Sun gear oriented spline machining method and machining device

Abstract

The application relates to the spline machining technical field, in particular to a spline machining method and device for a sun gear, the method comprising the following steps: collecting vibration signals of a milling cutter during the whole process of milling spline treatment of any sun gear by using a spline milling machine; obtaining a milling signal section set and an interval signal section set; obtaining similar characteristic values, periodic characteristic values and reserved characteristic values of each modal component of each signal section in the milling signal section set; reconstructing each signal section and merging the signal section with a signal section in the interval signal section set after smoothing processing; and evaluating whether to replace the milling cutter before milling spline treatment of a next sun gear of the any sun gear. The application aims to improve the effectiveness of the vibration signals of the milling cutter during the milling spline process of the sun gear, improve the accuracy of the cutter wear state evaluation, and ensure the machining precision and efficiency of the sun gear spline.

Description

Technical Field

[0001] This application relates to the field of spline machining technology, specifically to a spline machining method and apparatus for sun gears. Background Technology

[0002] Planetary reducers are precision reducers whose transmission structure mainly consists of three parts: planetary gears, a sun gear, and an internal gear. They are often used in servo motors to reduce speed and increase torque. The sun gear is a crucial component in the transmission structure of a precision planetary reducer. A spline is a high-precision, high-load-bearing connection structure composed of multiple circumferentially distributed key teeth on the shaft or hub of the sun gear. It is used to connect other gears or couplings. The machining accuracy of the sun gear spline directly determines the transmission accuracy and load-bearing capacity of the planetary reducer. Spline milling machines are commonly used spline machining devices, and the wear condition of the milling cutters used in spline milling machines is a key factor affecting the machining accuracy of the sun gear spline. Therefore, during the spline machining of the sun gear, it is necessary to monitor the wear condition of the milling cutters used in the spline milling machine and replace them in a timely manner to ensure the machining accuracy of the subsequent sun gear spline.

[0003] Existing methods typically use signals related to machine tool wear, such as vibration signals, to monitor the wear condition of machine tool tools in real time. However, these methods neglect vibrations caused by non-tool vibration sources during the cutting process, such as machine tool component installation errors and workpiece material inhomogeneity, as well as interference from sensor noise. This reduces the effectiveness of the acquired vibration signals, affecting the assessment of the wear condition of the spline milling machine tool used in the spline machining of the sun gear. Consequently, misjudgments of the wear condition of the spline milling machine tool are prone to occur, impacting the overall machining accuracy and efficiency of the sun gear spline. Summary of the Invention

[0004] In view of the above, it is necessary to provide a spline machining method and apparatus for sun gears. Compared with traditional spline machining methods and apparatus for sun gears, this method improves the effectiveness of the vibration signal of the milling cutter during the spline milling process of the sun gear, and enhances the accuracy of tool wear condition assessment, thereby ensuring the overall machining accuracy and efficiency of the sun gear spline.

[0005] In a first aspect, embodiments of this application provide a spline machining method for a sun gear, the method comprising the following steps:

[0006] Vibration signals of the milling cutter were collected throughout the entire process of milling splines on any sun gear using a spline milling machine.

[0007] By analyzing the amplitude distribution in the vibration signal, the vibration signal is divided into a set of milling signal segments and a set of interval signal segments;

[0008] For each signal segment in the milling signal segment set, modal decomposition is performed separately. By the similarity between each modal component of each signal segment and the modal components of other signal segments, the similarity feature values ​​of each modal component of each signal segment are obtained. Then, combined with the autocorrelation of each modal component of each signal segment, the periodic feature values ​​of each modal component of each signal segment are obtained, and the retained feature values ​​of each modal component of each signal segment are obtained.

[0009] By retaining the feature values, each signal segment is reconstructed; all reconstructed signal segments are merged with the smoothed signal segments in the interval signal segment set to obtain the processed vibration signal; by using the time-domain and frequency-domain characteristics of the processed vibration signal, it is evaluated whether the milling cutter should be replaced before milling splines on the next sun gear of any sun gear.

