Finished product error analysis method and system for numerical control machine tool

Abstract

The invention relates to the technical field of numerical control machine tool machining quality inspection work, in particular to a numerical control machine tool finished product error analysis method and system. Comprising workpiece sizes, machine tool states, machining parameters, environmental conditions, operator input, historical machining data, sensor data, quality control data and maintenance and calibration records; through comprehensive application of data preprocessing, statistical analysis, a machine learning algorithm and a model evaluation and optimization technology, accurate prediction and effective control of errors in the machining process of the numerical control machine tool are realized; in addition, data accuracy and analysis reliability are enhanced through weighted data standardization processing, and data features which are crucial to error analysis are reserved while dimensionality reduction is achieved through application of weighted principal component analysis. Through the combination of the technologies, the error prediction precision is improved, and the production efficiency and the product quality are remarkably improved through real-time monitoring and dynamic adjustment of machining parameters.

Description

Technical Field

[0001] The present invention relates to the technical field of quality inspection work in numerical control machine tool processing, and particularly to a method and system for error analysis of finished products of numerical control machine tools. Background Art

[0002] In the field of machining, the processing quality of raw materials is directly related to the performance and reliability of the final product. Traditional processing procedures often focus on the control of the processing process, while relatively less attention is paid to the inspection of raw materials and the quality monitoring of finished products. This may lead to defective products during the processing, increasing production costs and affecting the overall quality of the products.

[0003] To improve the processing quality, various devices and technological processes have been adopted in the prior art. For example, CN117047424A proposes a processing inspection process method for bevel worm gear pair reducers. This method effectively avoids the generation of defective products during the processing by strictly inspecting the raw materials before processing. In addition, after processing, the finished products will go through a series of precise inspection processes to ensure that the quality of the parts meets the standards. However, the prior art has limitations in real-time data processing and multi-dimensional error analysis, multi-source data processing, comprehensive evaluation of machine tool status, and the accuracy of error prediction. Summary of the Invention

[0004] In view of the problem in the above or the prior art that the prior art has limitations in real-time data processing and multi-dimensional error analysis, multi-source data processing, comprehensive evaluation of machine tool status, and the accuracy of error prediction, the present invention is proposed.

[0005] Therefore, the object of the present invention is to provide a method for error analysis of finished products of numerical control machine tools.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: including collecting the original data during the numerical control machine tool processing, including workpiece dimensions, machine tool status, processing parameters, environmental conditions, operator input, historical processing data, sensor data, quality control data, maintenance and calibration records, and production batch information; preprocessing the original data, including data cleaning, data standardization, and dimensionality reduction processing; statistical analysis, using statistical analysis methods to perform preliminary error analysis on the preprocessed data; adopting machine learning algorithms to deeply analyze the error data and establish an error prediction model; predicting and evaluating the finished product error according to the prediction model; and proposing an error correction scheme according to the evaluation result.

[0007] As a preferred embodiment of the method for analyzing the finished product errors of the numerically controlled machine tools of the present invention, further comprising: performing fusion processing on the collected multi-source data to ensure the consistency and integrity of the data; applying wavelet transform to denoise the signals to improve the accuracy of subsequent analysis; using principal component analysis to reduce the dimension of high-dimensional data and reduce the complexity of the data.

[0008] As a preferred embodiment of the method for analyzing the finished product errors of the numerically controlled machine tools of the present invention, wherein: the statistical analysis step further comprises performing variance analysis to determine which factors have a significant impact on the errors; performing correlation analysis to identify the relationships between different error sources; using covariance analysis to evaluate the combined impact of multiple independent variables on the errors.

[0009] As a preferred embodiment of the method for analyzing the finished product errors of the numerically controlled machine tools of the present invention, wherein: the machine learning algorithms include using the random forest algorithm to classify the data set to identify the patterns of error generation; using the gradient boosting decision tree to perform regression analysis on the errors to predict the magnitude of the errors; using the recurrent neural network in deep learning to model the time series data to analyze the changing trend of the errors over time.

[0010] As a preferred embodiment of the method for analyzing the finished product errors of the numerically controlled machine tools of the present invention, wherein: the establishment of the error prediction model includes using the cross-validation method to evaluate the generalization ability of the model; applying the Akaike information criterion or the Bayesian information criterion to select the optimal model; using the model fusion technology to improve the accuracy of the prediction.

[0011] As a preferred embodiment of the method for analyzing the finished product errors of the numerically controlled machine tools of the present invention, wherein: the error evaluation includes calculating the mean, variance, and standard deviation of the errors to evaluate the central tendency and dispersion degree of the errors; using control charts to monitor the changes in the errors over time to ensure the stability of the machining process; applying the six sigma methodology to evaluate and improve the quality of the machining process.

