Efficient machining method and device for cycloidal gear of speed reducer based on machining machine tool

Abstract

The invention relates to the technical field of cycloidal gear machining control, in particular to a speed reducer cycloidal gear efficient machining method and device based on a machining machine tool, and the method comprises the steps that machining data in the cycloidal gear machining process are obtained and preprocessed; preliminarily judging whether a flutter phenomenon occurs or not by analyzing the periodic change of the cutting depth and the periodic fluctuation of the cutting force under different temperature changes; vibration, current and cutting force data are combined to evaluate the chatter occurrence possibility; classifying the processing data through clustering analysis by utilizing the chatter occurrence possibility so as to screen out high-quality processing data and flaw processing data; and a dragonfly optimization algorithm is adopted, initial position distribution of the algorithm is optimized through the high-quality machining data and the defect machining data, and the optimal cutting speed is solved to serve as the cutting speed in the current cycloidal gear machining process. According to the method, the flutter phenomenon is comprehensively analyzed through the multi-dimensional machining data, so that the overall machining efficiency in the current cycloidal gear machining process is improved.

Description

Technical Field

[0001] This application relates to the technical field of cycloid gear machining control, and specifically to an efficient machining method and device for the cycloid gear of a speed reducer based on a machining tool. Background Art

[0002] The RV speed reducer is a high-precision speed reduction device. With advantages such as a compact structure, large torque, and high transmission efficiency, it is widely used in industrial machinery fields such as machine tools, robots, and automation equipment. Among them, the cycloid gear, as the core component of the RV speed reducer, its tooth profile accuracy (such as tooth profile shape, meshing clearance) has a crucial impact on the transmission efficiency, load-bearing capacity, and service life of the speed reducer. Although traditional machining methods (such as gear shaping, gear grinding) can meet certain accuracy requirements, in modern manufacturing requirements of high precision, large quantities, and low costs, they still show insufficient machining efficiency and lack of process stability. Therefore, optimizing the machining method of the cycloid gear will help accurately control the dimensional accuracy, shape and position accuracy, and surface roughness of the cycloid gear, thereby ensuring the stability and reliability of the RV speed reducer.

[0003] Since the cycloid gear belongs to a precision machined part, in order to improve the machining accuracy of the cycloid gear, existing research has proposed an adaptive machining control technology for the cycloid gear, which can adaptively adjust machining parameters based on data such as temperature, vibration, and current. However, the existing adaptive machining control technology for cycloid gears often focuses on optimizing machining parameters through the mean value or single index of machining data to achieve the purpose of improving machining accuracy. It does not fully consider that in the cycloid gear machining process where the cutting depth is constantly changing, the change of machining parameters is likely to cause chatter phenomenon, resulting in a decrease in the overall machining efficiency. Summary of the Invention

[0004] To solve the above technical problems, this application provides an efficient machining method and device for the cycloid gear of a speed reducer based on a machining tool. The specific technical solutions adopted are as follows: In the first aspect, an embodiment of this application provides an efficient machining method for the cycloid gear of a speed reducer based on a machining tool. The method includes the following steps: Obtain the machining data during the cycloid gear machining process and perform preprocessing; the machining data includes vibration data, cutting temperature, cutting force, cutting speed, cutting depth, and feed motor current data, and construct sequences in time order; Divide the cutting force into multiple subsequences by using the periodicity of the cutting depth; analyze the discrete degree of the similarity between any subsequence and the remaining subsequences, and the periodic fluctuations of the subsequences under different temperature changes, and construct the cutting disorder coefficient of the cutting force sequence to preliminarily judge whether chatter phenomenon occurs; Record the vibration, current, and cutting force data as target machining data; calculate the significance index of each peak based on the difference between the peak value and its respective average peak value in the target machining data and the local fluctuation characteristics of this difference, and cluster all the peaks based on this to extract the detected peaks; and combine the cutting disorder coefficient of the cutting force sequence, the range mean of the means of the internal elements of all clusters in all target machining data, and the distance mean between the detected peak position sequences of all target machining data to construct the chatter confidence level to evaluate the likelihood of chatter occurrence; Use the likelihood of chatter occurrence to classify the machining data through cluster analysis to screen out high-quality machining data and defective machining data; Adopt the dragonfly optimization algorithm to optimize the initial position distribution of the algorithm with high-quality machining data and defective machining data, and solve for the optimal cutting speed to be used as the cutting speed during the current cycloid gear machining process.