[0010] In one embodiment, the process of obtaining the milling signal segment set and the interval signal segment set is as follows:

[0011] Obtain the segmentation threshold of the amplitude of all sampling points in the acquired vibration signal, and form the signal segments composed of amplitudes greater than or less than or equal to the segmentation threshold into a milling signal segment set and an interval signal segment set.

[0012] In one embodiment, the process of obtaining the similarity feature values ​​is as follows:

[0013] The center frequency of each modal component of each signal segment is obtained by signal decomposition algorithm. The modal component with the smallest difference value between the center frequency of each modal component of the remaining signal segments and the modal components of each signal segment is taken as the reference modal component of each modal component of each signal segment.

[0014] By measuring the similarity between each modal component of each signal segment and all its corresponding modal components, the similarity feature values ​​of each modal component of each signal segment are obtained.

[0015] In one embodiment, the similarity feature value is calculated as follows:

[0016] Calculate the mean difference between each modal component of each signal segment and all its control modal components;

[0017] The similarity feature value is inversely proportional to the mean value.

[0018] In one embodiment, the periodic characteristic value is the maximum value of the autocorrelation coefficient of each modal component of each signal segment under all preset lag orders.

[0019] In one embodiment, the retained feature value is the average of the normalized values ​​of the similar feature values ​​and the normalized values ​​of the periodic feature values.

[0020] In one embodiment, the process of reconstructing each signal segment is as follows:

[0021] Obtain the segmentation threshold for retaining the feature values ​​of all modal components in each signal segment, filter out the modal components in each signal segment whose retain feature values ​​are greater than the segmentation threshold, and reconstruct each signal segment based on all the filtered modal components.

[0022] In one embodiment, the time-domain features and frequency-domain features include at least the mean, standard deviation, root mean square deviation, peak factor, kurtosis index, and skewness index, and the frequency-domain features include at least the frequency variance, root mean square frequency, and centroid frequency.

[0023] In one embodiment, the process of evaluating whether to change the milling cutter before milling the spline on the next sun gear of any of the sun gears is as follows:

[0024] The wear state identification vector of the vibration signal is formed by combining the time-domain and frequency-domain characteristics of the processed vibration signal.

[0025] Based on the wear state identification vectors of vibration signals during spline milling using milling cutters under different wear states, a binary classification model is trained to obtain a tool wear state identification model. If the spline obtained by milling spline using a milling cutter under any wear state identification vector meets the production standard, the label of the wear state identification vector is set to a first preset value; otherwise, the label of the wear state identification vector is set to a second preset value.

[0026] The label of the wear state identification vector corresponding to any sun gear is obtained by the tool wear state identification model. If the label is the first preset value, the milling cutter is not replaced before the spline milling of the next sun gear; otherwise, the milling cutter is replaced.

[0027] Secondly, embodiments of this application also provide a spline machining apparatus for a sun gear, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of the spline machining method for a sun gear described above.

[0028] This application has at least the following beneficial effects:

[0029] This application analyzes the spline milling process of the sun gear and divides the vibration signal into a set of milling signal segments and a set of interval signal segments. It can clearly distinguish between the stage when the tool is actually milling the spline teeth and the stage when the tool is separated from the workpiece. This allows for targeted analysis of signals at different stages, effectively reducing the interference of vibration signals in the non-milling stage on the analysis of signals in the milling stage, improving the accuracy of subsequent signal processing, and thus improving the accuracy of assessing the wear state of the milling cutter based on the vibration signal.

[0030] Furthermore, by analyzing the characteristics of vibration signals generated by the milling cutter itself during the spline milling process of the sun gear, as well as the characteristics of vibration signals generated by non-tool vibration sources and sensor noise interference, similar feature values ​​and periodic feature values ​​are obtained. The consistency of tool vibration signals during different spline milling processes is evaluated by using similar feature values, and the periodic feature values ​​are used to identify periodic cutting vibration characteristics in the tool vibration signals. Then, by combining similar feature values ​​and periodic feature values, tool vibration signals and non-tool vibration signals are distinguished. Feature information related to tool wear status is extracted from the vibration signals, and the vibration signals are reconstructed to improve the accuracy of tool wear status assessment.

