Intelligent prediction and compensation method for wear of numerical control machining tool

Abstract

The invention provides an intelligent prediction and compensation method for wear of a numerical control machining tool, and belongs to the technical field of industrial control. The method comprises the steps that cutting process data of a numerical control machining tool are collected and preprocessed, and standardized data are obtained; performing symbol conversion on the standardized data to obtain symbol sequence data; obtaining a tool wear state prediction value based on the symbol sequence through a tool wear state prediction model; determining a wear compensation strategy according to the predicted value of the tool wear state; and the wear compensation strategy serves as a control instruction and is executed through a numerical control system. A symbol dynamic system and an adaptive symbol mapping mechanism are introduced, a tool wear process is regarded as a symbol dynamic system, and evolution of a wear state is represented by a symbol sequence; the problems that in the prior art, a numerical control machining tool abrasion prediction and compensation method is low in prediction precision, poor in model generalization ability, lack of flexibility of compensation strategies and the like are solved.

Description

Technical Field

[0001] The invention relates to the field of industrial control technology, and in particular to a method for intelligently predicting and compensating wear of numerical control machining tools. Background Art

[0002] In CNC machining, tool wear is a complex and dynamic process, which is affected by many factors, such as cutting force, temperature, vibration, etc. The interaction and changes between these factors are highly nonlinear and uncertain, making it difficult to accurately predict and timely compensate for tool wear.

[0003] Traditional methods often use a single signal feature to predict tool wear based on empirical models; however, wear prediction relies only on a single signal feature, such as judging the degree of wear only by the change in the cutting force signal. This method cannot fully capture the complex dynamic behavior of the wear process and easily ignores the influence of other important factors, such as changes in temperature and vibration signals, resulting in insufficient reliability of the prediction results. Moreover, the prediction effect of the empirical model under specific conditions is acceptable, but when the processing conditions change, such as replacing workpieces of different materials or adjusting cutting parameters, the generalization ability of the model is poor and cannot adapt to the new wear law. For example, the wear prediction model established when processing aluminum alloys may be completely invalid when processing stainless steel, and it needs to be remodeled, which increases the workload and time cost. Traditional compensation strategies are often fixed, such as replacing or grinding the tool only when the wear reaches a certain level. This strategy cannot be flexibly adjusted according to the real-time changes in the wear state, which easily leads to untimely compensation or over-compensation. For example, the cutting parameters are not adjusted in time to slow down the wear rate at the beginning of wear, and the tool is hastily replaced when the wear is serious, which affects the processing efficiency and product quality.

[0004] Therefore, an intelligent prediction and compensation method for CNC machining tool wear is needed. Summary of the invention

[0005] In view of this, the present invention provides an intelligent prediction and compensation method for CNC machining tool wear, introduces a symbolic dynamic system and an adaptive symbol mapping mechanism, regards the tool wear process as a symbolic dynamic system, and uses symbol sequences to represent the evolution of the wear state, so as to solve the problems of low prediction accuracy, poor model generalization ability, and lack of flexibility in compensation strategies in the prior art of CNC machining tool wear prediction and compensation methods.

[0006] To this end, the present invention provides the following technical solutions:

[0007] A method for intelligent prediction and compensation of tool wear in numerical control machining, comprising:

[0008] Collect the cutting process data of the CNC machining tool and perform preprocessing to obtain standardized data;

[0009] Perform symbol conversion on the standardized data to obtain symbol sequence data;

[0010] Through the tool wear state prediction model, obtain the tool wear state prediction value based on the symbol sequence data;

[0011] Determine the wear compensation strategy according to the tool wear state prediction value;

[0012] Take the wear compensation strategy as a control instruction and execute it through the CNC system.

[0013] Furthermore, it also includes:

[0014] Characterize the future wear trend of the tool through the symbol transition probability matrix;

[0015] Combine the tool wear state prediction value and the future wear trend to obtain the trend compensation strategy.

[0016] Furthermore, the cutting process data of the CNC machining tool includes:

[0017] Cutting force signal, temperature signal and vibration signal.

