A Method and System for Analyzing the State of a Five-Axis Machine Tool Spindle Based on Multidimensional Data

Abstract

This application belongs to the technical field of machine tool safety detection, and particularly relates to a method and system for analyzing the state of a five-axis machine tool spindle based on multi-dimensional data. The method includes: collecting the state data of the spindle in real time through multiple sensors; extracting features from the preprocessed state data to obtain multi-dimensional feature data; determining the health state of the spindle based on the multi-dimensional feature data through a machine learning model; predicting the impact data on workpiece forming caused by the health state of the spindle according to empirical data, and determining the type of defects occurring during the workpiece forming process based on the impact data through a regression model; generating an adjustment strategy according to the type of defects; generating an adjustment instruction according to the adjustment strategy, and performing workpiece processing according to the adjustment instruction. This application can comprehensively evaluate the health state of the spindle, avoid workpiece forming defects caused by spindle failures, improve the quality of workpiece processing, and reduce downtime and production losses.

Description

Technical Field

[0001] This application belongs to the technical field of machine tool safety detection, and particularly relates to a method and system for analyzing the spindle state of a five-axis machine tool based on multi-dimensional data. Background Art

[0002] With the development of technology, the demand for ultra-precision machining of complex-shaped parts is increasing, and increasing the number of machine tool axes has become a development trend of today's ultra-precision machine tools. If ultra-precision cutting of complex components such as off-axis free-form surfaces and non-rotationally symmetric surfaces is required, a multi-axis ultra-precision machine tool is needed. A five-axis machine tool is a high-precision machining device with five motion axes, which can simultaneously control the movement of the tool on the linear and rotational axes, thereby achieving high-precision machining of complex workpieces. As the core component of a five-axis machine tool, the state of the spindle directly affects the machining efficiency and quality. A high-performance spindle can provide stable rotational speed, power, and torque, ensuring the stability and accuracy of the machining process. If the spindle fails or its performance deteriorates, it will lead to a decrease in machining efficiency and quality, and even damage the workpiece or tool.

[0003] The existing methods for monitoring the spindle state usually include two types. One is to collect the spindle operation data through sensors and set thresholds or rules to judge the spindle state. Although this method is simple to implement, it cannot effectively capture the influence of the coupling effect between multiple factors on the spindle state. The other is to achieve intelligent monitoring of the spindle state by training a model. This method can handle more complex multi-dimensional feature relationships, but it has a strong dependence on high-quality labeled data. In addition, neither of the above two spindle state monitoring methods can accurately predict the influence of the spindle state on the subsequent workpiece forming. This limitation makes it difficult for the existing technology to provide forward-looking guidance for process optimization and preventive maintenance, restricting the improvement of the intelligent manufacturing level.

[0004] Therefore, it is desirable to have a technical solution to overcome or mitigate at least one of the above defects of the existing technology. Summary of the Invention

[0005] The purpose of this application is to provide a method and system for analyzing the spindle state of a five-axis machine tool based on multi-dimensional data to solve at least one problem existing in the prior art.

[0006] The technical solution of this application is as follows:

[0007] The first aspect of this application provides a method for analyzing the spindle state of a five-axis machine tool based on multi-dimensional data, including:

[0008] S100. Real-time collect the state data of the spindle through multiple sensors;

[0009] S200. Extract multi-dimensional feature data from the pre-processed state data;

[0010] S300. Determine the health state of the main shaft based on the multi-dimensional feature data through a machine learning model;

[0011] S400. Predict the impact data on workpiece forming caused by the health state of the main shaft according to empirical data, and determine the defect type occurring during the workpiece forming process based on the impact data through a regression model;

[0012] S500. Generate an adjustment strategy according to the defect type;

[0013] S600. Generate an adjustment instruction according to the adjustment strategy, and perform workpiece processing according to the adjustment instruction.

[0014] In at least one embodiment of the present application, the state data includes vibration data, temperature data, torque data, and rotational speed data.