[0031] Furthermore, after smoothing the signal segments in the interval signal segment set, they are merged with the reconstructed signal segments. The vibration signal obtained by merging is used to evaluate the wear state of the milling cutter used in the spline milling of the sun gear. Based on the evaluation results, it is determined whether the milling cutter needs to be replaced. This can effectively reduce the interference of vibrations caused by non-tool vibration sources such as spline milling machine part installation errors and workpiece material inhomogeneity, as well as the noise of the sensor itself in the collected vibration signal. This improves the effectiveness of the collected vibration signal, enabling a more accurate evaluation of the wear state of the spline milling machine cutter used in the spline milling of the sun gear. It avoids the impact on the overall machining accuracy and efficiency of the sun gear spline due to misjudgment of the spline milling machine cutter wear state. Attached Figure Description

[0032] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 A flowchart illustrating the steps of a spline machining method for a sun gear, as provided in one embodiment of this application;

[0034] Figure 2 This is a schematic diagram of the vibration signal processing flow. Detailed Implementation

[0035] In the description of the embodiments in this application, the words "exemplary," "or," and "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary," "or," and "for example" is intended to present the relevant concepts in a specific manner.

[0036] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. It should be understood that, unless otherwise stated, " / " in this application means "or".

[0037] It should also be noted that the terms "first" and "second" in this application are used to distinguish similar objects, rather than to describe a specific order or sequence.

[0038] The following description, in conjunction with the accompanying drawings, details a specific scheme for a spline machining method and machining apparatus for a sun gear provided in this application.

[0039] Please see Figure 1 The diagram illustrates a step-by-step flowchart of a spline machining method for a sun gear according to an embodiment of this application. The method includes the following steps:

[0040] Step 1: Collect the vibration signal of the milling cutter during the entire process of milling splines on any sun gear using a spline milling machine.

[0041] In this application, the spline machining process of the sun gear mainly includes three processes: spline milling, quenching, and spline grinding.

[0042] In the spline milling process, a vibration sensor is installed on the milling cutter of the spline milling machine used for the sun gear. The vibration sensor collects the vibration signal of the milling cutter during the entire process of milling splines on any sun gear in any batch. This allows for analysis of the vibration generated by the milling cutter during the entire milling process of the spline teeth of any sun gear, thereby assessing whether to replace the milling cutter before milling splines on the next sun gear, ensuring the accuracy and efficiency of spline machining for subsequent sun gears.

[0043] In this embodiment, the sampling frequency of the vibration sensor is 10kHz. The sampling frequency is preset by the user and can be set by the user according to the actual situation. This application does not impose any special restrictions.

[0044] Step 2: Analyze and process the acquired vibration signals.

[0045] Step 2.1: Divide the vibration signal into milling signal segment set and interval signal segment set by the amplitude distribution in the vibration signal.

[0046] Under normal circumstances, when using a milling cutter on a spline milling machine to mill splines on a sun gear, the cutter face will continuously rub against the sun gear surface, resulting in significant cutting vibration. After each spline tooth on the sun gear surface is milled, the sun gear will rotate in an indexing motion so that the cutter can mill the next spline tooth on the sun gear surface. However, during the indexing rotation of the sun gear, the cutter will not vibrate significantly because it will detach from the sun gear surface.

[0047] Based on the above analysis, the segmentation threshold of the amplitude of all sampling points in the acquired vibration signal is obtained. The vibration signal is divided into multiple signal segments with amplitudes greater than the segmentation threshold and multiple signal segments with amplitudes less than or equal to the segmentation threshold. All signal segments with amplitudes greater than the segmentation threshold are combined into a milling signal segment set, and all signal segments with amplitudes less than or equal to the segmentation threshold are combined into an interval signal segment set. The milling signal segment set and the interval signal segment set are used to characterize the sets of vibration signal segments in the vibration signal that are in the time period of the spline key tooth milling process on the surface of the sun gear and not in the vibration signal.