[0018] Furthermore, the performing symbol conversion on the standardized data to obtain symbol sequence data includes:

[0019] Calculate the mean and standard deviation of the standardized data as statistical features;

[0020] Divide the symbol interval based on the statistical features and assign symbols to the standardized data.

[0021] Furthermore, the determining the wear compensation strategy according to the tool wear state prediction value includes:

[0022] If the tool wear state prediction value is slight wear, adjust the cutting parameters;

[0023] If the tool wear state prediction value is moderate wear, perform local grinding on the tool;

[0024] If the tool wear state prediction value is severe wear, replace the tool;

[0025] If the tool wear state prediction value is abnormal state, stop the machining.

[0026] Furthermore, the combining the tool wear state prediction value and the future wear trend to obtain the trend compensation strategy includes:

[0027] If the predicted value of the tool wear state is mild wear and the future wear trend is moderate wear, increase the adjustment range of the cutting parameters;

[0028] If the predicted value of the tool wear state is moderate wear and the future wear trend is severe wear, replace the tool;

[0029] If the predicted value of the tool wear state is mild wear and the future wear trend is abnormal, repair the equipment;

[0030] If the predicted value of the tool wear state is abnormal and the future wear trend is abnormal, stop work for treatment.

[0031] Further, obtaining the predicted value of the tool wear state based on the symbol sequence data through the tool wear state prediction model includes:

[0032] Pre-train the tool wear state prediction model;

[0033] Use the pre-trained tool wear state prediction model to input the symbol sequence data and output the predicted value of the tool wear state.

[0034] Further, pre-training the tool wear state prediction model includes:

[0035] The tool wear state prediction model is an unsupervised learning model;

[0036] Perform clustering analysis on the symbol sequence; and label the clustering results, mapping each cluster to the corresponding wear mode;

[0037] Use cross-validation to optimize the model parameters;

[0038] Minimize the difference between the input symbol sequence and the reconstructed symbol sequence.

[0039] Further, dividing the symbol interval based on the statistical features and assigning symbols to the standardized data includes:

[0040]

[0041] Wherein, F represents the cutting force signal, T represents the temperature signal; V represents the vibration signal; are the mean values of the cutting force signal, temperature signal and vibration signal respectively, S F 、S T 、S V are the standard deviations of the cutting force signal, temperature signal and vibration signal respectively.

[0042] Advantages and positive effects of the present invention:

[0043] By modeling the tool wear process as a symbolic dynamic system and transforming the continuous wear state into a discrete symbol sequence, the present invention can accurately capture the non-linear wear dynamics. The adaptive symbol mapping mechanism dynamically optimizes the symbol segmentation rules by real-time sensing of data characteristics, enabling the symbolization process to adapt to different working conditions and individual differences, and significantly enhancing the robustness of state representation. Combining the spatio-temporal correlation analysis of the trend prediction database, multi-scale laws of wear evolution are mined from historical symbol sequences, which not only enhances the early perception ability of sudden wear but also reduces the prediction delay of progressive wear. This hybrid architecture integrating dynamic symbolization and adaptive learning enables the compensation strategy to quickly respond to sudden anomalies and perform prospective optimization based on long-term trends, ultimately achieving a coordinated improvement in three dimensions: prediction accuracy, real-time performance, and generalization ability. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0045] Figure 1 It is a flowchart of the intelligent prediction and compensation method for the wear of a numerical control machining tool in Embodiment 1 of the present invention;

[0046] Figure 2 It is a flowchart of the intelligent prediction and compensation method for the wear of a numerical control machining tool in Embodiment 2 combined with a symbol transition probability matrix;