[0015] In at least one embodiment of the present application, in S200, extracting multi-dimensional feature data from the pre-processed state data includes:

[0016] S210. Sort the state data according to the time stamp and synchronize the data through interpolation;

[0017] S220. Pre-process the state data, and the pre-processing methods include at least cleaning, denoising, and filtering;

[0018] S230. Extract multi-dimensional feature data related to the health state of the main shaft from the state data;

[0019] S240. Fuse the multi-dimensional feature data through the principal component analysis method.

[0020] In at least one embodiment of the present application, in S230, extracting multi-dimensional feature data related to the health state of the main shaft from the state data includes:

[0021] Extract vibration feature data related to the health state of the main shaft from the vibration data, including vibration amplitude and vibration frequency;

[0022] Extract temperature feature data related to the health state of the main shaft from the temperature data, including temperature change rate and temperature change trend;

[0023] Extract torque feature data related to the health state of the main shaft from the torque data, including load fluctuation amplitude and load fluctuation frequency;

[0024] Extract the rotational speed characteristic data related to the health state of the main shaft from the rotational speed data, including the amplitude of rotational speed fluctuation.

[0025] In at least one embodiment of the present application, in S240, the multi-dimensional feature data is fused by the principal component analysis method, including:

[0026] S241. Perform standardization processing on the multi-dimensional feature data, including:

[0027] Calculate the mean value of the multi-dimensional feature data ;

[0028] Calculate the standard deviation of the multi-dimensional feature data ;

[0029] Standardize the multi-dimensional feature data through the mean value and the standard deviation:

[0030] ;

[0031] Wherein, is the i th multi-dimensional feature data in the dataset, is the mean value of the multi-dimensional feature data, is the standard deviation of the multi-dimensional feature data, is the data point obtained by standardizing the i th multi-dimensional feature data in the dataset;

[0032] S242. Construct a covariance matrix, and solve the eigenvalues and eigenvectors, including:

[0033] Construct a covariance matrix:

[0034] ;

[0035] Wherein, Z is the multi-dimensional feature matrix composed of data points , n is the number of samples, is Z transpose matrix, is the covariance matrix;

[0036] Construct a characteristic polynomial according to the covariance matrix:

[0037] ;

[0038] Wherein, I is the identity matrix, is the characteristic polynomial;

[0039] By solving the characteristic polynomial , all eigenvalues , m is the total number of eigenvalues;

[0040] For each eigenvalue , by solving the system of equations , the corresponding eigenvector is obtained , k = 1, 2, 3, …, m ;

[0041] S243. Sort the eigenvalues from largest to smallest, and select a preset number of principal components in the front;

[0042] S244. Project the multi-dimensional feature data onto the selected several principal components to obtain the multi-dimensional feature data after dimensionality reduction.

[0043] In at least one embodiment of the present application, in S300, based on the multi-dimensional feature data, determining the health status of the main shaft through a machine learning model includes:

[0044] S310. Divide the multi-dimensional feature data into a training set, a validation set, and a test set;

[0045] S320. Build a fully connected neural network model using the deep learning framework Keras;

[0046] S330. Train the fully connected neural network model through the training set, select the model and tune the hyperparameters through the validation set, and evaluate the fully connected neural network model through the test set;

[0047] S340. Determine the health status of the main shaft through the fully connected neural network model, and the types of the health status of the main shaft include healthy, slightly abnormal, and severely abnormal.

[0048] In at least one embodiment of the present application, in S400, predicting the influence data of the health status of the main shaft on the workpiece forming according to empirical data, and based on the influence data, determining the type of defect occurring during the workpiece forming process through a regression model includes:

[0049] S410. Predict the influence data of the health status of the main shaft on the workpiece forming according to empirical data, and the influence data includes a health status score and various processing parameters;

[0050] S420. Input the health status score and various processing parameters into the regression model to calculate the workpiece quality:

[0051] ;

[0052] where Y is the workpiece quality, is the intercept, is the corresponding regression coefficient, is the health status score or the processing parameter;

[0053] S430. Determine the type of defect that occurs during the workpiece forming process according to the workpiece quality.