[0048] In this embodiment, the Otsu threshold segmentation algorithm is used to obtain the segmentation threshold of the amplitude of all sampling points in the acquired vibration signal. The Otsu threshold segmentation algorithm is a well-known technology and will not be described in detail in this application. As other implementation methods, based on the ability to obtain the segmentation threshold of the amplitude of all sampling points in the acquired vibration signal, the implementer may use other existing technologies, such as iterative threshold segmentation, global threshold segmentation, etc. This application does not impose any special restrictions.

[0049] During the spline milling process of the sun gear, when the cutting tool is outside the time frame of the spline teeth milling process, the acquired vibration signal segments do not contain information related to the wear state of the milling cutter. These vibration signal segments are mainly affected by vibrations from non-milling cutter vibration sources and the noise of the vibration sensor itself. To reduce the impact of this non-milling cutter vibration information and the noise of the vibration sensor itself on subsequent analysis of the milling cutter wear state using the acquired vibration signals, the signal segments in the interval signal segment set are smoothed.

[0050] In this embodiment, the Savitzky-Golay filtering algorithm is used to smooth the signal segments in the interval signal segment set. The Savitzky-Golay filtering algorithm is a well-known technology and will not be described in detail here. As other implementation methods, based on the ability to smooth the signal segments in the interval signal segment set, implementers may use other existing feasible technologies, and this application does not impose any special restrictions.

[0051] Step 2.2: For each signal segment in the milling signal segment set, perform mode decomposition and obtain the similarity feature values ​​of each mode component of each signal segment by comparing the similarity between each mode component of each signal segment and the mode components of the other signal segments.

[0052] In the process of milling splines on the sun gear, to ensure the consistency of the dimensions of all spline teeth on the sun gear surface, milling parameters that are close or consistent are usually used to mill the spline teeth, such as cutting speed, feed rate, and cutting time. This ensures that the tool vibration signal components generated by the tool vibration should have similar signal distributions during the time period of the milling process of each spline tooth on the sun gear surface. However, the non-tool vibration signal components generated by vibration sources other than milling cutter vibration sources, such as installation errors of milling machine parts or uneven material of the sun gear workpiece, as well as the noise vibration signal components introduced by the vibration sensor itself, usually do not have this characteristic.

[0053] Based on the above analysis, for each signal segment in the milling signal segment set, a signal decomposition algorithm is used to perform modal decomposition on each signal segment to obtain each modal component and the center frequency of each modal component, which are used to characterize the different signal components in each signal segment.

[0054] In this embodiment, variational mode decomposition algorithm is used to obtain each mode component and the center frequency of each mode component of each signal segment. Variational mode decomposition algorithm is a well-known technology and will not be described in detail in this application. As other implementation methods, based on the ability to obtain each mode component and the center frequency of each mode component of each signal segment, the implementer may use other existing feasible technologies. This application does not impose any special restrictions.

[0055] Furthermore, by analyzing the similarity between each modal component of each signal segment and the modal components of other signal segments, similarity feature values ​​for each modal component of each signal segment are obtained. The specific process is as follows:

[0056] Taking any modal component b1 of any signal segment b in the milling signal segment set as an example, the modal component with the smallest difference in center frequency between the modal components of each other signal segment in the milling signal segment set and the modal component b1 of signal segment b is taken as each reference modal component of modal component b1 of signal segment b. This reference modal component is used to characterize the vibration signal component in each other signal segment that is in the same frequency range as the vibration signal component corresponding to modal component b1. Then, the difference between the modal component b1 of signal segment b and each reference modal component is calculated, and the difference between the modal component b1 of signal segment b and its reference modal component is calculated. The mean difference between all control modal components, and the similarity feature value of modal component b1 of signal segment b are inversely proportional to the mean. The similarity feature value is used to evaluate whether there is a vibration signal component with a similar signal distribution to the vibration signal component corresponding to modal component b1 among the vibration signals collected during the time period of the milling process of each spline tooth other than the spline tooth corresponding to signal segment b during the spline milling process of the sun gear. The larger the similarity feature value, the more likely there is a vibration signal component with a similar signal distribution to the vibration signal component corresponding to modal component b1.