[0047] Figure 3 It is a flowchart for formulating the compensation strategy in Embodiment 1 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0048] In order to enable those skilled in the art to better understand the solution of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some of the embodiments of the present invention, rather than all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0049] It should be noted that the terms "first", "second", etc. in the description, claims and above-mentioned drawings of the present invention are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present invention described here can be implemented in an order other than those illustrated or described here. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0050] The present invention provides an intelligent prediction and compensation method for tool wear in numerical control machining, which collects the cutting process data of the numerical control machining tool in real time, cleans and normalizes the collected cutting process data, and converts the processed standardized data into symbol sequence data according to the adaptive symbol mapping rule; trains a model using the symbol sequence data and its corresponding tool wear pattern definition rule to form a trained wear state prediction model; inputs the symbol sequence data generated in real time into the trained wear state prediction model, and combines the trend prediction database of the symbol sequence data updated in real time to predict the future wear trend and predict the current tool wear state; formulates corresponding compensation strategies according to the predicted tool wear state, in combination with the processing technology requirements and tool characteristics. Through the above method, the accuracy and response speed of wear state prediction are improved, and a more intelligent and flexible compensation strategy formulation is realized. Specifically, the present invention includes the following steps:

[0051] S1. Data acquisition;

[0052] Receive the cutting process data of the numerical control machining tool, including cutting force signal, temperature signal and vibration signal;

[0053] S2. Data preprocessing and symbol conversion;

[0054] 1) Preprocess the received cutting process data to obtain standardized data;

[0055] Specifically, the preprocessing includes: using a low-pass filter to filter out high-frequency noise; unifying data with different dimensions to the same numerical range.

[0056] 2) Convert the standardized data into symbol sequence data according to the adaptive symbol mapping rule. Specifically,

[0057] Calculate the mean and standard deviation of the standardized cutting force signal, temperature signal and vibration signal respectively as statistical features;

[0058] Dynamically divide the symbol intervals: Dynamically divide the symbol intervals of the cutting force signal, temperature signal, and vibration signal according to statistical characteristics. The cutting force signal is divided into three symbol intervals of "low", "medium", and "high", which are represented by the symbols "0", "1", and "2" respectively. The temperature signal is divided into three symbol intervals of "low temperature", "medium temperature", and "high temperature", which are represented by the symbols "3", "4", and "5" respectively. The vibration signal is divided into three symbol intervals of "low frequency", "medium frequency", and "high frequency", which are represented by the symbols "6", "7", and "8" respectively;

[0059] Generate symbol sequence data: Convert it into symbol sequence data according to the adaptive symbol mapping rule.

[0060] The adaptive symbol mapping rule automatically adjusts the division of symbol intervals according to the changes in real-time data.

[0061] S3. Obtain the predicted value of the tool wear state based on the symbol sequence data through the tool wear state prediction model;

[0062] 1) Use the symbol sequence data and its corresponding tool wear mode definition rules to train the model. Specifically,

[0063] Data preparation: Use unsupervised learning methods to perform clustering analysis on the symbol sequence, perform expert annotation on the clustering results, and map each cluster to a specific wear mode;

[0064] Model selection and training: Select a suitable unsupervised learning model, use cross-validation to optimize the model parameters, and ensure the generalization ability of the model;

[0065] Training process: Minimize the difference between the input symbol sequence x and the reconstructed symbol sequence therebetween;

[0066] Model evaluation: Use the reserved test set to evaluate the model performance to ensure that it can accurately predict the wear state corresponding to the new symbol sequence.

[0067] 2) Input the symbol sequence data generated in real time into the trained wear state prediction model to predict the tool wear state; and the symbol transition probability matrix characterizes the future wear trend of the tool;

[0068] S4. Develop a compensation strategy;

[0069] According to the predicted tool wear state and future wear trend, combined with the processing technology requirements and tool characteristics, develop corresponding compensation strategies.

[0070] The method of the present invention is further described by specific embodiments:

[0071] S1. Data acquisition:

[0072] In this embodiment, a variety of high-precision sensors (cutting force, temperature, vibration) are installed at key parts of the numerical control machining equipment, and data is collected in real time at a sampling frequency of 100 Hz.