[0054] In at least one embodiment of the present application, the processing parameters at least include cutting speed, feed rate, cutting depth, tool wear amount, and coolant state.

[0055] In at least one embodiment of the present application, in S500, the adjustment strategy generated according to the type of defect includes:

[0056] Adjust the spindle speed or the feed speed;

[0057] Adjust the spindle load;

[0058] Optimize the cooling system or replace the lubricating oil;

[0059] Replace or repair the faulty components.

[0060] The second aspect of the present application provides a five-axis machine tool spindle state analysis system based on multi-dimensional data. Based on the above-mentioned five-axis machine tool spindle state analysis method based on multi-dimensional data, it includes:

[0061] A data acquisition module for real-time collecting the state data of the spindle through a variety of sensors;

[0062] A data processing module for extracting features from the preprocessed state data to obtain multi-dimensional feature data;

[0063] A state analysis module for determining the spindle health status based on the multi-dimensional feature data through a machine learning model;

[0064] A defect prediction module for predicting the influence data of the spindle health status on the workpiece forming according to the empirical data, and determining the type of defect that occurs during the workpiece forming process based on the influence data through a regression model;

[0065] An adjustment module for generating an adjustment strategy according to the type of defect;

[0066] An output module for generating an adjustment instruction according to the adjustment strategy and performing workpiece processing according to the adjustment instruction.

[0067] The invention has at least the following beneficial technical effects:

[0068] The five-axis machine tool spindle status analysis method based on multi-dimensional data of the present application can comprehensively evaluate the spindle health status and detect potential problems in advance by using a machine learning model to monitor and analyze multi-dimensional feature data in real time; predict the impact data of the spindle health status on the subsequent workpiece forming according to empirical data, and identify the defect types in combination with a regression model to avoid workpiece forming defects caused by spindle failures; avoid machining errors caused by spindle abnormalities through early adjustment, improve the workpiece machining quality, reduce downtime and production losses; real-time monitoring and intelligent adjustment help to reduce the working load of the spindle, reduce spindle failures, and extend the service life of the spindle. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 is a flowchart of the five-axis machine tool spindle status analysis method based on multi-dimensional data according to an embodiment of the present application;

[0070] Figure 2 is a schematic diagram of the five-axis machine tool spindle status analysis system based on multi-dimensional data according to an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0071] To make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described in more detail below with reference to the accompanying drawings in the embodiments of the present application. In the drawings, the same or similar reference numerals denote the same or similar elements or elements with the same or similar functions throughout. The described embodiments are some, but not all, of the embodiments of the present application. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to explain the present application and should not be construed as limiting the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application. The embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0072] In the description of the present application, it should be understood that the orientation or positional relationships indicated by the terms "center", "longitudinal", "transverse", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. are based on the orientation or positional relationships shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as limiting the protection scope of the present application.

[0073] The following will be further described in detail with reference to the attached Figures 1 to 2 to the present application.

[0074] The first aspect of the present application provides a five-axis machine tool spindle status analysis method based on multi-dimensional data, such asFigure 1 As shown, it includes the following steps:

[0075] S100. Real-time collect the state data of the spindle through multiple sensors;

[0076] S200. Extract features from the preprocessed state data to obtain multi-dimensional feature data;

[0077] S300. Based on the multi-dimensional feature data, determine the health state of the spindle through a machine learning model;

[0078] S400. Predict the influence data on workpiece forming caused by the health state of the spindle according to empirical data, and based on the influence data, determine the defect types occurring during the workpiece forming process through a regression model;

[0079] S500. Generate an adjustment strategy according to the defect type;

[0080] S600. Generate an adjustment instruction according to the adjustment strategy, and perform workpiece processing according to the adjustment instruction.