[0005] Preferably, the method for preliminarily determining whether chatter occurs is as follows: Obtain the period length of the cutting depth sequence; Use the period length to divide the cutting force into multiple cutting force subsequences, and calculate the degree of dispersion of the similarity between the first cutting force subsequence and all the other cutting force subsequences to determine the first degree of dispersion of the cutting force sequence; Divide the cutting temperature into two temperature subsequences through the sequence segmentation algorithm, and record them as the rising temperature subsequence and the stable temperature subsequence in order of position sequence respectively; Calculate the position sequence mean of each cutting force subsequence in the cutting force sequence, and record the cutting force subsequence whose position sequence mean belongs to the position sequence range of the rising temperature subsequence as the temperature-rising cutting force subsequence, and record the remaining cutting force subsequences as the temperature-stable cutting force subsequences; Sort the temperature-rising cutting force subsequences in ascending order according to their position sequence means, and calculate the difference between the cutting force means of two adjacent temperature-rising cutting force subsequences in turn. Count the number of negative elements among all the differences, and record the ratio between the number of negative elements and the number of all differences as the first ratio of the cutting force sequence; Obtain the degree of dispersion between the cutting force means of all the temperature-stable cutting force subsequences, which is recorded as the second degree of dispersion of the cutting force sequence; Use the first and second degrees of dispersion of the cutting force sequence and the first ratio of the cutting force sequence to construct the cutting disorder coefficient of the cutting force sequence to preliminarily determine whether chatter occurs.

[0006] Preferably, the period length is determined by the lag amount corresponding to the first peak among several autocorrelation coefficients of the cutting depth.

[0007] Preferably, the method for evaluating the likelihood of chatter occurrence is: For various target machining data, obtain all the peaks and valleys in the target machining data, and obtain the mean value between all the peaks; calculate the difference between each peak and the peak mean value, and record it as the first difference of each peak; Taking each peak as the center, calculate the degree of dispersion of all elements within the range of the previous nearest neighbor valley value and the next nearest neighbor valley value of each peak, and record it as the local fluctuation value of each peak; Calculate the product of the first difference of each peak and the local fluctuation value, and record it as the significant index of each peak; use the significant indices of all peaks as the input of the clustering algorithm for clustering; Calculate the mean value of the elements within each clustering cluster, and find the range of all the mean values, and record it as the first range of the target machining data; at the same time, record the peak corresponding to the elements within the clustering cluster with the largest mean value among all the clustering clusters as the detection peak; Obtain the ranks of all the detection peaks in the target machining data, and construct a rank sequence of the target machining data in ascending order of the ranks; Utilize the cutting disorder coefficient of the cutting force sequence, the mean value of the first ranges of all the target machining data, and the mean value of the distances between the rank sequences of all the target machining data to construct a flutter confidence level to evaluate the possibility of flutter occurrence.

[0008] Preferably, the method for classifying the machining data through cluster analysis to screen out high-quality machining data is as follows: Utilize the flutter confidence level, machining duration, and machining error in each machining process to construct an attribute vector of the machining data in each machining process; Perform clustering on the attribute vectors of the machining data in all the collected machining processes, and record the clustering cluster with the smallest mean value of the inner element modulus length as the high-quality clustering cluster, and record the clustering cluster with the smallest mean value of the inner element modulus length as the defective clustering cluster; Among them, the machining data corresponding to all the elements in the defective clustering cluster is defective machining data, and the machining data corresponding to all the elements in the high-quality clustering cluster is high-quality machining data.

[0009] Preferably, the method for using the dragonfly optimization algorithm to optimize the initial position distribution of the algorithm with high-quality machining data and defective machining data and solve the optimal cutting speed is as follows: Take the cutting speed in the high-quality machining data as the food position in the dragonfly optimization algorithm, and take the cutting speed in the defective machining data as the natural enemy position; During initialization, the dragonflies are first placed at the food position, and the remaining dragonflies are randomly distributed at non-natural enemy positions; Construct a global objective function: ; where Y is the value of the global objective function; is the function to obtain the minimum value; S is the modulus length of the attribute vector corresponding to the machining data; The number of dragonflies, the initial positions, the preset relevant parameters, the global objective function, and the cutting speed of the machining data during the current machining process are used as the inputs of the dragonfly optimization algorithm. The output of the dragonfly optimization algorithm is the optimal cutting speed that minimizes the value of the global objective function; Among them, the number of dragonflies is determined by the number of machining data obtained from the historical database.

[0010] Preferably, the calculation method of the cutting disorder coefficient is as follows: The ratio result of the sum value of the first discreteness degree and the second discreteness degree of the cutting force sequence to the first ratio of the cutting force sequence is used as the cutting disorder coefficient of the cutting force sequence.

[0011] Preferably, the calculation formula of the chatter confidence level is as follows: Denote the chatter confidence level of the u-th machining data as , ; In the formula, is the cutting disorder coefficient of the cutting force sequence during the u-th machining process; is the first range mean value of all target machining data in the u-th time; is the distance mean value between the rank sequences of all target machining data in the u-th time.