[0073] The cutting force sensor is installed at the tool clamping part to monitor the force changes during the cutting process in real time; the temperature sensor is placed near the tool to measure the temperature of the cutting area; the vibration sensor is fixed near the tool spindle to capture the tool vibration signal. The sensors are connected to the numerical control system through a data acquisition card, and data is collected in real time at a sampling frequency of 100 Hz. Preferably, when machining a complex part, the sensors will continuously collect the signal data to form a complete data set for subsequent analysis.

[0074] S2. Data processing:

[0075] S21. Filter and denoise the data. A low-pass filter is used to filter out high-frequency noise and retain the main features of the cutting force, temperature, and vibration signals. Specifically, for the cutting force signal, the high-frequency noise caused by equipment vibration is filtered out, and the low-frequency components related to tool wear are retained.

[0076] S22. Unify the data with different dimensions to the same numerical range;

[0077] Perform normalization processing on the data to unify the data with different dimensions to the same numerical range, which is convenient for subsequent symbol mapping and analysis. Specifically, data such as cutting force is normalized to the range of [0, 1] to make it comparable with temperature and vibration data. The preprocessed data forms standardized data, which can more accurately reflect the actual changes during tool wear and provide high-quality input for the generation of symbol sequence data.

[0078] S23. Calculate the mean and standard deviation of the standardized data as statistical features:

[0079] Cutting force:

[0080] Temperature:

[0081] Vibration:

[0082] Among them, F represents the cutting force signal, T represents the temperature signal; V represents the vibration signal; are the means of the cutting force signal, temperature signal, and vibration signal respectively, and S F , S T , S V are the standard deviations of the cutting force signal, temperature signal, and vibration signal respectively.

[0083] Step 2.4: Convert the preprocessed cutting force, temperature, and vibration signals into symbolic sequence data according to the adaptive symbol mapping rule.

[0084] Dynamically divide the symbol intervals: Dynamically divide the symbol intervals of the cutting force signal, temperature signal, and vibration signal according to statistical characteristics. The cutting force signal is divided into three symbol intervals: "low", "medium", and "high", which are represented by the symbols "0", "1", and "2" respectively. The temperature signal is divided into three symbol intervals: "low temperature", "medium temperature", and "high temperature", which are represented by the symbols "3", "4", and "5" respectively. The vibration signal is divided into three symbol intervals: "low frequency", "medium frequency", and "high frequency", which are represented by the symbols "6", "7", and "8" respectively.

[0085]

[0086] Moreover, the adaptive symbol mapping rule can automatically adjust the division of symbol intervals according to the changes in real-time data, making the symbolic sequence more accurately reflect the actual changes in the wear state.

[0087] S25: Generate symbolic sequence data: Convert it into symbolic sequence data according to the above adaptive symbol mapping rule. For example, at a certain time point, if the cutting force signal falls within the "low" symbol interval, it is marked as "0"; if the temperature signal falls within the "medium temperature" symbol interval, it is marked as "4"; if the vibration signal falls within the "high frequency" symbol interval, it is marked as "8", then the symbolic sequence at this time point is "048", that is, each time point consists of three symbols (cutting force, temperature, vibration). This symbolic representation not only simplifies the data structure but also facilitates subsequent pattern recognition and trend prediction. The symbol combinations of multiple time points form a complete symbolic sequence, such as "048047138...". The symbolic sequence can effectively capture the dynamic change law in the wear process, has high information compression ability and anti-noise interference ability, and provides a basis for subsequent wear state prediction and compensation strategy formulation.

[0088] S3: Build a tool wear state prediction model and perform pre-training;

[0089] S31: Data preparation:

[0090] Use unsupervised learning methods to perform clustering analysis on the symbolic sequence and label the clustering results; Define wear pattern classifications, including: slight wear, moderate wear, severe wear, and abnormal states.

[0091] Define specific wear patterns according to the multi-signal combination in combination with the tool wear pattern definition rules to ensure coverage of all possible situations, including:

[0092] Slight wear: The symbol combinations are "046" and "147", characterized by relatively low cutting force, and relatively stable temperature and vibration.

[0093] Moderate wear: The symbol combination is "157", characterized by medium cutting force, and increased temperature and vibration.

