Computer-implemented method for adjusting the center of force or lateral force of a coil spring

Abstract

The invention relates to a computer-implemented method for adjusting the center of force or lateral force of a coil spring (34) during the cold winding process using a cold winding machine (20), having the following steps: - determining the geometry of the coil spring (34) during the cold winding process, - ascertaining, on the basis of the determined geometry, a deviation between the center of force or the lateral force of the shaped coil spring (34) and a specified center of force or a specified lateral force using a set shaping parameter of the cold winding machine (20); and - adapting the set shaping parameter of the cold winding machine during the process of shaping the coil spring if the deviation is greater than a specified threshold.

Description

[0001] Computer-implemented method for adjusting a force center or a transverse force of a coil spring

[0002] Description

[0003] The invention relates to the automatic adjustment of the center of force, i.e. the transverse force, of a coil spring.

[0004] The cold coiling process for coil springs is generally subject to very high tolerances. A deviation from a specified spring geometry can, among other things, lead to a shift in the center of force of the coil spring, causing the coil spring to build up an undesirable transverse force. The shift in the center of force can be due to the starting product, the spring wire, or to the forming process, for example due to wear on the coiling tools. Tools include, in particular, deflection elements, e.g., deflection pulleys for adjusting the diameter of the coil spring or pitch elements for adjusting the pitch of the coil spring. The spring wire can fluctuate in strength and thus lead to an inhomogeneity of the shear modulus. This fluctuating shear modulus results in areas within a wire coil with different springbacks after the coil spring has been formed.

[0005] The shift of the force center can be recognized, for example, by the length of the spring, the measurement of the force penetration points or by geometric variations such as the coil spacing, the (especially average) diameter of the coil spring or the (especially average) pitch of the coil spring.

[0006] The force penetration points can be measured, e.g., in Cartesian coordinates, using a measuring device and reflect the spring's transverse forces. Transverse forces generated by a coil spring that deviate from the target values ​​have a negative impact on ride comfort in the car and are largely responsible for shock absorber wear. The object of the present invention is therefore to create an improved concept for the cold winding of coil springs.

[0007] This problem is solved by the subject matter of the independent patent claims. Further advantageous embodiments are the subject matter of the dependent patent claims.

[0008] Embodiments show a computer-implemented method for adjusting a center of force or a transverse force of a helical spring during cold winding using a cold winding machine. During cold winding, the spring wire for producing the helical spring is usually already tempered and thus already has its later strength. The method initially comprises determining a geometry of the helical spring during cold winding, preferably continuously or at predetermined times during winding. When determining the geometry, coil spacing and / or the (in particular average) diameter of the helical spring and / or the (in particular average) pitch of the helical spring can be determined. In particular, a profile of the diameter and / or a profile of the pitch of the helical spring can be determined.The geometry of the coil spring is determined, for example, by means of one or more lasers and / or by means of a camera system.

[0009] After determining the geometry of the coil spring, a deviation of the force center or shear force of the (fully) formed coil spring from a specified force center or shear force is estimated using a set forming parameter of the cold coiling machine. This means that an estimate is made during the forming process as to whether or not the (fully) formed coil spring will have a force center or shear force within a tolerance.

[0010] If the deviation exceeds a specified threshold, i.e., if the center of force or the transverse force is outside a tolerance, the set forming parameter of the cold coiling machine is adjusted during the forming of the coil spring. By adjusting the forming parameter, the geometry of the formed coil spring can be adjusted so that the center of force or the transverse force of the formed coil spring is below the threshold, i.e., within the tolerance.

[0011] The idea is therefore to reduce the variation in shaping caused by the manufacturing process (e.g., due to tool aging) using an automated process. Due to the uncontrollable fluctuations in material properties, it is all the more important that the shaping process has as little tolerance as possible.

[0012] In exemplary embodiments, the pitch of the final coil is set as a forming parameter. Adjusting the pitch of the final coil offers a simple way to correct the spring's geometry right at the end of the spring forming process in order to adjust the center of force or the transverse force. A further advantage is that at this point, almost the entire coil spring has already been wound, thus allowing the best possible estimate of the deviation of the center of force from a predetermined center of force. The pitch of the coil spring can vary along the length of the coil spring and therefore does not need to be constant.

[0013] In further embodiments, the diameter of the coil spring is set as a forming parameter. A coiling machine deforms the spring wire along two axes. One axis adjusts the spring pitch, the other axis controls the spring diameter. The center of force of the coil spring can be adjusted both along the spring diameter axis and along the pitch axis. Adjusting the diameter of the coil spring thus represents a supplement or alternative to adjusting the pitch of the coil spring's coils. The diameter of the coil spring can vary along the length of the coil spring and therefore does not need to be constant.