[0081] For the five-axis machine tool spindle state analysis method based on multi-dimensional data of the present application, first, in S100, state data such as vibration data, temperature data, torque data, and rotational speed data of the spindle are real-time collected through multiple sensors. Among them, the vibration amplitude and frequency of the spindle can be monitored through a vibration sensor to determine whether there is an abnormal vibration mode; the temperature of the spindle can be monitored through a temperature sensor to prevent damage caused by overheating; the torque sensor can measure the torque of the spindle to analyze the change of the spindle load; the rotational speed sensor is used to real-time monitor the rotational speed of the spindle.

[0082] For the five-axis machine tool spindle state analysis method based on multi-dimensional data of the present application, in S200, the collected state data is preprocessed and feature-extracted to extract key features related to the health state of the spindle. Different sensors usually collect different time series data. To ensure the time series consistency among the multi-dimensional feature data, these data need to be time-synchronized, and each sensor's collected data is attached with a timestamp. In the preferred embodiment of the present application, in S200, multi-dimensional feature data is obtained by extracting features from the preprocessed state data, including:

[0083] S210. Sort the state data according to the timestamp and perform data synchronization through interpolation;

[0084] Sort the status data of all sensors according to the time stamps, and align all data sequences at a unified time step through interpolation. During the data synchronization process, if data is missing at certain time points (such as a sensor fails or is lost), the missing data needs to be repaired through data filling (such as interpolation method based on surrounding valid data) to ensure data continuity.

[0085] S220. Preprocess the status data, and the preprocessing methods include at least cleaning, denoising, and filtering;

[0086] By preprocessing the synchronized status data, noise and unnecessary high-frequency interference are removed to improve the accuracy of subsequent analysis.

[0087] S230. Extract multi-dimensional feature data related to the health status of the main shaft from the status data;

[0088] In this embodiment, the extraction of multi-dimensional feature data mainly includes vibration feature extraction, temperature feature extraction, torque feature extraction, and rotational speed feature extraction. Specifically:

[0089] Vibration feature extraction: Extract vibration feature data related to the health status of the main shaft from the vibration data, including vibration amplitude and vibration frequency;

[0090] Temperature feature extraction: Extract temperature feature data related to the health status of the main shaft from the temperature data, including temperature change rate and temperature change trend;

[0091] Torque feature extraction: Extract torque feature data related to the health status of the main shaft from the torque data, including load fluctuation amplitude and load fluctuation frequency;

[0092] Rotational speed feature extraction: Extract rotational speed feature data related to the health status of the main shaft from the rotational speed data, including rotational speed fluctuation amplitude.

[0093] Among them, regarding vibration feature extraction, in addition to vibration amplitude and vibration frequency, the features of the extracted vibration signal can also include statistical features in the time domain and frequency domain, as well as kurtosis. Time domain features are some statistical features directly extracted from the original vibration data, such as mean, variance, skewness, kurtosis, etc., which reflect the volatility and asymmetry of the vibration signal and can identify irregular vibrations in the system; frequency domain features are obtained by converting the vibration signal from the time domain to the frequency domain through Fourier transform to analyze the spectral features of the vibration, including the main frequency, frequency bandwidth, and power spectral density; kurtosis can detect abnormal fluctuations during the operation of the main shaft by calculating the sharpness of the vibration signal. Regarding temperature feature extraction, the key temperature features include the rate of temperature change and the long-term temperature change trend. The rate of change of the main shaft temperature reflects the change in the main shaft load. Monitoring the long-term trend of the temperature, if the temperature continues to rise, it may indicate potential problems with the main shaft. Regarding torque feature extraction, by analyzing the amplitude and frequency of the load fluctuation, the load fluctuation amplitude and load fluctuation frequency features can be extracted. A relatively large load fluctuation amplitude may mean problems such as tool wear, uneven workpiece material, or unstable fixture. For the frequency components of the load fluctuation, if abnormal frequencies appear, it may indicate a fault in the main shaft or external interference. Regarding rotational speed feature extraction, the rotational speed feature mainly includes the amplitude of rotational speed fluctuation. A large rotational speed fluctuation may indicate a fault in the main shaft drive system.