[0094] Severe wear: The symbol combination is "258", characterized by high cutting force, and significantly increased temperature and vibration.

[0095] Abnormal state: The symbol combinations are "168" or "249", characterized by abnormal high temperature or severe vibration, which may indicate tool failure or other problems.

[0096] During the process, historical data is used to confirm the pattern to ensure that the defined wear patterns can accurately reflect the actual wear situation.

[0097] Use unsupervised learning methods to perform clustering analysis on the symbol sequences. Train by minimizing the difference between the input symbol sequence and the reconstructed symbol sequence to optimize the model parameters. Ensure that the model can accurately predict the tool wear state without explicit labels.

[0098] Symbol sequence clustering: In the absence of wear labels, it is first necessary to perform clustering analysis on the symbol sequences. The purpose of this step is to find natural groupings or patterns in the data through unsupervised learning methods, providing a basis for subsequent wear pattern annotation. The original symbol sequence is generated by sensor acquisition and through preprocessing and symbol mapping rules. The symbol sequence at each time point consists of three symbols (cutting force, temperature, vibration), such as "046", "147", etc. Suppose there is a symbol sequence: "046147258046", which represents the state change of the tool at different time points.

[0099] In this embodiment, the goal of K-means clustering is to assign the symbol sequences to k clusters, so that the symbol sequences within each cluster are as similar as possible, while the differences between different clusters are as large as possible. The specific K-means clustering process is as follows:

[0100] Initialization: Randomly select k initial center points.

[0101] Assignment: Assign each symbol sequence to the cluster where the nearest center point is located.

[0102] Update: Recalculate the center point of each cluster.

[0103] Repeat: Repeat the assignment and update steps until convergence, that is, the clusters no longer change.

[0104] Annotation: After clustering, it will be found that certain symbol combinations frequently appear in the same cluster. This is because these symbol combinations have similar characteristics and may correspond to similar wear states. For example, if multiple symbol sequences all contain "046" or "147" and they frequently appear in the same cluster, then according to the characteristics of these symbol combinations, they are labeled as slightly worn. The reason is that according to historical experience and professional knowledge, these symbol combinations usually correspond to slightly worn states.

[0105] In this embodiment, "046" indicates low cutting force, moderate temperature, and small vibration, which are usually manifestations of slight wear. In this way, unlabeled data can be converted into labeled data, providing a basis for subsequent supervised learning.

[0106] S32. Build a tool wear state prediction model and perform pre-training;

[0107] The tool wear state prediction model includes: an autoencoder (Autoencoder) based on deep learning, a variational autoencoder (VAE), or a clustering-based model.

[0108] Use cross-validation to optimize the model parameters to ensure the generalization ability of the model.

[0109] In this embodiment, the tool wear state prediction model is an autoencoder; an autoencoder is a neural network designed to reconstruct input data and consists of an encoder and a decoder. The encoder maps the input symbol sequence x to a low-dimensional representation (hidden layer) z, and the decoder reconstructs the low-dimensional representation z back to the original symbol sequence

[0110] Training process: The goal of the autoencoder is to minimize the difference between the input symbol sequence x and the reconstructed symbol sequence The loss function usually uses the mean square error (MSE):

[0111]

[0112] where x i is the original symbol sequence, is the reconstructed symbol sequence, and N is the number of samples.

[0113] Suppose there is an original symbol sequence "046", and after encoding and decoding by the autoencoder, the reconstructed symbol sequence is "046". If the reconstruction result is very close to the original sequence, it means that the model can well capture the characteristics of the symbol sequence.

[0114] To ensure the generalization ability of the model, use cross-validation to optimize the model parameters. Divide the dataset into a training set and a validation set, repeatedly train and evaluate the model, and finally select the model configuration with the optimal performance.

[0115] Evaluate the model performance using the reserved test set to ensure that it can accurately predict the wear state corresponding to new symbol sequences. Evaluation metrics include accuracy, recall rate, F1 score, etc. Suppose there are 100 symbol sequences in the test set, and 50 of them are correctly classified as slightly worn, and the accuracy of the model is 50%.