[0012] As a preferred embodiment of the method for analyzing the finished product errors of the numerically controlled machine tools of the present invention, wherein: the proposal of the error correction scheme includes automatically adjusting the control parameters of the numerically controlled machine tool, such as tool compensation, feed rate, etc. based on the error evaluation results; using the simulated annealing algorithm or the genetic algorithm to optimize the machining path to reduce the machining errors; dynamically adjusting the machining strategy according to the real-time monitoring data to adapt to the changes in the machining conditions.

[0013] Beneficial effects of the finished product error analysis method for numerically controlled machine tools of the present invention: First, by accurately collecting data and using an advanced error prediction model, the accuracy of the processed finished product is significantly improved, effectively reducing the output of defective products. Second, by reducing material waste and rework requirements, the production cost is significantly reduced, and at the same time, the material utilization rate is increased. In addition, the robust design of this method enables it to adapt to changing processing environments and maintain the stability of processing quality. The real-time monitoring and dynamic parameter adjustment capabilities optimize the processing flow, shorten the response time, and reduce the possible downtime. The data-driven decision support provides a scientific basis for production management, and the scalability of the method enables it to easily adapt to the needs of different machine tools and processes. Finally, the method of the present invention also reflects the characteristics of environmental friendliness, helping enterprises achieve green production by reducing waste and energy consumption.

[0014] To solve the above technical problems, the present invention also provides the following technical solution: A finished product error analysis system for numerically controlled machine tools, including a data collection module for collecting raw data during the processing of numerically controlled machine tools; a data preprocessing module for cleaning, standardizing, and dimension reduction processing of the collected raw data; a statistical analysis module for preliminary error analysis of the preprocessed data; a machine learning module for in-depth analysis of error data using machine learning algorithms and establishing an error prediction model; an error prediction and evaluation module for predicting and evaluating the finished product error according to the prediction model; and an error correction scheme generation module for proposing an error correction scheme according to the evaluation result.

[0015] As a preferred solution of the finished product error analysis system for numerically controlled machine tools of the present invention, wherein: The data preprocessing module further includes

[0016] A weighted data standardization unit for dynamically adjusting the weight coefficient w according to data characteristics and applying the following formula where x i is the i-th data point, μ is the average value of the data, σ is the standard deviation of the data, and w i is the weight coefficient of the i-th data point.

[0017] As a preferred solution of the finished product error analysis system for numerically controlled machine tools of the present invention, wherein: A weighted principal component analysis unit for assigning different weights according to data characteristics and applying the following weighted formula

[0018] PC w =cov w (X) 1 / 2 X

[0019] where X is the original data matrix, cov w (X) is the covariance matrix with different weights assigned according to data characteristics, and PC wIt is the weighted principal component score matrix.

[0020] Beneficial effects of the finished product error analysis system for numerical control machine tools of the present invention: By comprehensively applying data preprocessing, statistical analysis, machine learning algorithms, and model evaluation and optimization techniques, accurate prediction and effective control of errors during the machining process of numerical control machine tools are achieved; in addition, the processing of weighted data standardization enhances the accuracy of data and the reliability of analysis, and the application of weighted principal component analysis retains the data features crucial for error analysis while reducing the dimensionality; the combination of these technologies not only improves the accuracy of error prediction, but also significantly improves production efficiency and product quality through real-time monitoring and dynamic adjustment of machining parameters; in addition, the present invention also effectively reduces production costs by reducing waste products and rework, and at the same time optimizes the maintenance plan and reduces downtime. Brief Description of the Drawings

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0022] Figure 1 It is a schematic diagram of the overall process of the method for analyzing the errors of finished products of numerical control machine tools. Detailed Embodiments

[0023] To make the above objects, features, and advantages of the present invention more obvious and understandable, the detailed embodiments of the present invention will be described below in conjunction with the drawings in the specification.

[0024] Many specific details are set forth in the following description in order to provide a thorough understanding of the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the spirit of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0025] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that can be included in at least one implementation of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or alternative embodiments that exclude each other with other embodiments.