[0014] In exemplary embodiments, the deviation of the force center or the transverse force of the formed coil spring is determined using a set forming parameter of the cold coiling machine compared to a predetermined force center or a predetermined transverse force by a model trained by means of machine learning, in particular an artificial intelligence. Using the possibilities of artificial intelligence, a database can be filled with possible pitch and / or diameter values, which are correlated with the force penetration point. For example, in a batch of a production series, each spring can be analyzed with regard to its geometry, i.e. in particular the pitch and radius of the coils, for example based on one or more camera images. Furthermore, the corresponding, determined force center can be assigned to each spring.Thus, using the camera images as input parameters and the center of force as output parameters, a model can be created, particularly for subsequent batches of this production series. This enables supervised learning of the model.

[0015] The model can then be applied to one or more images of a spring from the subsequent batches to determine the predicted center of force during coiling and, in particular, to adapt a final coil to the predicted center of force in order to set or at least approximate a desired center of force. In this case, it is advantageous to create a separate model for each production batch. A more generalized model can be created by using springs from different production batches. It is then possible to use one model for all springs. Using a laser to determine the geometry, a model can be trained accordingly, with the camera images then being replaced by the output data from the measuring laser.

[0016] The applicant, as a manufacturer of coil springs, has an enormous pool of springs whose properties were used to train and validate the artificial intelligence. If the cold coiling machine is equipped with a suitable sensor, for example a measuring laser, in particular one measuring laser per axis to be measured (pitch or diameter), or a camera system, to record the diameter profile and the pitch, especially of the end coil, continuously or at specified times, the artificial intelligence can detect a fluctuation in the force penetration point values ​​by adjusting the spring geometry along the diameter axis or by changing the pitch, especially the pitch of the end coil, and react directly based on an adjusted geometry. This allows for control to adjust the center of force or the transverse forces.Any model that uses machine learning to determine the center of force of coil springs during forming based on training data can be used as artificial intelligence. The model can be a neural network, a regression algorithm, a slime mold algorithm, or a combination thereof. Other well-known models suitable for machine learning, and combinations thereof, can also be used.

[0017] In particular, the center of force can be determined using artificial intelligence in the data, particularly image data or laser data. Artificial intelligence comprises the model. A model describes a trained algorithm. Most common, known algorithms are suitable as algorithms, in particular artificial neural networks, Bayes classifiers, support vector machines (SVMs), linear regression, but deep learning approaches such as artificial neural networks based on transformer technology and / or convolutional artificial neural networks are preferred. This means that the model is preferably implemented as an artificial neural network. The artificial neural network can, for example, be based on transformer technology and / or be or comprise a convolutional artificial neural network.

[0018] In particular, it has been shown that no special algorithms need to be used to create the model. It is sufficient to train algorithms offered for commercial use with appropriate data. Such algorithms are also referred to as COTS (Components-Off-The-Shelf). It has been shown that the structure of the algorithm—for example, the number of neurons or the number of layers in artificial neural networks—is irrelevant for training the model as long as a minimum level of complexity is achieved. Algorithms that also perform well in public benchmarks are particularly suitable. Examples of benchmarks include the "COCO dev-test" and the "COCO val2017." Examples include models based on the DINO (DETR with Improved deNoising anchOr box) algorithm or the YOLO (You Only Look Once) algorithm.

[0019] In “Tianhe Ren, Jianwei Yang, Shilong Liu, Ailing Zeng, Feng Li, Hao Zhang, Hongyang Li, Zhaoyang Zeng, Lei Zhang. A Strong and Reproducible Object Detector with Only Public Datasets. arXiv:2304.13027 [cs.CV], (htps: / / doi.org / 10.48550 / arXiv.2304.13027)“ und „Hao Zhang, Feng Li, Shilong Liu, Lei Zhang, Hang Su, Jun Zhu, Lionel M. Ni, Heung-Yeung Shum. DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection. arXiv:2203.03605 [cs.CV], (htps: / / doi.org / 10.48550 / arXiv.2203.036Q5)“ sind Beispiele für die Modelle offenbart.