[0116] If the model performance is not ideal, semi-supervised learning methods can be considered to guide the learning process of the model using a small amount of labeled data: fine-tune by combining the labeled partial data to improve the accuracy of the model. Manual adjustment can also be combined with historical experience, especially for those situations where the model is difficult to distinguish, and historical experience can provide valuable supplementary information. Suppose there is a series of symbol sequences: "046147258046", after K-means clustering, it is found that some symbol combinations (such as "046" and "147") frequently appear in the same cluster. According to the characteristics of these symbol combinations, they are labeled as slightly worn. Then, use an autoencoder to train the model to minimize the difference between the original symbol sequence and the reconstructed symbol sequence. Optimize the model parameters through cross-validation and evaluate the model performance on the test set to finally obtain a model that can accurately predict the wear state.

[0117] S4. Obtain the tool wear state prediction value based on the symbol sequence data through the pre-trained tool wear state prediction model;

[0118] S5. According to the prediction value output by the tool wear state prediction model, combine the processing technology requirements and tool characteristics to formulate corresponding compensation strategies, such as Figure 3 as shown.

[0119] For the tool wear prediction value of slightly worn (such as symbol combinations "046", "147"), adjust the cutting parameters; in this embodiment, preferably, reduce the cutting speed by 5% and increase the feed rate by 10% to slow down the wear rate.

[0120] For the tool wear prediction value of moderately worn (such as symbol combination "157"): perform local grinding of the tool and recalibrate the tool position; in this embodiment, preferably, the grinding amount is 0.1mm.

[0121] For the tool wear prediction value of severely worn (such as symbol combination "258"): immediately replace the tool with a new one and reset the parameters of the tool and the lathe according to the characteristics of the new tool.

[0122] For the tool wear prediction value of abnormal state (such as symbol combinations "168" or "249"): immediately stop the machining, check the tool and other possible problems to avoid further damage.

[0123] Convert the compensation strategy into specific numerical control instructions and control the equipment to execute through the numerical control system. Specifically, the instructions for adjusting the cutting parameters can be directly sent to the parameter setting module of the numerical control system to achieve real-time update of the parameters; the instructions for tool grinding and replacement need to be operated through the tool management module of the numerical control system to ensure the accurate execution of the compensation strategy.

[0124] Embodiment 2

[0125] Combined with Figure 2 As shown, based on steps S1 - S5 of Embodiment 1, in this embodiment, a symbol transition probability matrix is introduced to characterize the future wear trend of the tool; a tool wear compensation strategy is further formulated according to the future wear trend of the tool.

[0126] S6. Input the real-time generated symbol sequence data into the trained wear state prediction model, and combine the trend prediction database of the symbol sequence data updated in real time to predict the future wear trend and predict the current tool wear state.

[0127] A dynamic trend prediction database is formed through the symbol transition probability matrix for trend prediction. When a real-time data is input into the tool wear state prediction model, while obtaining a prediction, the real-time data will also be compared with the database formed by the symbol transition probability matrix for further prediction and verification. This can not only improve the accuracy of prediction, but also capture potential trend changes and abnormal conditions.

[0128] Specifically, based on the trained tool wear state prediction model, the wear state of the real-time data is initially predicted. Combining with the symbol transition probability matrix, the probability of transferring from the current state to the future state is evaluated to provide a more detailed trend prediction. In this way, by combining the two, the system can more accurately judge the current and future wear states and take corresponding compensation strategies in a timely manner.

[0129] S61. The symbol transition probability matrix describes the transition relationship between different symbols and is used as a dynamic database for trend prediction. Its construction steps are as follows:

[0130] 1) Count the transition times: For each symbol category (cutting force, temperature, vibration), count the transition times N between its internal symbols ij , and i and j must be adjacent transitions within the same category and cannot cross categories. Specifically, the cutting force symbol "0" can only transfer to "1" or "2", and cannot directly jump to the temperature symbol "4".