[0057] In this embodiment, the difference between center frequencies is the absolute value of the difference. As other implementation methods, based on the ability to measure the degree of difference between center frequencies, the implementer may use other calculation methods, such as the square of the difference, etc. This application does not impose any special restrictions.

[0058] In this embodiment, the difference between modal components is the DTW (Dynamic Time Warping) distance between modal components. As other implementations, based on the measurable difference between modal components, implementers may adopt other existing technologies, such as Euclidean distance, etc. This application does not impose any special restrictions.

[0059] In this embodiment, the reciprocal of the sum of the mean and the preset positive number is used as the similarity feature value of the modal component b1 of the signal segment b. The preset positive number is used to avoid the denominator being 0. The value of the preset positive number is preset by the user and can be set by the implementer according to the actual situation. In this embodiment, the value of the preset positive number is 0.001.

[0060] Based on the method for calculating the similarity eigenvalues ​​of the modal component b1 of signal segment b, the similarity eigenvalues ​​of each modal component of each signal segment in the milling signal segment set are calculated.

[0061] Step 2.3: Obtain the periodic characteristic values ​​of each modal component of each signal segment by the autocorrelation of each modal component of each signal segment.

[0062] However, the frequency components and frequency band energy of the vibration signal generated by the milling cutter usually change with the amount of tool wear. Therefore, during the milling process of a spline tooth on the surface of the sun gear, if the wear of the milling cutter intensifies, the tool vibration signal component in the vibration signal collected during the milling of that spline tooth will no longer have similar signal distribution characteristics to the tool vibration signal components collected during the milling of several previous spline teeth. Furthermore, during the milling process of the spline teeth on the surface of the sun gear using a milling cutter, the periodic entry and exit of the cutter will generate periodic cutting vibration. This periodicity is mainly reflected in the tool vibration signal component, while non-tool vibration signal components and noise signal components typically do not possess this characteristic. Therefore, to effectively reduce the impact of interference information introduced by non-milling cutter vibration sources and the noise of the vibration sensor itself on the subsequent tool wear condition assessment during the milling process of each spline tooth on the surface of the sun gear, the following processing is performed.

[0063] Based on the above analysis, taking modal component b1 of signal segment b as an example, the autocorrelation function is used to calculate the autocorrelation coefficient of modal component b1 of signal segment b at each preset lag order. The maximum value of the autocorrelation coefficient of modal component b1 of signal segment b at all preset lag orders is taken as the periodic characteristic value of modal component b1 of signal segment b. This value is used to evaluate whether the vibration signal component corresponding to modal component b1 has obvious periodic signal distribution characteristics. The larger the periodic characteristic value, the more obvious the periodic signal distribution characteristics. The autocorrelation function is a well-known technique and will not be described in detail in this application.

[0064] In this embodiment, when calculating the autocorrelation coefficient of the modal component b1 of signal segment b under preset lag orders, the range of the lag order is [1, n], where n represents the number of sampling points in the modal component b1. The implementer can set the range of the lag order according to the actual situation, and this application does not impose any special restrictions.

[0065] Based on the method for calculating the periodic characteristic value of the modal component b1 of signal segment b, calculate the periodic characteristic value of each modal component of each signal segment in the milling signal segment set.

[0066] Step 2.4: Combine the similarity feature values ​​and periodic feature values ​​of each modal component of each signal segment to obtain the retained feature values ​​of each modal component of each signal segment.