[0094] S240. Fuse the multi-dimensional feature data through the principal component analysis method.

[0095] In this embodiment, the process of fusing the multi-dimensional feature data is as follows:

[0096] S241. Perform standardization processing on the multi-dimensional feature data, including:

[0097] Calculate the mean of the multi-dimensional feature data :

[0098] ;

[0099] where N is the number of multi-dimensional feature data;

[0100] Calculate the standard deviation of the multi-dimensional feature data :

[0101] ;

[0102] Standardize the multi-dimensional feature data through the mean and the standard deviation:

[0103] ;

[0104] where is the i th multi-dimensional feature data in the dataset, is the mean of the multi-dimensional feature data, is the standard deviation of the multi-dimensional feature data, is the i th data point obtained by normalizing the multi-dimensional feature data in the dataset;

[0105] The data points obtained by the normalization process

[0106] S242. Construct a covariance matrix, solve for the eigenvalues and eigenvectors, including:

[0107] Construct the covariance matrix:

[0108] ;

[0109] where, Z is the multi-dimensional feature matrix composed of the data points , n is the number of samples, and each sample is Z 's transpose matrix, is the covariance matrix, representing the covariance between each feature;

[0110] Construct the characteristic polynomial according to the covariance matrix:

[0111] ;

[0112] where I is the identity matrix, is the characteristic polynomial;

[0113] By solving the characteristic polynomial , all the eigenvalues of the covariance matrix are obtained, m is the total number of eigenvalues;

[0114] For each eigenvalue , by solving the system of equations , the corresponding eigenvector is obtained, k = 1, 2, 3,..., m ;

[0115] S243. Sort the eigenvalues from largest to smallest and select a preset number of principal components in the front;

[0116] S244. Project the multi-dimensional feature data onto the selected several principal components to obtain the multi-dimensional feature data after dimensionality reduction.

[0117] In the method for analyzing the state of the spindle of a five-axis machine tool based on multi-dimensional data of the present application, in S300, based on the multi-dimensional feature data obtained above, the health state of the spindle is determined through a machine learning model. The specific process includes:

[0118] S310: Divide the multi-dimensional feature data into a training set, a validation set, and a test set;

[0119] S320: Build a fully-connected neural network model using the deep learning framework Keras;

[0120] S330: Train the fully-connected neural network model with the training set, select the model and tune the hyperparameters with the validation set, and evaluate the fully-connected neural network model with the test set;

[0121] S340: Determine the health state of the spindle through the fully-connected neural network model. The types of the health state of the spindle include healthy, slightly abnormal, and severely abnormal.

[0122] In the preferred embodiment of the present application, the multi-dimensional feature data is divided into a training set, a validation set, and a test set; among them, the training set is used to train the model, accounting for 70% of the total data, the validation set is used for model selection and hyperparameter tuning, accounting for 15% of the total data, and the test set is used to evaluate the performance of the final model, accounting for 15% of the total data. The fully-connected neural network model consists of multiple fully-connected layers, each layer having a certain number of neurons, and performing non-linear transformation through an activation function; among them, the size of the input layer depends on the number of features; according to the complexity of the problem, multiple hidden layers are used, and the hidden layers usually have 128 or 256 neurons, or larger; the number of neurons in the output layer corresponds to the number of classifications of the health state of the spindle. In this embodiment, there are 3 classifications: healthy, slightly abnormal, and severely abnormal. Train the model with the training set, monitor the performance during the training process, avoid overfitting, deploy the trained model to the actual system, and update the model regularly to maintain high efficiency.