[0131] For the symbol sequence: "046147258046", the data at each time point after decomposition is shown in Table 1:

[0132] Table 1

[0133] Time point Cutting force (F) Temperature (T) Vibration (V) <![CDATA[t1]]> 0 4 6 <![CDATA[t2]]> 1 4 7 <![CDATA[t3]]> 2 5 8 <![CDATA[t4]]> 0 4 6

[0134] As can be seen from Table 1:

[0135] Cutting force symbols (0, 1, 2): The symbol "0" appears 2 times, the symbol "1" appears 1 time, and the symbol "2" appears 1 time.

[0136] Transfer from 0 to 1: 1 time, transfer from 1 to 2: 1 time, transfer from 2 to 0: 1 time.

[0137] Temperature symbols (3, 4, 5): The symbol "4" appears 3 times, and the symbol "5" appears 1 time.

[0138] Transfer from 4 to 4: 1 time, transfer from 4 to 5: 1 time, transfer from 5 to 4: 1 time.

[0139] Vibration symbols (6, 7, 8): The symbol "6" appears 2 times, the symbol "7" appears 1 time, and the symbol "8" appears 1 time.

[0140] Transfer from 6 to 7: 1 time, transfer from 7 to 8: 1 time, transfer from 8 to 6: 1 time.

[0141] 2) Calculate the probability P of transferring from symbol i to symbol j according to the number of transfers ij :

[0142]

[0143] Where N ij is the number of times of transferring from symbol i to symbol j, and ∑ k N ik is the total number of times symbol i appears.

[0144] Cutting force symbol transfer probability matrix:

[0145] 0 appears 2 times and transfers to 1 once:

[0146]

[0147] 1 appears 1 time and transfers to 2 once:

[0148]

[0149] 2 appears 1 time and transfers to 0 once:

[0150]

[0151] Therefore, the cutting force symbol transfer probability matrix is:

[0152]

[0153] Similarly, the temperature symbol transition probability matrix is as follows:

[0154]

[0155] Similarly, the vibration symbol transition probability matrix is as follows:

[0156]

[0157] The symbol transition probability matrix is not only a static probability table but also a dynamic database that is updated in real time and used to predict future wear trends.

[0158] S62. Using the symbol transition probability matrix to assist in predicting the wear state:

[0159] If the real-time symbol sequence is "046", the predicted value of the tool wear state prediction model is slight wear, and the symbol transition probability matrix shows that the most likely future transition is to "147", i.e., moderate wear. Then, based on the compensation strategy for slight wear, the adjustment range of the cutting parameters can be increased. In this embodiment, preferably, the cutting speed is reduced by 10% and the feed rate is increased by 15% to slow down the wear speed in advance.

[0160] If the real-time symbol sequence is "157", the predicted value of the tool wear state prediction model is moderate wear, and the symbol transition probability matrix shows a high probability of future transition to "258", i.e., severe wear. Then, the tool can be replaced to avoid further damage.

[0161] If the real-time symbol sequence is "046", the predicted value of the tool wear state prediction model is slight wear, and the symbol transition probability matrix shows a possible future transition to an abnormal state (such as "168"). This indicates that some unforeseen working condition changes have occurred, and the tool and other equipment should be inspected immediately.

[0162] If the real-time symbol sequence is "168", the predicted value of the tool wear state prediction model is an abnormal state, and the symbol transition probability matrix shows that the future may continue to stay in the abnormal state or turn to a more severe state (such as "269"). Then, the machining should be stopped immediately to inspect the tool and other equipment to prevent further damage.

[0163] S63. The symbol transition probability matrix can help explain why certain symbol combinations occur frequently, providing a more intuitive understanding. If the model predicts moderate wear, but the symbol transition probability matrix shows that the future may return to slight wear, this can be explained as the tool having recovered a good working state after being reground.