[0012] In a second aspect, another embodiment of the present application provides an efficient machining device for a reducer cycloid gear based on a machining tool. This device implements the efficient machining method for a reducer cycloid gear based on a machining tool as described above. This device includes: A data acquisition and preprocessing module, which is used to acquire vibration, temperature, cutting force, cutting speed, cutting depth, and current data during the cycloid gear machining process and perform preprocessing; A chatter detection module, which is used to analyze the periodic change of the cutting depth and the periodic fluctuation of the cutting force to detect the chatter phenomenon; A chatter confidence level analysis module, which is used to evaluate the possibility of chatter occurrence by combining multi-dimensional data; An attribute clustering module, which is used to classify the machining data; An optimization and solution module, which is used to optimize the parameters of the dragonfly optimization algorithm by using the classification result, and solve the optimal cutting speed through the dragonfly optimization algorithm to improve the machining efficiency of the current machining process and avoid chatter.

[0013] Preferably, the steps of data acquisition in the data acquisition and preprocessing module include: Arrange vibration sensors on the spindle of the machining tool to acquire vibration data during the machining process in real time; The cutting temperature, cutting force, cutting speed, cutting depth during the machining process, and the current data of the feed motor are obtained in real time through the central control system of the machining tool.

[0014] The present application has at least the following beneficial effects: By initially judging whether chatter occurs, the present application can reflect whether the cutting force remains stable and periodic under the periodic change of the cutting depth of the cycloid gear, so as to judge the stability of the machining process; by evaluating the possibility of chatter occurrence, the present application can reflect the coupling and synchronization between multi-dimensional machining data during the machining process, so as to accurately identify whether chatter occurs; by classifying the machining data using the possibility of chatter occurrence, the present application can distinguish the machining quality, thereby optimizing the relevant parameters of the optimal solution algorithm, and then improving the operation efficiency. In view of the problem that the prior art does not fully consider the change of the dynamic cutting depth of the cycloid gear, which may lead to chatter and reduce the machining efficiency, the present application evaluates the possibility of chatter occurrence and optimizes the initial position distribution in the algorithm through high-quality machining data, so as to more efficiently find the optimal cutting speed, and then maximize the overall machining efficiency during the current cycloid gear machining process while ensuring machining accuracy and avoiding chatter. Description of the Drawings

[0015] Figure 1 It is a flowchart of the steps of an efficient machining method for a reducer cycloid gear based on a machining tool provided by an embodiment of the present application; Figure 2 It is a schematic diagram of the modules of an efficient machining device for a reducer cycloid gear based on a machining tool provided by an embodiment of the present application. Detailed Description of the Embodiment

[0016] Embodiment 1 This embodiment provides a flowchart of the steps of an efficient machining method for a reducer cycloid gear based on a machining tool. As shown in the attached Figure 1 figure, the method includes the following steps: Step 1: Obtain the machining data during the cycloid gear machining process and perform preprocessing.

[0017] The data acquisition and preprocessing in the present application are carried out through the data acquisition and preprocessing module built in the efficient machining device for the reducer cycloid gear based on the machining tool, specifically as follows: A vibration sensor is arranged on the main shaft of the machining tool to collect the vibration data of the machining tool during the machining process in real time; the cutting temperature, cutting force, cutting speed, cutting depth during the cycloid gear machining process, and the current data of the feed motor are obtained in real time through the central control system of the machining tool.

[0018] All data is collected synchronously and in real time, with a collection frequency of 10HZ. The data collection duration is from the start of cycloid gear machining until the end of machining. After each machining, the machining accuracy (percentage data) of the machined cycloid gear is collected by the three-coordinate point-by-point scanning method.

[0019] According to the time sequence of data collection, vibration sequences, temperature sequences, cutting force sequences, cutting speed sequences, cutting depth sequences, and current sequences are constructed respectively, which are recorded as one-time machining data. Obtain the machining data of the cycloid gear N times from the historical database of the machining machine tool. In this embodiment, N takes the value of 100.

[0020] To eliminate the influence of dimension between data, all data is normalized. The normalization methods include Z-score normalization, sigmoid normalization, maximum-minimum normalization, etc. In this embodiment, maximum-minimum normalization is adopted.

[0021] Step 2: Initially judge whether chatter occurs.

[0022] Under normal circumstances, the faster the cutting speed, the higher the cutting efficiency at this time, and the shorter the machining duration. However, during high-speed cutting operations, small changes in machining parameters will also cause large fluctuations in the operating stress between the tool and the workpiece, which is likely to cause chatter, and then it is necessary to adjust the machining parameters again to ensure the stability of the machining rhythm, which will instead reduce the machining efficiency.

[0023] During the machining of the cycloid gear, since the tooth profile of the cycloid gear is multi-lobe meshing and the cutting depth changes periodically with the tooth phase, the cycloid gear is more likely to chatter compared with conventional machining parts, thus affecting the machining accuracy and efficiency.