[0026] Example 1, refer to Figure 1, which is the first embodiment of the present invention. This embodiment provides a method for analyzing the errors of finished products of numerically controlled machine tools, including collecting the original data during the machining process of numerically controlled machine tools, including workpiece dimensions, machine tool status, machining parameters, environmental conditions, operator inputs, historical machining data, sensor data, quality control data, maintenance and calibration records, and production batch information; preprocessing the original data, including data cleaning, data standardization, and dimensionality reduction; statistical analysis, using statistical analysis methods to conduct preliminary error analysis on the preprocessed data; adopting machine learning algorithms to conduct in-depth analysis on the error data and establish an error prediction model; predicting and evaluating the finished product errors according to the prediction model; and proposing an error correction plan according to the evaluation results. Among them, the method for analyzing the errors of finished products of numerically controlled machine tools of the present invention provides a solid data foundation for error analysis by collecting comprehensive original data; the collected data types are extensive, covering multiple aspects from workpiece dimensions to production batch information, ensuring the comprehensiveness and accuracy of the analysis results. Selecting to collect these data is to ensure that the analysis method can capture the factors that may cause errors from different angles, thereby providing more accurate error prediction and correction plans.

[0027] Specifically, it further includes fusing the collected multi-source data to ensure the consistency and integrity of the data; applying wavelet transform to denoise the signals to improve the accuracy of subsequent analysis; adopting principal component analysis to conduct dimensionality reduction on high-dimensional data and reduce the complexity of the data. The present invention fuses multi-source data and conducts denoising and dimensionality reduction on the data through wavelet transform and principal component analysis to improve the accuracy and efficiency of the analysis; fusing multi-source data is to obtain a more comprehensive data perspective, while denoising and dimensionality reduction are to remove unnecessary data redundancy and retain the most critical information for error analysis, which helps to improve the performance of the model and the speed of analysis.

[0028] Furthermore, the statistical analysis step further includes conducting variance analysis to determine which factors have a significant impact on the errors; implementing correlation analysis to identify the relationships between different error sources; using covariance analysis to evaluate the combined impact of multiple independent variables on the errors. In the statistical analysis stage, the present invention adopts variance analysis, correlation analysis, and covariance analysis to determine the influencing factors of the errors and their relationships. The selection of these statistical methods is to quantify the error sources from different angles and provide data support for subsequent in-depth analysis.

[0029] Among them, the machine learning algorithms include using the random forest algorithm to classify the data set to identify the patterns of error generation; using the gradient boosting decision tree to perform regression analysis on the errors to predict the magnitude of the errors; using the recurrent neural network in deep learning to model the time series data to analyze the changing trend of the errors over time. The selection of machine learning algorithms, including random forest, gradient boosting decision tree, and recurrent neural network, aims to improve the accuracy of error prediction through the advantages of different algorithms. These algorithms each have their unique advantages in classification, regression, and time series analysis, and their comprehensive application can capture the characteristics of error data more comprehensively.

[0030] Preferably, the establishment of the error prediction model includes using the cross-validation method to evaluate the generalization ability of the model; applying the Akaike information criterion or the Bayesian information criterion to select the optimal model; using the model fusion technology to improve the prediction accuracy. When establishing the error prediction model, the cross-validation, Akaike information criterion or Bayesian information criterion, and model fusion technology are adopted. The selection of these technologies is to ensure that the model not only performs well on the training data, but also has good generalization ability and prediction accuracy.

[0031] It should be noted that the error evaluation includes calculating the mean, variance, and standard deviation of the errors to evaluate the central tendency and dispersion degree of the errors; using control charts to monitor the changes of the errors over time to ensure the stability of the processing process; applying the six sigma methodology to evaluate and improve the quality of the processing process. In the error evaluation stage, the stability and quality of the processing process are comprehensively evaluated by calculating the statistical quantities of the errors, using control charts, and applying the six sigma methodology. The selection of these evaluation means helps to understand the characteristics of the errors from different dimensions and provides a basis for formulating effective error correction schemes.

[0032] When in use, the proposal of the error correction scheme includes automatically adjusting the control parameters of the numerical control machine tool based on the error evaluation results, such as tool compensation, feed rate, etc.; using the simulated annealing algorithm or genetic algorithm to optimize the machining path to reduce the machining errors; dynamically adjusting the machining strategy according to the real-time monitoring data to adapt to the changes in the machining conditions. In the stage of proposing the error correction scheme, the present invention can automatically adjust the control parameters, optimize the machining path, and dynamically adjust the machining strategy according to the real-time monitoring data. The selection of these measures is to achieve the real-time optimization of the machining process, reduce the errors caused by human factors, and improve the production efficiency and product quality.