[0164] S64. Using the symbol transition probability matrix to monitor abnormal situations;

[0165] If the symbol transition probability matrix shows a sudden transition from a "normal" symbol combination (such as "046") to an abnormal symbol combination (such as "168"), this may indicate an impending tool failure or other problems, and trigger the alarm mechanism in a timely manner. When the symbol sequence stays in the abnormal state (such as "168") for a long time, the symbol transition probability matrix can help identify whether this abnormality is temporary or persistent, so as to decide whether it is necessary to immediately stop the machining.

[0166] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent prediction and compensation method for the wear of a numerically controlled machining tool, characterized in that, Including: Collecting the cutting process data of the CNC machining tool, and performing preprocessing to obtain standardized data; Performing symbol conversion on the standardized data to obtain symbol sequence data; Based on the symbol sequence data, obtaining the tool wear state prediction value through the tool wear state prediction model; Determining the wear compensation strategy according to the tool wear state prediction value; Taking the wear compensation strategy as a control instruction and executing it through the CNC system.

2. The intelligent prediction and compensation method for the wear of a numerically controlled machining tool according to claim 1, characterized in that Also including: Characterizing the future wear trend of the tool through the symbol transition probability matrix; Obtaining the trend compensation strategy by combining the tool wear state prediction value and the future wear trend.

3. The intelligent prediction and compensation method for tool wear in numerical control machining according to claim 1, wherein The cutting process data of the CNC machining tool includes: Cutting force signal, temperature signal, and vibration signal.

4. The intelligent prediction and compensation method for the wear of a numerically controlled machining tool according to claim 1, characterized in that, The performing symbol conversion on the standardized data to obtain symbol sequence data includes: Calculating the mean and standard deviation of the standardized data as statistical features; Dividing the symbol interval based on the statistical features and assigning symbols to the standardized data.

5. The intelligent prediction and compensation method for the wear of a numerically controlled machining tool according to claim 1, wherein The determining the wear compensation strategy according to the tool wear state prediction value includes: If the tool wear state prediction value is slight wear, adjusting the cutting parameters; If the tool wear state prediction value is moderate wear, performing local grinding on the tool; If the tool wear state prediction value is severe wear, replacing the tool; If the tool wear state prediction value is an abnormal state, stopping the machining.

6. The intelligent prediction and compensation method for tool wear in numerical control machining according to claim 2, characterized in that, The obtaining the trend compensation strategy by combining the tool wear state prediction value and the future wear trend includes: If the tool wear state prediction value is slight wear and the future wear trend is moderate wear, increasing the adjustment range of the cutting parameters; If the tool wear state prediction value is moderate wear and the future wear trend is severe wear, replacing the tool; If the tool wear state prediction value is slight wear and the future wear trend is an abnormal state, overhauling the equipment; If the tool wear state prediction value is an abnormal state and the future wear trend is an abnormal state, performing a shutdown process.

7. The intelligent prediction and compensation method for tool wear in numerical control machining according to claim 1, wherein, The obtaining the tool wear state prediction value based on the symbol sequence data through the tool wear state prediction model includes: Performing pre-training on the tool wear state prediction model; Using the pre-trained tool wear state prediction model to input the symbol sequence data and output the tool wear state prediction value.

8. The intelligent prediction and compensation method for tool wear in numerical control machining according to claim 1, characterized in that Performing pre-training on the tool wear state prediction model includes: The tool wear state prediction model is an unsupervised learning model; Performing clustering analysis on the symbol sequence; and annotating the clustering result, mapping each cluster to the corresponding wear mode; Using cross-validation to optimize the model parameters; Minimizing the difference between the input symbol sequence and the reconstructed symbol sequence.

9. The intelligent prediction and compensation method for tool wear in numerical control machining according to claim 4, characterized in that The dividing the symbol interval based on the statistical features and assigning symbols to the standardized data includes: Among them, F represents the cutting force signal, T represents the temperature signal, and V represents the vibration signal; are the mean values of the cutting force signal, temperature signal, and vibration signal, respectively, S F , S T , S V are the standard deviations of the cutting force signal, temperature signal, and vibration signal, respectively.

Images