[0024] This application analyzes the machining data through the chatter detection module built in the high-efficiency machining device of the reducer cycloid gear of the machining machine tool to initially detect whether chatter occurs, specifically as follows: During cutting, when the cutting speed is reasonable and no chatter is caused, due to the periodic change of the cutting depth, the cutting force should also show periodic fluctuations. However, the increase in temperature will cause the workpiece material to soften, resulting in a decrease in the average value of the cutting force, but the periodic change trend of the cutting force remains unchanged. As the temperature reaches a stable state, the average value of the cutting force also begins to stabilize. If chatter occurs due to too fast cutting speed, the change of the cutting force will no longer have periodic characteristics, and at the same time, it will also destroy the synchronous change between the cutting force and the temperature.

[0025] Accordingly, the present application obtains the period length of the cutting depth; uses the period length to divide the cutting force into multiple cutting force subsequences, and calculates the degree of dispersion of the similarity between the first cutting force subsequence and all the remaining cutting force subsequences to determine the first degree of dispersion of the cutting force sequence.

[0026] Preferably, as a preferred implementation manner, the period length is determined by the lag amount corresponding to the first peak among several autocorrelation coefficients of the cutting depth. In other implementation manners, the main frequency period of the cutting depth in the frequency domain can also be obtained through fast Fourier transform to determine its period length.

[0027] This embodiment takes the machining data of the u-th cycloid gear as an example for analysis: Taking the cutting depth sequence as the input of the autocorrelation coefficient, within the range of lag amount [1 - , the autocorrelation coefficients corresponding to each lag amount are output, and the lag amount corresponding to the first peak among all the autocorrelation coefficients is denoted as the period length of the cutting depth sequence. Wherein, N is the data length of the cutting depth sequence; t takes the value of 5 in this embodiment, indicating that there are at least 5 gears in the cycloid gear, and the implementer can adaptively take values according to the actual situation.

[0028] The cutting depth sequence is segmented into multiple depth subsequences through the period length; according to the position sequence range of each depth subsequence in the cutting depth sequence, multiple cutting force subsequences are extracted from the cutting force sequence. The similarity between the first cutting force subsequence and all the cutting force subsequences is calculated respectively, and the degree of dispersion among all the similarities is calculated, which is denoted as the first degree of dispersion of the cutting force sequence.

[0029] The calculation method of the similarity is not limited to the Pearson correlation coefficient, the Spearman correlation coefficient, and the cosine similarity. The Pearson correlation coefficient is adopted in this embodiment; the calculation method of the degree of dispersion is not limited to information entropy, variance, and coefficient of variation. Information entropy is adopted in this embodiment for calculation. The first degree of dispersion can reflect whether the data change trends between the first cutting force subsequence and the remaining each cutting force subsequence are relatively consistent, so as to reflect whether the change trend in the cutting force sequence conforms to periodicity.

[0030] Furthermore, analyze whether the change of the cutting force is synchronized with the temperature.

[0031] Since during the cutting process, the cutting temperature will first gradually increase and then tend to be stable after reaching a certain temperature, showing two obvious data trends.

[0032] Accordingly, in this application, the cutting temperature is segmented into two temperature subsequences through a sequence segmentation algorithm, and are sequentially recorded as the rising temperature subsequence and the stable temperature subsequence according to the order. The sequence segmentation algorithm is not limited to the BG segmentation algorithm and the MK segmentation algorithm. In this embodiment, the MK sequence segmentation algorithm is adopted.

[0033] Calculate the order mean of each cutting force subsequence in the cutting force sequence, and record the cutting force subsequence whose order mean belongs to the order range of the rising temperature subsequence as the temperature-rise cutting force subsequence, and record the remaining cutting force subsequences as the temperature-stable cutting force subsequences.

[0034] Sort the temperature-rise cutting force subsequences in ascending order according to their order means, and sequentially calculate the difference between the cutting force means of the two adjacent temperature-rise cutting force subsequences before and after. Count the number of negative elements among all the differences, and record the ratio between the number of negative elements and the number of all differences as the first ratio of the cutting force sequence.

[0035] Obtain the degree of dispersion between the cutting force means of all the temperature-stable cutting force subsequences, which is recorded as the second degree of dispersion.

[0036] Accordingly, this application constructs a cutting disorder coefficient of the cutting force sequence by using the first and second degrees of dispersion of the cutting force sequence and the first ratio of the cutting force sequence, so as to preliminarily judge whether a chatter phenomenon occurs.

[0037] In this embodiment, the cutting disorder coefficient of the cutting force sequence in the u-th machining process is constructed The calculation expression is: In the formula, is the first degree of dispersion of the cutting force sequence in the u-th machining process; is the second degree of dispersion of the cutting force sequence in the u-th machining process; is the first ratio of the cutting force sequence in the u-th machining process.