[0067] The similarity feature values ​​and periodic feature values ​​of all modal components in all signal segments of the milling signal segment set are normalized respectively. The normalized values ​​of the similarity feature values ​​and the normalized values ​​of the periodic feature values ​​of each modal component of each signal segment are used as the retained feature values ​​of each modal component of each signal segment. This is used to evaluate whether the vibration signal components corresponding to each modal component of each signal segment need to be retained in the acquired vibration signal. The larger the retained feature value, the more the vibration signal components corresponding to each modal component of each signal segment should be retained.

[0068] In this embodiment, the Min-Max normalization method is used to normalize the similar feature values ​​and the periodic feature values ​​respectively. The Min-Max normalization method is a well-known technology and will not be described in detail in this application.

[0069] Step 2.5: Reconstruct each signal segment using the retained feature values; merge all the reconstructed signal segments with the smoothed signal segments in the interval signal segment set to obtain the processed vibration signal.

[0070] Taking signal segment b as an example, the segmentation threshold for retaining the feature values ​​of all modal components of signal segment b is obtained. Modal components with retaining feature values ​​greater than the segmentation threshold are filtered out. The filtered modal components are used to characterize the modal components corresponding to all vibration signal components generated by the vibration of the milling cutter itself in signal segment b. All the filtered modal components are reconstructed to obtain the reconstructed signal segment of signal segment b, which is used to characterize the vibration signal generated by the vibration of the milling cutter itself during the time period of the milling process of the spline teeth corresponding to signal segment b.

[0071] In this embodiment, the Otsu threshold segmentation algorithm is used to obtain the segmentation threshold for preserving the feature values ​​of all modal components of signal segment b. The Otsu threshold segmentation algorithm is a well-known technology and will not be described in detail in this application. As other implementation methods, based on the ability to obtain the segmentation threshold for preserving the feature values ​​of all modal components of signal segment b, the implementer may use other existing technologies, such as iterative threshold segmentation, global threshold segmentation, etc. This application does not impose any special restrictions.

[0072] According to the method for obtaining the reconstructed signal segment of signal segment b, obtain the reconstructed signal segment of each signal segment in the milling signal segment set.

[0073] Furthermore, the reconstructed signal segments of all signal segments in the milling signal segment set and all signal segments in the interval signal segment set are merged according to time sequence. The resulting vibration signal is recorded as the effective vibration signal, used to characterize the vibration signal generated by the milling cutter itself during the milling of the spline on the sun gear. A schematic diagram of the vibration signal processing flow is shown below. Figure 2 As shown.

[0074] Step 3: By analyzing the time-domain and frequency-domain characteristics of the processed vibration signal, assess whether to change the milling cutter before performing spline milling on the next sun gear of any sun gear.

[0075] Multiple time-domain and frequency-domain feature parameters of the effective vibration signal are extracted. These parameters are then combined to form a wear state identification vector for the effective vibration signal. In this embodiment, the time-domain feature parameters include mean, standard deviation, root mean square deviation, peak factor, kurtosis index, and skewness index. The frequency-domain feature parameters include frequency variance, mean square frequency, and centroid frequency. Implementers may add other existing time-domain and frequency-domain feature parameters as needed; this application does not impose any special restrictions. The calculation of peak factor, kurtosis index, skewness index, frequency variance, mean square frequency, and centroid frequency are all well-known techniques and will not be elaborated upon here.

[0076] A method for obtaining wear state identification vectors from effective vibration signals is employed. This method acquires wear state identification vectors from vibration signals generated during the spline milling of a sun gear using milling cutters under different wear states. The number of wear state identification vectors is 200. A professional determines whether the splines of the sun gear obtained under different wear states meet production standards. If the splines obtained by milling the splines using a milling cutter under any wear state identification vector meet production standards, the label of that wear state identification vector is set to a first preset value; otherwise, the label is set to a second preset value. A binary classification model is trained based on the acquired wear state identification vectors and their labels, and this trained binary classification model is used as the tool wear state identification model. The number 200 is merely one embodiment of this application; implementers can set its specific value according to actual conditions, and this application does not impose any special limitations.