[0033] Example 2, refer to Figure 1, which is the second embodiment of the present invention. Different from the previous embodiment, this embodiment provides a finished product error analysis system for numerical control machine tools, which includes a data collection module for collecting the original data during the machining process of the numerical control machine tool; a data preprocessing module for cleaning, standardizing, and dimension reduction processing of the collected original data; a statistical analysis module for preliminary error analysis of the preprocessed data; a machine learning module for in-depth analysis of the error data using machine learning algorithms and establishing an error prediction model; an error prediction and evaluation module for predicting and evaluating the finished product error according to the prediction model; an error correction scheme generation module for proposing an error correction scheme based on the evaluation results. The design of the finished product error analysis system for numerical control machine tools integrates multiple modules such as data collection, preprocessing, statistical analysis, machine learning, error prediction and evaluation, and error correction scheme generation. The design of the system aims to provide an integrated solution and improve the flexibility and scalability of the system through modular design.

[0034] Specifically, the data preprocessing module further includes a weighted data standardization unit for dynamically adjusting the weight coefficient w according to the data characteristics and applying the following formula,

[0035]

[0036] where x i is the i-th data point, μ is the average value of the data, σ is the standard deviation of the data, and w i is the weight coefficient of the i-th data point. In the data preprocessing module, a weighted data standardization unit is introduced to dynamically adjust the weight coefficient. Weighted data standardization is to consider the importance of different data points and improve the accuracy of subsequent analysis.

[0037] Furthermore, a weighted principal component analysis unit for assigning different weights according to the data characteristics and applying the following weighted formula,

[0038] PC w =cov w (X) 1 / 2 X

[0039] where X is the original data matrix, cov w (X) is the covariance matrix with different weights assigned according to the data characteristics, and PC w is the weighted principal component score matrix. The system also includes a weighted principal component analysis unit for assigning different weights according to the data characteristics. Weighted principal component analysis is to retain the most important data characteristics for error analysis while reducing the dimension, thereby improving the performance of the model.

[0040] The remaining structures are the same as those in Embodiment 1.

[0041] In summary, the error analysis process for CNC machining finished products is as follows:

[0042] S1: Data collection, collect the data generated during the CNC machining process, including but not limited to workpiece dimensions, machine tool status, machining parameters, environmental conditions, etc.

[0043] S2: Data preprocessing, clean the collected data to remove invalid or incorrect data points; perform data standardization to ensure that the data is on the same dimension and scale; use methods such as principal component analysis (PCA) for dimensionality reduction to reduce the complexity of the data.

[0044] S3: Error source identification, analyze the data to identify possible error sources, such as tool wear, machine tool vibration, material variation, etc.

[0045] S4: Deployment of real-time monitoring system, install necessary sensors on the machine tool to monitor the key parameters of the machining process in real time, and transmit the real-time data to the data processing center for further analysis.

[0046] S5: Error model establishment, use historical data and machine learning algorithms to establish an error prediction model; select appropriate algorithms, such as support vector machine (SVM), random forest, etc.

[0047] S6: Model training and verification, use the training dataset to train the error model.

[0048] Evaluate the prediction accuracy and generalization ability of the model through the validation dataset.

[0049] S7: Error prediction, input the data obtained from real-time monitoring into the error model to predict potential errors during the machining process.

[0050] S8: Error correction measures, according to the error prediction results, automatically or manually adjust the machining parameters to implement error correction measures.

[0051] S9: Finished product inspection, after machining, perform precise dimensional and geometric shape inspections on the finished product; use high-precision inspection equipment such as coordinate measuring machine (CMM), laser scanner, etc.

[0052] S10: Quality assessment and feedback, perform quality assessment on the finished product inspection data to determine whether the product meets the quality standards; feedback the quality assessment results to the error model for continuous optimization and improvement of the model.

[0053] Importantly, it should be noted that the construction and arrangement of the present application shown in multiple different exemplary embodiments are merely illustrative. Although only a few embodiments are described in detail in this disclosure, those who refer to this disclosure should easily understand that many modifications are possible without materially departing from the novel teachings and advantages of the subject matter described in this application (e.g., changes in the dimensions, scales, structures, shapes and proportions of various elements, as well as parameter values (such as temperature, pressure, etc.), installation arrangements, use of materials, colors, orientations, etc.). For example, an element shown as integrally formed may be composed of multiple parts or elements, the position of the element may be inverted or otherwise changed, and the nature, number or position of discrete elements may be altered or changed. Accordingly, all such modifications are intended to be included within the scope of the present invention. The order or sequence of any process or method steps may be changed or reordered according to alternative embodiments. In the claims, any "means-plus-function" clause is intended to cover the structures that perform the recited function described herein, and not only structural equivalents but also equivalent structures. Other substitutions, modifications, changes and omissions may be made in the design, operating conditions and arrangement of the exemplary embodiments without departing from the scope of the present invention. Accordingly, the present invention is not limited to the specific embodiments, but extends to various modifications that still fall within the scope of the appended claims.

