​is further obtained. FD .

[0065] According to the above reference Figures 3 to 5 In the described embodiment, in step S103, each frequency domain representation A of the set of observed angle values FD will be further associated with the tightening category, so in step S104, the training of the ML model will further consider the frequency domain representation A of the set of observed angle values FD Finally, in step S105, the frequency domain representation A of the set of further observed angle values ​​is provided to the trained ML model. FD , so that the trained ML model outputs the estimated tightening category.

[0066] Likewise, it is advantageous to also consider the frequency domain representation A of the set of observed angle values FD , the ML model can obtain more information, which provides a more robust and accurate classification compared to considering only the time domain representation of the set of observed torque and angle values.

[0067] In an embodiment, by performing FFT on the time domain representation of the obtained observed angle value, that is, A FD =FFT(A TD ) to obtain the frequency domain representation of the observed angle value.

[0068] It is conceivable that steps S101 to S105 of the method according to the embodiment may be performed by a control device 21 arranged in the tool 10 itself. However, it is also conceivable that the torque value and the angle value are measured by the torque sensor 19 and the angle sensor 22 and transmitted via the transmitter 15 to a cloud server 30 equipped with corresponding control devices for performing all steps S101 to S105, which will reduce the computational burden of the processing unit 16. The cloud server 30 may send information representing the estimated tightening category to the tool 15 (i.e., the transmitter 15 may be a transceiver), which presents a warning of the tightening category to the operator in the form of sound via the display device 14 or via a small speaker included in the tool 10.

[0069] It is further envisioned that the tool 10 performs all steps S101 to S105 and then alerts the cloud server 30 (possibly together with an operator alert) of the estimated tightening category, which may maintain a database of estimated tightening categories for many tools, or provides an alert to, for example, a monitoring room.

[0070] Aspects of the invention have mainly been described above with reference to a few embodiments and examples thereof. However, as readily appreciated by a person skilled in the art, other embodiments than the ones disclosed above are equally possible within the scope of the invention as defined by the appended claims of the present application.

[0071] Therefore, although various aspects and embodiments have been disclosed herein, other aspects and embodiments will be apparent to those skilled in the art. The various aspects and embodiments disclosed herein are for illustrative purposes and are not intended to be limiting, with the true scope and spirit being represented by the appended claims.

Claims

1. A method for determining a tightening class of a tightening operation performed by a tightening tool (10), the method comprising: Obtaining a time domain representation of a set of observed torque values ​​and angle values ​​for a fastener (20) that has been tightened by a tightening tool (10); obtaining a frequency domain representation of a set of observed torque values ​​for a fastener that has been tightened by a tightening tool (10); associating a tightening category with a time domain representation of each acquired set of observed torque values ​​and angle values ​​and a frequency domain representation of each acquired set of observed torque values, the tightening category identifying a type of tightening operation that has been applied to the fastener (20); Training a machine learning model with a time domain representation of a set of acquired observed torque values ​​and angle values, a frequency domain representation of a set of acquired observed torque values, and a tightening category associated with each set; and A further acquired time domain representation of a set of observed torque values ​​and angle values ​​for a fastener that has been tightened by a tightening tool (10) and a frequency domain representation of a further set of observed torque values ​​are provided to the trained machine learning model, wherein the trained machine learning model outputs an estimated tightening category for the provided set of observed torque values ​​and angle values.

2. The method according to claim 1, wherein: Obtaining a frequency domain representation of a set of observed torque values ​​for a fastener (20) that has been tightened by a tightening tool (10) further comprises: obtaining a frequency domain representation of a set of observed angle values ​​for a tightened fastener; and associating further comprises: Associating a tightening category with a frequency domain representation of a set of angle values ​​of each acquired observation; training the machine learning model further comprises: Training a machine learning model with a frequency domain representation of a set of acquired observed angle values; and wherein providing further comprises: The trained machine learning model is provided with frequency domain representations of further sets of further acquired observed angle values ​​for fasteners that have been tightened by the tightening tool (10).

3. The method according to claim 1 or 2, wherein: The frequency domain representation of the observed torque value and the observed angle value is obtained by performing a fast Fourier transform (FFT) on the obtained time domain representation of the observed torque value and the observed angle value.

4. The method according to claim 1, wherein: The machine learning is based on one or more of neural networks, random forest-based classification, and regression analysis.

5. The method according to claim 1, further comprising: Provides warning of estimated tightening category.

6. The method according to claim 5, wherein: A warning is provided to an operator of the tightening tool (10), the tightening tool (10) itself, a monitoring room or a remote cloud function (30).

7. The method according to claim 6, wherein: The tightening tool (10) provides an audible and / or visual warning to an operator of the tool (10).

8. The method according to claim 1, wherein: The tightening category is selected from a group of tightening operations that have been applied to the fastener, representing breakout, socket slip, stick-slip, thread locking, and high screw-in torque.

9. A computer program product, comprising a computer-readable medium (18), the computer-readable medium having a computer program (17) implemented on the computer-readable medium, the computer program (17) comprising computer-executable instructions, which, when executed on a processing unit (16) included in a tightening tool (10), cause the tightening tool (10) to perform the steps recited in any one of claims 1 to 8.

10. A control device (21) configured to determine a tightening category of a tightening operation performed by a tightening tool (10), the control device (21) comprising a processing unit (16) and a memory (18), the memory comprising instructions (17) executable by the processing unit (16), whereby the control device (21) operates to: Obtaining a time domain representation of a set of observed torque values ​​and angle values ​​for a fastener (20) that has been tightened by a tightening tool (10); Obtaining a frequency domain representation of a set of observed torque values ​​for a fastener (20) that has been tightened by a tightening tool (10); associating a tightening category with a time domain representation of each acquired set of observed torque values ​​and angle values ​​and a frequency domain representation of each acquired set of observed torque values, the tightening category identifying a type of tightening operation that has been applied to the fastener (20); Training a machine learning model with a time domain representation of a set of acquired observed torque values ​​and angle values, a frequency domain representation of a set of acquired observed torque values, and a tightening category associated with each set; and A further acquired time domain representation of a set of observed torque values ​​and angle values ​​for a fastener that has been tightened by a tightening tool and a frequency domain representation of a further set of observed torque values ​​are provided to the trained machine learning model, wherein the trained machine learning model outputs an estimated tightening category for the provided set of observed torque values ​​and angle values.

11. The control device (21) according to claim 10, wherein: The control device (21) is arranged inside the tightening tool (10) or attached to the outside of the main body of the tightening tool (10), or is a control device remotely located from the tightening tool (10) and communicating therewith.

12. The control device (21) according to any one of claims 10 to 11, further operative to obtain a set of observed torque values ​​from a torque measurement sensor (19) of the tightening tool (10), and to obtain a set of observed angle values ​​from an angle measurement sensor (22) of the tightening tool (10).

13. The control device (21) according to any one of claims 10 to 11, further comprising a communication interface (15), via which data can be sent and / or received by wired or wireless communication.

14. A tightening tool (10) comprising a control device (21) according to any one of claims 10 to 12.

Images