[0123] In the method for analyzing the state of the spindle of a five-axis machine tool based on multi-dimensional data of the present application, in S400, based on the current health state of the spindle, combined with historical experience data and a regression model, predict the impact data on workpiece forming caused by the current health state of the spindle, determine whether it will have an impact on subsequent workpiece forming, and analyze the possible types of defects that may occur during the workpiece forming process. In the preferred embodiment of the present application, the specific process is as follows:

[0124] S410: Predict the impact data on workpiece forming caused by the health state of the spindle according to the experience data. The impact data includes the health state score and various processing parameters;

[0125] S420. Input the health status score and various processing parameters into a regression model to calculate the workpiece quality:

[0126] ;

[0127] where Y is the workpiece quality, is the intercept, is the corresponding regression coefficient, is the health status score or a processing parameter;

[0128] Intercept represents the expected value of the workpiece quality Y when the health status score and all processing parameters are 0; the regression coefficient represents the degree of influence of the health status score or each processing parameter on the workpiece quality Y; the health status score or a processing parameter represents a factor that may affect the workpiece quality.

[0129] S430. Determine the type of defect that occurred during the workpiece forming process based on the workpiece quality.

[0130] Among them, the processing parameters include cutting speed, feed rate, cutting depth, tool wear amount, coolant state, etc. Substitute the health status score and processing parameters related to the possible forming defects of the workpiece into the regression model corresponding to the workpiece quality, and calculate the workpiece quality through the regression model. Finally, analyze the possible types of defects that may occur during the workpiece forming process based on the workpiece quality, such as unqualified surface roughness, excessive dimensional deviation, increased tool wear, etc.

[0131] For the five-axis machine tool spindle state analysis method based on multi-dimensional data in this application, if the prediction analysis shows that the type of defect determined by the spindle health status may have a negative impact on the subsequent workpiece forming, it is necessary to propose processing adjustment suggestions. Specifically, in S500, the adjustment strategies generated according to the type of defect include:

[0132] Adjust the spindle speed or feed rate;

[0133] Adjust the spindle load;

[0134] Optimize the cooling system or replace the lubricating oil;

[0135] Replace or repair faulty components.

[0136] Among them, by adjusting the spindle load, overloading operations can be avoided, and by optimizing the cooling system or replacing the lubricating oil, the spindle temperature can be reduced.

[0137] In the five-axis machine tool spindle status analysis method based on multi-dimensional data of the present application, in S600, adjustment instructions are generated according to the adjustment strategy, and the adjustment instructions are output to the operator or the machine tool control system in real time to ensure that the spindle always operates in the best working state. Through continuous monitoring and data feedback, the adjustment strategy is continuously optimized to improve the service life of the spindle and the workpiece processing quality.

[0138] The five-axis machine tool spindle status analysis method based on multi-dimensional data of the present application can comprehensively evaluate the spindle health status and detect potential problems in advance by real-time monitoring and analyzing multi-dimensional feature data through a machine learning model; predict the influence data of the spindle health status on the subsequent workpiece forming according to empirical data, and identify the defect types in combination with a regression model to avoid forming defects caused by spindle failures; avoid machining errors caused by spindle abnormalities through early adjustment, improve the workpiece processing quality, reduce downtime and production losses; real-time monitoring and intelligent adjustment help to reduce the working load of the spindle, reduce spindle failures, and extend the service life of the spindle.

[0139] Based on the above five-axis machine tool spindle status analysis method based on multi-dimensional data, the second aspect of the present application provides a five-axis machine tool spindle status analysis system based on multi-dimensional data, including:

[0140] A data acquisition module for real-time collecting the status data of the spindle through a variety of sensors;

[0141] A data processing module for extracting features from the preprocessed status data to obtain multi-dimensional feature data;

[0142] A status analysis module for determining the spindle health status based on the multi-dimensional feature data through a machine learning model;

[0143] A defect prediction module for predicting the influence data of the spindle health status on the workpiece forming according to empirical data, and determining the defect types occurring in the workpiece forming process based on the influence data through a regression model;

[0144] An adjustment module for generating an adjustment strategy according to the defect type;

[0145] An output module for generating adjustment instructions according to the adjustment strategy and performing workpiece processing according to the adjustment instructions.