[0038] It should be understood that the first degree of dispersion can reflect whether the change of the cutting force conforms to the periodic trend when the cycloid gear is machined at the current cutting speed under the cutting depth with periodic change characteristics. The larger the value, the worse the reflection of the periodic trend. The second degree of dispersion can reflect the fluctuation difference of the cutting force in different cutting cycles when the cutting temperature is stable. The larger the value, the greater the violent fluctuation of the cutting force. The first ratio can reflect whether the cutting force mean has a continuous downward trend in the temperature-rise stage, so as to reflect whether the cutting change conforms to the physical law of material softening. The cutting disorder coefficient It can comprehensively reflect the overall stability of the cutting force during the cycloid gear machining at the current cutting speed. The larger the value, the more likely it is that the current cutting speed has exceeded the most appropriate speed range and the more likely chatter has occurred.

[0039] Step 3: Evaluate the possibility of chatter occurrence.

[0040] Furthermore, when chatter does not occur, the machine tool vibration is mainly concentrated on the tooth passing frequency and its harmonics, and the amplitude of the vibration data is small and the fluctuation is relatively stable; while if the current cutting speed is inappropriate and chatter occurs, it will lead to abnormal mutation peaks in the vibration data and a significant increase in the variance of the vibration data; at the same time, when chatter occurs, the variance of the cutting force fluctuation and the current in the feed motor will also mutate. Therefore, it is possible to further analyze whether chatter has occurred and evaluate the possibility of chatter occurrence by combining the vibration, current, and cutting force data during the machining process.

[0041] Accordingly, in this application, the vibration, current, and cutting force data are all recorded as target machining data; for various target machining data, all the peaks and valleys in the target machining data are obtained, and the mean value between all the peaks is obtained; the difference between each peak and the peak mean value is calculated and recorded as the first difference of each peak; centered on each peak, the dispersion degree of all elements within the range of the previous nearest neighbor valley value and the next nearest neighbor valley value of each peak is calculated and recorded as the local fluctuation value of each peak; the product of the first difference of each peak and the local fluctuation value is calculated and recorded as the significant index of each peak; the significant indices of all the peaks are used as the input of the clustering algorithm for clustering; the mean value of the elements within each clustering cluster is calculated, and the range of all the mean values is obtained and recorded as the first range of the target machining data; at the same time, the peak corresponding to the elements within the clustering cluster with the largest mean value among all the clustering clusters is recorded as the detection peak; the ordinal positions of all the detection peaks in the target machining data are obtained, and an ordinal sequence of the target machining data is constructed in ascending order of the ordinal positions; the chatter confidence is constructed by using the cutting disorder coefficient of the cutting force sequence, the mean value of the first ranges of all the target machining data, and the mean value of the distances between the ordinal sequences of all the target machining data to evaluate the possibility of chatter occurrence.

[0042] In this embodiment, the following analysis is carried out taking the vibration data as an example: All the peaks and valleys in the vibration sequence are obtained through the peak-valley value detection algorithm, and the mean value between all the peaks is obtained. The difference between each peak and the peak mean value is calculated and recorded as the first difference of each peak. It should be noted that the first difference can be positive or negative.

[0043] Taking each peak as the center, calculate the degree of dispersion between all elements within the range of the nearest neighbor valley value before each peak and the nearest neighbor valley value after each peak, and record it as the local fluctuation value of each peak. The local fluctuation value can reflect the degree of mutation of each peak in the local range. It should be noted that: when the peak is near the two ends of the sequence and there is no valley value on the front or back side, calculate the degree of dispersion between the peak and all elements within the range of the first or last element of the sequence.

[0044] The product of the first difference of each peak value and the local fluctuation value is calculated and recorded as the significance index of each peak value. The significance index of all peak values ​​is used as the input of the clustering algorithm for clustering. The clustering algorithm is not limited to the k-means algorithm, the DPC algorithm, and the DBSCAN algorithm. This embodiment uses the K-means algorithm for clustering, obtains the optimal number of clustering clusters through the elbow rule, and outputs multiple clustering clusters.

[0045] Calculate the mean of the internal elements of each cluster, and find the range of all means, which is recorded as the first range of the vibration sequence; at the same time, record the peak value corresponding to the internal element of the cluster with the largest mean among all clusters as the detection peak value.

[0046] It should be understood that the significance index can reflect the degree of deviation of each vibration peak value compared to the vibration mean and whether the local fluctuation is severe; the clustering algorithm can cluster all peak values, and extract a group of vibration peak values ​​with the largest deviation from the cluster mean and the most severe local fluctuation by comparing the cluster means, thereby improving subsequent detection accuracy and efficiency.

[0047] The position sequence of all detected peaks in the vibration sequence is obtained, and the position sequence of the vibration sequence is constructed according to the order of the position sequence from small to large. In the same way, the position sequence and the first extreme difference of the current sequence and the cutting force sequence are obtained respectively.