[0020] Furthermore, a cold winding machine for forming a coil spring is disclosed. The above-described method can be carried out on this cold winding machine, but also on any other cold winding machine. The cold winding machine comprises a wire feeder designed to introduce a wire for forming the coil spring into the cold winding machine. A first and a second deflection element are designed to set a current diameter of the coil spring as a first forming parameter. It is also possible to use a plurality of deflection elements with more than two deflection elements. A pitch element is designed to set a current pitch of the coil spring as a second forming parameter. A deflection roller, for example, can be used as the deflection element. A deflection roller, for example, can be used as the pitch element.Instead of the respective rollers, wedges or other suitable fixing elements can also be used, which form a fixed point for the wire running against the fixing element.

[0021] A sensor determines the geometry of the coil spring during forming and outputs corresponding sensor data. A signal processing unit is configured to determine, based on the sensor data, a deviation of the center of force or the transverse force of the formed coil spring from a predetermined center of force or a predetermined transverse force using the adjusted first forming parameter and the adjusted second forming parameter, and to adjust the first and / or second forming parameters during forming of the coil spring if the deviation is greater than a predetermined threshold value.

[0022] Using the wire feed, the automatic winch can work against the deflection element and the pitch element to form the coil spring. This means that the wire feed guides the spring wire and pushes it against the essentially rigidly arranged deflection element or pitch element, thereby shaping the coil spring and the two axes.

[0023] Analogous to the method, a computer program comprising instructions is disclosed which, when the program is executed by a computer, cause the computer to carry out the method.

[0024] Preferred embodiments of the present invention are explained below with reference to the accompanying drawings. They show:

[0025] Fig. 1 : a schematic perspective view of a cold winding machine with which the above-mentioned method can be carried out;

[0026] Fig. 2: a schematic side view of three coil springs with different centers of force, shown in Fig. 2a, Fig. 2b and Fig. 2c.

[0027] Before exemplary embodiments of the present invention are explained in more detail below with reference to the drawings, it is pointed out that identical, functionally equivalent or equivalent elements, objects and / or structures are provided with the same reference numerals in the different figures, so that the description of these elements shown in different exemplary embodiments is interchangeable or can be applied to one another. Fig. 1 shows a schematic perspective view of a cold coiling machine 20. The cold coiling machine 20 comprises two deflection elements 22a, 22b, shown here as deflection rollers. It is also possible to use further deflection elements and thus any desired plurality of deflection elements. The deflection elements can be moved at least along an axis 22a', 22b', preferably three-dimensionally in space, in order to adjust a (current) diameter of the formed helical spring.The current position of the deflection elements 22a, 22b, ie the (current) diameter, can be referred to as the first shaping parameter.

[0028] The cold coiling machine 20 further comprises a pitch element 24. The pitch element 24 is freely positionable at least along an axis 24' to adjust the (current) pitch of the formed coil spring. A working cylinder, for example, can be used as the pitch element. However, other configurations are also possible. The position of the pitch element, ie, the (current) pitch, can be referred to as the second forming parameter.

[0029] The cold coiling machine 20 further includes a wire feed 26, symbolized by a directional arrow. The wire feed feeds the wire 27 from the coil spring, either from a wire coil or already cut to the correct length, into the cold coiling machine to form the coil spring. The feed allows the wire to work against the deflection element and the pitch element to form the coil spring. A first coil 27' of the coil spring is shown as an example. The wire feed speed can be referred to as the third forming parameter.

[0030] The cold coiling machine 20 further comprises a sensor 28. The sensor 28 can be used to determine the geometry of the coil spring during forming, and to output sensor data 30 corresponding to the geometry. The sensor data 30 can be processed by a signal processing unit 32. Based on the sensor data 30, the signal processing unit 32 can determine a deviation of the center of force or the transverse force of the formed coil spring using the set first forming parameter and the set second forming parameter compared to a predetermined center of force or a predetermined transverse force. Furthermore, the signal processing unit can adjust the first and / or the second forming parameter during forming of the coil spring if the deviation is greater than a predetermined threshold value.

[0031] It should be noted again that the execution of the method is not limited to the cold winding machine shown here, but can also be carried out on other cold winding machines.

[0032] Fig. 2a, Fig. 2b and Fig. 2c show a schematic side view of three coil springs 34 with different force centers. The lines labeled F show the force action line 36, the dashed line 38 shows the center axis of the spring. L denotes the length of the spring, so that at the outer end points 40a, 40b the spring rests against the vehicle, for example on spring plates. The intersection point of the force action line 36 with the end points 40a, 40b defines a force center 42a, 42b of the coil spring 34. A transverse force of the spring 34 can be adjusted via the deviation of the two force centers 42a, 42b from the center axis 38 of the spring 34. The transverse force of the spring is specified, for example, by the customer.The described method now makes it possible to automatically compensate deviations from a target geometry, knowing the spring geometry wound up to a measurement point (actual geometry), in order to set a predetermined force action line.