[0146] The five-axis machine tool spindle status analysis system based on multi-dimensional data of the present application ensures that sensors such as vibration, temperature, torque, and rotational speed installed on the spindle have been correctly calibrated and can stably transmit data, and correctly connects each sensor to the data acquisition module to ensure the accuracy and real-time nature of data transmission; the data acquisition module collects key status parameters such as vibration, temperature, torque, and rotational speed of the spindle in real time through the sensors installed on the spindle, and transmits the collected status data to the data processing module; the data processing module processes the collected status data, including synchronizing the collected various status data in time, and using multi-dimensional data fusion technology to extract multi-dimensional feature data related to the spindle health status; the status analysis module based on the extracted multi-dimensional feature data uses a machine learning model to perform real-time analysis on the spindle health status to obtain the spindle health status classification and health status score; the defect prediction module based on the current spindle health status, combined with historical experience data and a regression model, predicts whether the current spindle health status will have an impact on the subsequent workpiece forming, and analyzes the possible types of defects that may occur during the workpiece forming process; if the prediction analysis shows that the defect type determined according to the spindle health status may have a negative impact on the subsequent workpiece forming, the adjustment module will propose an adjustment strategy according to the predicted defect type; the output module generates adjustment instructions according to the adjustment strategy, and the adjustment instructions can be output in the form of visual charts, reports, etc. to provide intuitive and clear reference information for machine tool operators and maintenance personnel, and the adjustment instructions can also be directly output to the machine tool control system in the form of electronic instructions to adjust the machining parameters, paths or logics to achieve the control of workpiece machining, and the spindle health status can also be viewed in real time through the monitoring interface of the output module.

[0147] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed in the present application should be covered by the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A five-axis machine tool spindle state analysis method based on multi-dimensional data, characterized in that Including: S100. Real - time collect the state data of the spindle through multiple sensors; S200. Extract features from the pre - processed state data to obtain multi - dimensional feature data; S300. Based on the multi - dimensional feature data, determine the health state of the spindle through a machine learning model; S400. Predict the impact data on workpiece forming caused by the health state of the spindle according to empirical data, and based on the impact data, determine the defect types occurring during the workpiece forming process through a regression model; S500. Generate an adjustment strategy according to the defect type; S600. Generate an adjustment instruction according to the adjustment strategy, and perform workpiece processing according to the adjustment instruction; The state data includes vibration data, temperature data, torque data, and rotational speed data; In S200, extracting features from the pre - processed state data to obtain multi - dimensional feature data includes: S210. Sort the state data according to the time stamp and synchronize the data through interpolation; S220. Pre - process the state data, and the pre - processing methods at least include cleaning, denoising, and filtering; S230. Extract multi - dimensional feature data related to the health state of the spindle from the state data; S240. Fuse the multi - dimensional feature data through the principal component analysis method; In S300, based on the multi - dimensional feature data, determining the health state of the spindle through a machine learning model includes: S310. Divide the multi - dimensional feature data into a training set, a validation set, and a test set; S320. Build a fully - connected neural network model using the deep - learning framework Keras; S330. Train the fully - connected neural network model through the training set, select the model and tune the hyperparameters through the validation set, and evaluate the fully - connected neural network model through the test set; S340. Determine the health state of the spindle through the fully - connected neural network model, and the types of the spindle health state include healthy, slightly abnormal, and severely abnormal; In S400, predicting the impact data on workpiece forming caused by the health state of the spindle according to empirical data, and based on the impact data, determining the defect types occurring during the workpiece forming process through a regression model includes: S410. Predict the impact data on workpiece forming caused by the health state of the spindle according to empirical data, and the impact data includes a health state score and multiple processing parameters; S420. Input the health state score and multiple processing parameters into the regression model to calculate the workpiece quality; where Y is the workpiece quality, is the intercept, is the corresponding regression coefficient, is the health status score or the processing parameter; S430. Determine the defect types occurring during the workpiece forming process according to the workpiece quality.