[0048] In this embodiment, the chatter confidence of the u-th processing data is constructed The calculation expression is: In the formula, is the cutting turbulence coefficient of the cutting force sequence in the u-th machining process; is the first range mean of all target processing data for the uth time; is the mean distance between the bit sequence sequences of all target processing data for the uth time, expressed as the mean DTW distance between the three bit sequence sequences of vibration sequence, current sequence, and cutting force sequence. It should be noted that the DTW distance is used here to take into account the possible inconsistency in the data length of the bit sequence.

[0049] It should be understood that the first range mean Data fluctuations that can reflect whether abnormal data fluctuations have occurred in the vibration data, current data, and cutting force data in the target machining data at the u-th time; the distance from the mean value It can reflect whether the occurrence times of abnormal peaks among the three types of target machining data of vibration, current, and cutting force are relatively consistent. Chatter confidence level It can combine multi-dimensional data factors to comprehensively reflect whether chatter occurs during the u-th machining process when cutting at the current cutting speed. If the value is smaller, it indicates that chatter does not occur, and the cutting speed of the machining tool can still be considered to be increased to improve the machining efficiency.

[0050] Step 4: Screen out high-quality machining data and defective machining data.

[0051] Furthermore, after obtaining the chatter confidence level of the machining data, the cutting speed that minimizes the machining time and has good machining accuracy can be obtained by combining the required cutting duration and cutting accuracy at each cutting speed.

[0052] This application differentiates the machining data through the attribute clustering module built in the high-efficiency machining device for the cycloid gear of the reducer of the machining tool as follows: Obtain the chatter confidence level of each machining data; at the same time, take the data length of the cutting force sequence corresponding to each machining data as the machining duration of each machining data, and record the difference between 1 and the machining accuracy (percentage) as the machining error of each machining data.

[0053] To eliminate the influence of the dimension between data, the chatter confidence level, machining duration, and machining error in all collected machining processes are normalized respectively, and the attribute vector of each machining data is constructed according to the normalized data, that is, [chatter confidence level, machining duration, machining error]; the attribute vectors of the machining data in all collected machining processes are used as the input of the clustering algorithm, and multiple machining clustering clusters are output.

[0054] Calculate the mean of the inner element modulus lengths of each machining clustering cluster respectively. The machining clustering cluster with the largest mean is recorded as the defective clustering cluster, and the machining clustering cluster with the smallest mean is recorded as the high-quality clustering cluster. Correspondingly, the machining data corresponding to all elements in the defective clustering cluster is defective machining data, and the machining data corresponding to all elements in the high-quality clustering cluster is high-quality machining data.

[0055] Through the clustering algorithm, different machining qualities can be differentiated, which is convenient for finding the optimal cutting speed.

[0056] Step 5: Solve the optimal cutting speed.

[0057] This application obtains the optimal solution through the optimization solution module built in the high-efficiency processing device for the cycloid gear of the reducer based on the processing machine tool, specifically as follows: To further optimize the cutting speed, this application uses the dragonfly optimization algorithm for optimal solution. In the dragonfly optimization algorithm, the distribution of the initial positions has an important impact on the convergence speed and calculation accuracy of the algorithm. A good initial distribution can accelerate the search for the global optimal solution.

[0058] For this reason, this application takes the cutting speed in the high-quality processing data as the food position, that is, the local optimal solution, in the dragonfly optimization algorithm, and takes the cutting speed in the defective processing data as the natural enemy position, that is, the local worst solution. The number of dragonflies is determined by the number of processing data N obtained from the historical database.

[0059] During initialization, the dragonflies are first placed at the food position, and the remaining dragonflies are randomly distributed at non-natural enemy positions.

[0060] Construct the global objective function: . In the formula, Y is the value of the global objective function; is the function to take the minimum value; S is the modulus length of the attribute vector of the corresponding processing data.

[0061] In this embodiment, the relevant parameters of the preset dragonfly optimization algorithm are: separation weight is 0.1, alignment weight is 0.1, aggregation weight is 0.7, food factor is 1, natural enemy weight is 1, the inertia weight is initially 0.9 and linearly decreases to 0.4 with the number of iterations, and the number of iterations is 50.

[0062] Take the number of dragonflies, the initial positions, the preset relevant parameters, the global objective function, and the cutting speed of the processing data in the current processing as the input of the dragonfly optimization algorithm. The output of the dragonfly optimization algorithm is the optimal cutting speed that minimizes the value of the global objective function.

[0063] So far, on the premise of not causing chatter and ensuring the machining accuracy, the optimal cutting speed output is used as the cutting speed in the current cycloid gear machining process, so that the machining time of the current cycloid gear is the shortest and the machining efficiency is improved.