[0033] Fig. 2a shows, by way of example, a force action line 36 that runs parallel to the central axis 38 of the coil spring. Fig. 2b shows, by way of example, a force action line 36 in which the force centers 42a, 42b are arranged symmetrically to the central axis 38 of the coil spring. Fig. 2c shows, by way of example, a force action line 36 whose upper force center 42a lies on the central axis 38 of the coil spring and whose lower force center 42b is shifted to the left of the central axis 38. It should be noted that both the gradient and the horizontal displacement of the force action line relative to the central axis of the spring are variable, and only three examples are shown here.Although some aspects have been described in connection with a device, it is to be understood that these aspects also represent a description of the corresponding method, so that a block or component of a device can also be understood as a corresponding method step or as a feature of a method step. Analogously, aspects described in connection with or as a method step also represent a description of a corresponding block, detail, or feature of a corresponding device. The embodiments described above merely illustrate the principles of the present invention. It is to be understood that modifications and variations of the arrangements and details described herein will be apparent to others skilled in the art.Therefore, it is intended that the invention be limited only by the scope of the following claims and not by the specific details presented in the description and explanation of the embodiments herein.

[0034] List of reference symbols:

[0035] 20 cold winding machines

[0036] 22 Deflection element 24 Gradient element

[0037] 26 Wire feed

[0038] 27 wire

[0039] 27' first turn of the coil spring

[0040] 28 Sensor 30 Sensor data

[0041] 32 Signal processing unit

[0042] 34 coil spring

[0043] 36 Line of action of force

[0044] 38 Central axis 40 End points of the coil spring

[0045] 42 centers of force

Claims

Patent claims 1. A computer-implemented method for adjusting a force center or a transverse force of a coil spring (34) during cold winding with a cold winding machine (20), comprising the following steps: - Determining a geometry of the coil spring (34) during cold winding - Determining, based on the determined geometry, a deviation of the center of force or the transverse force of the formed coil spring (34) using a set forming parameter of the cold coiling machine (20) compared to a predetermined center of force or a predetermined transverse force; - Adjusting the set forming parameter of the cold coiling machine during the forming of the coil spring if the deviation is greater than a specified threshold value.

2. Method according to claim 1, wherein in determining the geometry of the helical spring (34) a pitch of the helical spring, in particular a profile of the pitch of the helical spring (34), is determined.

3. Method according to one of the preceding claims, wherein in determining the geometry of the helical spring (34) a diameter of the helical spring (34), in particular a profile of the diameter of the helical spring (34), is determined.

4. Method according to one of the preceding claims, wherein the pitch of the end turn of the helical spring (34) is set as a shaping parameter.

5. Method according to one of the preceding claims, wherein the diameter of the helical spring (34) is set as a shaping parameter.

6. Method according to one of the preceding claims, wherein the determination of the deviation of the center of force or the transverse force of the shaped Coil spring (34) is formed using a set forming parameter of the cold winding machine against a predetermined force center or a predetermined transverse force by a model trained by means of machine learning, in particular an artificial intelligence (32).

7. Cold winding machine (20) for forming a helical spring (34) with the following features: - a wire feeder (26) designed to introduce a wire (27) for shaping the helical spring (34) into the cold coiling machine (20); - a first and a second deflection element (22a, 22b) which are designed to set a current diameter of the helical spring (34) as a first shaping parameter; - a pitch element (24) which is designed to set a current pitch of the helical spring (34) as a second shaping parameter; - a sensor configured to determine a geometry of the coil spring (34) during forming and to output corresponding sensor data; - a signal processing unit configured to determine, based on the sensor data, a deviation of the center of force or the transverse force of the shaped coil spring (34) using the set first shaping parameter and the set second shaping parameter from a predetermined center of force or a predetermined transverse force; - adjusting the first and / or second forming parameter during the forming of the helical spring (34) if the deviation is greater than a predetermined threshold value.

8. Cold coiling machine according to claim 7, wherein the cold coiling machine (20) is designed to work by the wire feed (26) against the deflection element (22a, 22b) and / or the pitch element (24) in order to form the helical spring (24). A computer program comprising instructions which, when executed by a computer, cause the computer to carry out the method according to any one of claims 1 to 6.

Images