[0077] In this embodiment, the first preset value and the second preset value are 1 and 0 respectively. The values ​​of the first preset value and the second preset value are preset by humans. The implementer can set their specific values ​​by himself. This application does not impose any special restrictions.

[0078] In this embodiment, a support vector machine (SVM) model is trained, and the F1-Score is used as the evaluation metric for the SVM model. The training process of the SVM model is a well-known technique and will not be described in detail here.

[0079] The wear state identification vector of the effective vibration signal is used as the input of the tool wear state identification model, and the label of the wear state identification vector of the effective vibration signal is output. If the output label is the first preset value, it means that the wear of the milling cutter used in the spline machining of any sun gear is within the normal wear range. The milling cutter is not replaced before the next sun gear of any sun gear is milled. Otherwise, the milling cutter is replaced.

[0080] After the spline milling process, the sun gear is subjected to quenching. Before quenching, an anti-seepage material needs to be brushed onto the inner hole surface of the sun gear for protection to prevent the formation of a hardened layer on the inner hole surface during the quenching process.

[0081] The spline of the sun gear after quenching is ground using a gear grinding machine to achieve the desired finish.

[0082] Based on the same inventive concept as the above method, this application embodiment also provides a spline machining apparatus for a sun gear, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-described spline machining methods for a sun gear.

[0083] In summary, this application analyzes the spline milling process of the sun gear, dividing the vibration signal into a set of milling signal segments and a set of interval signal segments. This clearly distinguishes between the stage when the tool is actually milling the spline teeth and the stage when the tool is detached from the workpiece. This allows for targeted analysis of signals at different stages, effectively reducing the interference of vibration signals from non-milling stages on the analysis of signals from the milling stage, improving the accuracy of subsequent signal processing, and thus improving the accuracy of assessing the wear state of the milling cutter based on vibration signals.

[0084] Furthermore, by analyzing the characteristics of vibration signals generated by the milling cutter itself during the spline milling process of the sun gear, as well as the characteristics of vibration signals generated by non-tool vibration sources and sensor noise interference, similar feature values ​​and periodic feature values ​​are obtained. The consistency of tool vibration signals during different spline milling processes is evaluated by using similar feature values, and the periodic feature values ​​are used to identify periodic cutting vibration characteristics in the tool vibration signals. Then, by combining similar feature values ​​and periodic feature values, tool vibration signals and non-tool vibration signals are distinguished. Feature information related to tool wear status is extracted from the vibration signals, and the vibration signals are reconstructed to improve the accuracy of tool wear status assessment.

[0085] Furthermore, after smoothing the signal segments in the interval signal segment set, they are merged with the reconstructed signal segments. The vibration signal obtained by merging is used to evaluate the wear state of the milling cutter used in the spline milling of the sun gear. Based on the evaluation results, it is determined whether the milling cutter needs to be replaced. This can effectively reduce the interference of vibrations caused by non-tool vibration sources such as spline milling machine part installation errors and workpiece material inhomogeneity, as well as the noise of the sensor itself in the collected vibration signal. This improves the effectiveness of the collected vibration signal, enabling a more accurate evaluation of the wear state of the spline milling machine cutter used in the spline milling of the sun gear. It avoids the impact on the overall machining accuracy and efficiency of the sun gear spline due to misjudgment of the spline milling machine cutter wear state.

[0086] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than that shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in a different order than disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. Each block in a block diagram and / or flowchart, and combinations of blocks in a block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0087] It will be apparent to those skilled in the art that this application is not limited to the details of the exemplary embodiments described above, and that this application can be implemented in other specific forms without departing from its essential characteristics. Therefore, the embodiments described above should be considered exemplary and non-limiting in all respects.