[0054] In addition, in order to provide a concise description of the exemplary embodiments, not all features of the actual embodiments may be described (i.e., those features that are not relevant to the currently contemplated best mode of carrying out the present invention or those features that are not relevant to the implementation of the present invention).

[0055] It should be understood that in the development of any actual implementation, as in any engineering or design project, numerous specific implementation decisions may be made. Such development efforts may be complex and time-consuming, but for those of ordinary skill in the art who benefit from this disclosure, without undue experimentation, such development efforts will be a routine task of design, manufacturing and production.

[0056] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A method for analyzing the error of a finished product of a CNC machine tool, characterized in that: include, Collect raw data from CNC machine tool processing, including workpiece dimensions, machine tool status, processing parameters, environmental conditions, operator input, historical processing data, sensor data, quality control data, maintenance and calibration records, and production batch information; Preprocess the raw data, including data cleaning, data standardization and dimensionality reduction; Statistical analysis: preliminary error analysis of the preprocessed data is performed using statistical analysis methods; Use machine learning algorithms to conduct in-depth analysis of error data and establish an error prediction model; Predict and evaluate finished product errors based on the prediction model; Based on the evaluation results, an error correction plan is proposed.

2. The method for analyzing the error of a finished product of a CNC machine tool according to claim 1, characterized in that: Also includes, Fuse the collected multi-source data to ensure the consistency and integrity of the data; Apply wavelet transform to denoise the signal to improve the accuracy of subsequent analysis; Principal component analysis is used to reduce the dimensionality of high-dimensional data and reduce the complexity of the data.

3. The method for analyzing the error of a finished product of a CNC machine tool according to claim 1, characterized in that: The statistical analysis step further comprises, Conduct ANOVA to determine which factors have a significant effect on the error; Conduct correlation analysis to identify the interrelationships between different error sources; Covariance analysis was used to assess the combined effects of multiple independent variables on the error.

4. The method for analyzing the error of a finished product of a CNC machine tool according to claim 3, characterized in that: The machine learning algorithm includes, Use the random forest algorithm to classify the data set to identify patterns in error generation; The error is regressed using a gradient boosting decision tree to predict the size of the error; Time series data is modeled through recurrent neural networks in deep learning to analyze the trend of errors over time.

5. The method for analyzing the error of a finished product of a CNC machine tool according to claim 4, characterized in that: The establishment of the error prediction model includes: Use cross-validation methods to evaluate the generalization ability of the model; Apply Akaike Information Criterion or Bayesian Information Criterion to select the optimal model; The accuracy of prediction can be improved through model fusion technology.

6. The method for analyzing the error of a finished product of a CNC machine tool according to claim 4 or 5, characterized in that: The error assessment includes, Calculate the mean, variance, and standard deviation of the errors to assess the central tendency and dispersion of the errors; Use control charts to monitor the change of errors over time to ensure the stability of the machining process; Apply Six Sigma methodology to evaluate and improve the quality of machining processes.

7. The method for analyzing the error of a finished product of a CNC machine tool according to claim 6, characterized in that: The error correction scheme includes: Automatically adjust the control parameters of CNC machine tools, such as tool compensation and feed rate, based on the error evaluation results; Use simulated annealing algorithm or genetic algorithm to optimize the processing path to reduce processing errors; Based on real-time monitoring data, the processing strategy is dynamically adjusted to adapt to changes in processing conditions.

8. A finished product error analysis system for a CNC machine tool, characterized in that: include, Data collection module, used to collect raw data during the CNC machine tool processing; Data preprocessing module, used to clean, standardize and reduce the dimension of the collected raw data; Statistical analysis module, used to perform preliminary error analysis on preprocessed data; The machine learning module is used to conduct in-depth analysis of error data using machine learning algorithms and establish an error prediction model; Error prediction and evaluation module, used to predict and evaluate the finished product error based on the prediction model; The error correction scheme generation module is used to propose an error correction scheme based on the evaluation results.

9. The finished product error analysis system of a numerically controlled machine tool according to claim 8, characterized in that: The data preprocessing module further includes: A weighted data normalization unit is used to dynamically adjust the weight coefficient w according to the data characteristics and apply the following formula, Among them, x i is the i-th data point, μ is the mean of the data, σ is the standard deviation of the data, and w i is the weight coefficient of the ith data point.

10. The finished product error analysis system of a numerically controlled machine tool according to claim 8, characterized in that: A weighted principal component analysis unit is used to assign different weights according to data characteristics and apply the following weighting formula, PC w =cov w (X) 1 / 2 X Where X is the original data matrix, cov w (X) is the covariance matrix with different weights assigned according to data characteristics, PC w is the weighted principal component score matrix.

Images