[0064] Embodiment 2 The module schematic diagram of the high-efficiency processing device for the cycloid gear of the reducer based on the processing machine tool provided in this embodiment is as shown in the appendix Figure 2 shown. To implement the above-mentioned high-efficiency processing method for the cycloid gear of the reducer based on the processing machine tool, the device is built-in with multiple modules, specifically as follows: Data acquisition and preprocessing module, which is used to collect vibration, temperature, cutting force, cutting speed, cutting depth and current data during the cycloid gear machining process and perform preprocessing; The chatter detection module is used to analyze the periodic changes in cutting depth and the periodic fluctuations in cutting force to detect chatter phenomena; The chatter confidence analysis module is used to evaluate the possibility of chatter occurrence by combining multi-dimensional data; The attribute clustering module is used to classify the machining data; The optimization and solution module is used to optimize the parameters of the dragonfly optimization algorithm using the classification results, and solve for the optimal cutting speed through the dragonfly optimization algorithm to improve the machining efficiency of the current machining process and avoid chatter.

[0065] Those skilled in the art will readily conceive of other embodiments of the present application after considering the specification and practicing the invention herein. The present application is intended to cover any variations, uses, or adaptations of the present application, which follow the general principles of the present application and include known common knowledge or conventional technical means in the technical field not invented by the present application.

[0066] It should be understood that the present application is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. An efficient machining method for the cycloid gear of a speed reducer based on a machining tool, characterized in that The method includes: Obtaining the processing data during the cycloid gear machining process and performing preprocessing; the processing data includes vibration data, cutting temperature, cutting force, cutting speed, cutting depth, and feed motor current data, and constructing sequences in order according to time; Dividing the cutting force into multiple subsequences by using the periodicity of the cutting depth; analyzing the discrete degree of the similarity between any subsequence and the remaining subsequences, as well as the periodic fluctuations of the subsequences under different temperature changes, and constructing the cutting disorder coefficient of the cutting force sequence to preliminarily determine whether chatter occurs; Denoting the vibration, current, and cutting force data as target processing data; calculating the significant index of each peak according to the difference between the peak value and its respective average peak value in the target processing data and the local fluctuation characteristics of the difference, and clustering all the peaks with this to extract the detected peaks; and combining the cutting disorder coefficient of the cutting force sequence, the range mean of the means of the elements within all clustering clusters in all target processing data, and the distance mean between the detected peak position sequences of all target processing data to construct the chatter confidence level to evaluate the possibility of chatter occurrence; Using the possibility of chatter occurrence to classify the processing data through cluster analysis to screen out high-quality processing data and defective processing data; Adopting the dragonfly optimization algorithm to optimize the initial position distribution of the algorithm with the high-quality processing data and defective processing data, and solving for the optimal cutting speed to be used as the cutting speed during the current cycloid gear machining process.

2. The high-efficiency machining method of the cycloid gear of the speed reducer based on the machining tool according to claim 1, characterized in that The method for preliminarily determining whether chatter occurs is as follows: Obtaining the period length of the cutting depth sequence; Dividing the cutting force into multiple cutting force subsequences by using the period length, and calculating the discrete degree of the similarity between the first cutting force subsequence and all the remaining cutting force subsequences to determine the first discrete degree of the cutting force sequence; Dividing the cutting temperature into two temperature subsequences by using the sequence segmentation algorithm, and respectively denoting them as the rising temperature subsequence and the stable temperature subsequence in order according to the position sequence; Calculating the position sequence mean of each cutting force subsequence in the cutting force sequence, and denoting the cutting force subsequences whose position sequence means belong to the position sequence range of the rising temperature subsequence as the temperature-rising cutting force subsequences, and denoting the remaining cutting force subsequences as the temperature-stable cutting force subsequences; Sorting the temperature-rising cutting force subsequences in ascending order according to their position sequence means, and successively calculating the difference between the cutting force means of the two adjacent temperature-rising cutting force subsequences, counting the number of negative elements among all the differences, and denoting the ratio between the number of negative elements and the number of all differences as the first ratio of the cutting force sequence; Obtaining the discrete degree between the cutting force means of all the temperature-stable cutting force subsequences, and denoting it as the second discrete degree of the cutting force sequence; Using the first and second discrete degrees of the cutting force sequence and the first ratio of the cutting force sequence to construct the cutting disorder coefficient of the cutting force sequence to preliminarily determine whether chatter occurs.

3. The high-efficiency machining method of the cycloid gear of the speed reducer based on the machining tool according to claim 2, characterized in that The period length is determined by the lag amount corresponding to the first peak among several autocorrelation coefficients of the cutting depth.