Claims

1. A method for machining splines on a sun gear, characterized in that, The method includes the following steps: Vibration signals of the milling cutter were collected throughout the entire process of milling splines on any sun gear using a spline milling machine. By analyzing the amplitude distribution in the vibration signal, the vibration signal is divided into a set of milling signal segments and a set of interval signal segments; For each signal segment in the milling signal segment set, modal decomposition is performed separately. By the similarity between each modal component of each signal segment and the modal components of other signal segments, the similarity feature values ​​of each modal component of each signal segment are obtained. Then, combined with the autocorrelation of each modal component of each signal segment, the periodic feature values ​​of each modal component of each signal segment are obtained, and the retained feature values ​​of each modal component of each signal segment are obtained. By retaining the feature values, each signal segment is reconstructed; all reconstructed signal segments are merged with the smoothed signal segments in the interval signal segment set to obtain the processed vibration signal; by using the time-domain and frequency-domain characteristics of the processed vibration signal, it is evaluated whether the milling cutter should be replaced before milling splines on the next sun gear of any sun gear.

2. The spline machining method for a sun gear as described in claim 1, characterized in that, The process of obtaining the set of milling signal segments and the set of interval signal segments is as follows: Obtain the segmentation threshold of the amplitude of all sampling points in the acquired vibration signal, and form the signal segments composed of amplitudes greater than or less than or equal to the segmentation threshold into a milling signal segment set and an interval signal segment set.

3. The spline machining method for a sun gear as described in claim 1, characterized in that, The process for obtaining the similarity feature values ​​is as follows: The center frequency of each modal component of each signal segment is obtained by signal decomposition algorithm. The modal component with the smallest difference value between the center frequency of each modal component of the remaining signal segments and the modal components of each signal segment is taken as the reference modal component of each modal component of each signal segment. By measuring the similarity between each modal component of each signal segment and all its corresponding modal components, the similarity feature values ​​of each modal component of each signal segment are obtained.

4. The spline machining method for a sun gear as described in claim 3, characterized in that, The method for calculating the similarity feature values ​​is as follows: Calculate the mean difference between each modal component of each signal segment and all its control modal components; The similarity feature value is inversely proportional to the mean value.

5. The spline machining method for a sun gear as described in claim 1, characterized in that, The periodic characteristic value is the maximum value of the autocorrelation coefficient of each modal component of each signal segment under all preset lag orders.

6. The spline machining method for a sun gear as described in claim 1, characterized in that, The retained feature value is the average of the normalized values ​​of the similar feature values ​​and the normalized values ​​of the periodic feature values.

7. The spline machining method for a sun gear as described in claim 1, characterized in that, The process of reconstructing each signal segment is as follows: Obtain the segmentation threshold for retaining the feature values ​​of all modal components in each signal segment, filter out the modal components in each signal segment whose retain feature values ​​are greater than the segmentation threshold, and reconstruct each signal segment based on all the filtered modal components.

8. The spline machining method for a sun gear as described in claim 1, characterized in that, The time-domain features and frequency-domain features include at least the mean, standard deviation, root mean square deviation, peak factor, kurtosis index, and skewness index in the time-domain features, and at least the frequency variance, root mean square frequency, and centroid frequency in the frequency-domain features.

9. The spline machining method for a sun gear as described in claim 1, characterized in that, The process of assessing whether to change the milling cutter before milling the spline on the next sun gear of any of the sun gears is as follows: The wear state identification vector of the vibration signal is formed by combining the time-domain and frequency-domain characteristics of the processed vibration signal. Based on the wear state identification vectors of vibration signals during spline milling using milling cutters under different wear states, a binary classification model is trained to obtain a tool wear state identification model. If the spline obtained by milling spline using a milling cutter under any wear state identification vector meets the production standard, the label of the wear state identification vector is set to a first preset value; otherwise, the label of the wear state identification vector is set to a second preset value. The label of the wear state identification vector corresponding to any sun gear is obtained by the tool wear state identification model. If the label is the first preset value, the milling cutter is not replaced before the spline milling of the next sun gear; otherwise, the milling cutter is replaced.

10. A spline machining apparatus for a sun gear, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the spline machining method for a sun gear as described in any one of claims 1-9.

Images