2. The five-axis machine tool spindle state analysis method based on multi-dimensional data according to claim 1, wherein In S230, extracting multi - dimensional feature data related to the health state of the spindle from the state data includes: Extract vibration feature data related to the health state of the spindle from the vibration data, including vibration amplitude and vibration frequency; Extract temperature feature data related to the health state of the spindle from the temperature data, including temperature change rate and temperature change trend; Extract torque feature data related to the health state of the main shaft from the torque data, including the amplitude of load fluctuation and the frequency of load fluctuation; Extract rotational speed feature data related to the health state of the main shaft from the rotational speed data, including the amplitude of rotational speed fluctuation.

3. The five-axis machine tool spindle state analysis method based on multi-dimensional data according to claim 2, characterized in that In S240, fuse the multi-dimensional feature data by the principal component analysis method, including: S241. Perform normalization processing on the multi-dimensional feature data, including: Calculate the mean of multi-dimensional feature data ; Calculate the standard deviation of multi-dimensional feature data ; Normalize the multi-dimensional feature data by the mean value and the standard deviation: Among them, is the i th multi-dimensional feature data in the dataset, is the mean of the multi-dimensional feature data, is the standard deviation of the multi-dimensional feature data, is the data point obtained by standardizing the i th multi-dimensional feature data in the dataset; S242. Construct a covariance matrix, solve the eigenvalues and eigenvectors, including: Construct a covariance matrix: Among them, Z is a multi-dimensional feature matrix composed of data points, n is the number of samples, is Z the transposed matrix of is the covariance matrix; Construct a characteristic polynomial according to the covariance matrix: where I is the identity matrix, is the characteristic polynomial; By solving the characteristic polynomial , all the eigenvalues of the covariance matrix are obtained , m is the total number of eigenvalues; For each eigenvalue , by solving the system of equations , the corresponding eigenvector is obtained k = 1, 2, 3, …, m ; S243. Sort the eigenvalues from largest to smallest, and select a preset number of principal components in the front; S244. Project the multi-dimensional feature data onto the selected several principal components to obtain the multi-dimensional feature data after dimensionality reduction.

4. The five-axis machine tool spindle state analysis method based on multi-dimensional data according to claim 3, characterized in that The machining parameters at least include cutting speed, feed rate, cutting depth, tool wear amount, and coolant state.

5. The method for analyzing the spindle state of a five-axis machine tool based on multi-dimensional data according to claim 4, wherein In S500, the adjustment strategies generated according to the defect type include: Adjust the rotational speed or feed speed of the main shaft; Adjust the load of the main shaft; Optimize the cooling system or replace the lubricating oil; Replace or repair the faulty components.

6. A five-axis machine tool spindle status analysis system based on multi-dimensional data, based on the five-axis machine tool spindle status analysis method based on multi-dimensional data according to any one of claims 1 to 5, characterized in that, Include: A data acquisition module for real-time collecting the state data of the main shaft through a variety of sensors; A data processing module for extracting features from the preprocessed state data to obtain multi-dimensional feature data; A state analysis module for determining the health state of the main shaft based on the multi-dimensional feature data through a machine learning model; A defect prediction module for predicting the influence data on the workpiece forming caused by the health state of the main shaft according to the empirical data, and determining the defect type occurring in the workpiece forming process based on the influence data through a regression model; An adjustment module for generating adjustment strategies according to the defect type; An output module for generating adjustment instructions according to the adjustment strategies and performing workpiece machining according to the adjustment instructions.

Images