4. The high-efficiency machining method of the cycloid gear of the speed reducer based on the machining tool according to claim 1, wherein, The method for evaluating the possibility of chatter occurrence is as follows: For various target machining data, all the peaks and valleys in the target machining data are obtained, and the average value between all the peaks is obtained; the difference between each peak and the peak average value is calculated and denoted as the first difference of each peak; Centered on each peak, the degree of dispersion between all elements within the range of the previous nearest neighbor valley value and the next nearest neighbor valley value of each peak is calculated and denoted as the local fluctuation value of each peak; The product of the first difference of each peak and the local fluctuation value is calculated and denoted as the significant index of each peak; the significant indices of all peaks are used as the input of the clustering algorithm for clustering; The average value of the elements within each clustering cluster is calculated, and the range of all the average values is obtained and denoted as the first range of the target machining data; meanwhile, the peak corresponding to the elements within the clustering cluster with the largest average value among all the clustering clusters is denoted as the detected peak; The ordinal positions of all the detected peaks in the target machining data are obtained, and an ordinal sequence of the target machining data is constructed in ascending order of the ordinal positions; Using the cutting disorder coefficient of the cutting force sequence, the average value of the first ranges of all the target machining data, and the average distance between the ordinal sequences of all the target machining data, a chatter confidence level is constructed to evaluate the possibility of chatter occurrence.

5. The high-efficiency processing method of the cycloid gear of the speed reducer based on the processing machine tool according to claim 3, characterized in that, The method for classifying machining data through cluster analysis to screen out high-quality machining data is as follows: Using the chatter confidence level, machining duration, and machining error in each machining process, an attribute vector of the machining data in each machining process is constructed; The attribute vectors of the machining data in all the collected machining processes are clustered, and the clustering cluster with the smallest average modulus length of the internal elements is denoted as the high-quality clustering cluster, and the clustering cluster with the smallest average modulus length of the internal elements is denoted as the defective clustering cluster; Among them, the machining data corresponding to all the elements in the defective clustering cluster is defective machining data, and the machining data corresponding to all the elements in the high-quality clustering cluster is high-quality machining data.

6. The high-efficiency machining method of the cycloid gear of the speed reducer based on the machining tool according to claim 1, characterized in that The method for using the dragonfly optimization algorithm to optimize the initial position distribution of the algorithm with high-quality machining data and defective machining data to solve the optimal cutting speed is as follows: The cutting speed in the high-quality machining data is used as the food position in the dragonfly optimization algorithm, and the cutting speed in the defective machining data is used as the natural enemy position; During initialization, the dragonflies are first placed at the food position, and the remaining dragonflies are randomly distributed at non-natural enemy positions; Construct the global objective function: ; where Y is the value of the global objective function; is the minimum value function; S is the norm of the attribute vector of the corresponding processing data; The number of dragonflies, the initial positions, preset relevant parameters, the global objective function, and the cutting speed of the machining data in the current machining process are used as the input of the dragonfly optimization algorithm, and the output of the dragonfly optimization algorithm is the optimal cutting speed that minimizes the global objective function value; Among them, the number of dragonflies is determined by the number of machining data obtained from the historical database.

7. The high-efficiency machining method of the cycloid gear of the reducer based on the machining tool according to claim 2, characterized in that, The calculation method of the cutting disorder coefficient is as follows: The ratio result of the sum value of the first degree of dispersion and the second degree of dispersion of the cutting force sequence to the first ratio of the cutting force sequence is used as the cutting disorder coefficient of the cutting force sequence.

8. The high-efficiency machining method of the cycloid gear of the speed reducer based on the machining tool according to claim 4, characterized in that, The calculation formula of the chatter confidence level is as follows: Denote the chatter confidence of the $u$-th machining data as , ; In the formula, is the cutting disorder coefficient of the cutting force sequence in the u-th machining process; is the first range mean of all target machining data in the u-th time; is the mean distance between the rank sequences of all target machining data in the u-th time.

9. An efficient machining device for the cycloid gear of a speed reducer based on a machining tool, characterized in that, The device implements the high-efficiency machining method for the cycloid gear of the reducer based on the machining tool as described in any one of claims 1-8. The device includes: The data acquisition and preprocessing module is used to acquire vibration, temperature, cutting force, cutting speed, cutting depth, and current data during the cycloid gear machining process and perform preprocessing; The chatter detection module is used to analyze the periodic changes in cutting depth and the periodic fluctuations in cutting force to detect chatter phenomena; The chatter confidence analysis module is used to evaluate the possibility of chatter occurrence by combining multi-dimensional data; The attribute clustering module is used to classify the machining data; The optimization solution module is used to optimize the parameters of the dragonfly optimization algorithm using the classification results and solve for the optimal cutting speed through the dragonfly optimization algorithm to improve the machining efficiency of the current machining process and avoid chatter.

10. The high-efficiency machining device for the cycloid gear of a speed reducer based on a machining tool according to claim 9, characterized in that, The steps of data acquisition in the data acquisition and preprocessing module include: Arranging vibration sensors on the spindle of the machining tool to acquire vibration data during the machining process in real time; Obtaining the cutting temperature, cutting force, cutting speed, cutting depth, and current data of the feed motor during the machining process in real time through the central control system of the machining tool.

Images