Intelligent operation and maintenance management method and system for numerical control machine tool

Through artificial intelligence-based data analysis technology, time-series correlation analysis of the vibration, current and voltage data of CNC machine tools is carried out to identify motor status abnormalities, solving the problem that traditional systems cannot capture subtle fault signals, and improving the intelligence of operation and maintenance management and Predictive maintenance capability.

CN119067640AActive Publication Date: 2024-12-03HANGZHOU YOUFU CLOUD TECH CO LTD

Patent Information

Application Number
CN202411558443.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-11-04
Publication Date
2024-12-03
Estimated Expiration
2044-11-04

AI Technical Summary

Technical Problem

The traditional CNC machine tool operation and maintenance management system has limitations in data collection and analysis, and cannot fully reflect the operating status of the machine tool motor, especially ignoring subtle parameters such as vibration, sound and current fluctuations, making it difficult to capture complex equipment failure modes and early warning signals. .

Method used

Using data analysis and processing technology based on artificial intelligence, time sequence analysis is performed on the time queue of vibration amplitude value, current value and voltage value, and features are extracted through timing correlation, local timing coordination mode is identified, motor status detection results are generated, and early failures are achieved. Discover in time.

Benefits of technology

Improve the intelligent and predictive maintenance capabilities of operation and maintenance management, and can capture more information about the operating status of the equipment, especially those subtle changes that may indicate early failures, enhancing the ability to detect and deal with potential problems early.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119067640A_ABST
    Figure CN119067640A_ABST
Patent Text Reader

Abstract

The invention provides an intelligent operation and maintenance management method and system for a numerical control machine tool, relates to the field of intelligent management of equipment, and adopts a data analysis and processing technology based on artificial intelligence to perform time sequence analysis on a time queue of a vibration amplitude value, a time queue of a current value and a time queue of a voltage value. A motor state detection result is intelligently obtained according to a time sequence cooperative response characteristic between a characteristic obtained by carrying out time sequence correlation on current and voltage and a vibration amplitude time sequence characteristic, and more information about an equipment operation state, especially subtle changes possibly indicating an early fault, can be captured. And features in the data can be automatically extracted, and a local time sequence cooperation mode among the data can be identified, so that potential problems can be found as soon as possible, and the intelligent and predictive maintenance capability of operation and maintenance management can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of intelligent equipment management, and more specifically, to an intelligent operation and maintenance management method and system for CNC machine tools. Background Art

[0002] CNC machine tools are key equipment in modern manufacturing, widely used in industries such as automotive, aerospace, and precision instruments. With the continuous advancement of industrial automation, CNC machine tools play a vital role in improving production efficiency and ensuring machining accuracy. Because CNC machine tools play a crucial role in the production process, their operating status is directly related to product quality and production efficiency. Therefore, effective operation and maintenance management of CNC machine tools is crucial to ensuring smooth production.

[0003] However, traditional CNC machine tool operation and maintenance management systems may rely on a limited number of sensors or manual input for data collection, limiting the dimensionality and depth of the data and insufficiently reflecting the actual operating status of the machine tool motor. For example, traditional systems may only monitor basic operating parameters such as speed and temperature, overlooking more subtle vibrations, sounds, or current fluctuations that may contain early signs of a fault. Furthermore, traditional systems may employ simple threshold judgments or statistical analysis methods for data analysis. While easy to implement, these methods often lack the ability to identify deep patterns and anomalies within local time series data, making it difficult to capture complex equipment failure modes and early warning signals. For example, initial wear on a motor may not immediately cause a significant temperature increase or speed drop. However, monitoring even small local changes in vibration or current fluctuations can potentially detect these problems at an early stage.

[0004] Therefore, an optimized CNC machine tool operation and maintenance management solution is needed to solve the above technical problems. Summary of the Invention

[0005] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a method and system for intelligent operation and maintenance management of CNC machine tools, which adopts data analysis and processing technology based on artificial intelligence to perform time series analysis on the time queue of vibration amplitude value, the time queue of current value and the time queue of voltage value, so as to intelligently obtain the motor status detection result based on the characteristics obtained by time series correlation of current and voltage and the time series collaborative response characteristics between the vibration amplitude time series characteristics, and can capture more information about the operating status of the equipment, especially those subtle changes that may indicate early failures. And it can automatically extract features from the data and identify the local time series collaborative patterns between the data, so as to discover potential problems early, thereby improving the intelligence of operation and maintenance management and the ability of predictive maintenance.

[0006] According to one aspect of the present application, there is provided an intelligent operation and maintenance management system for CNC machine tools, comprising: The data acquisition layer includes a data acquisition module and a data transmission module. The data acquisition module is used to receive a time queue of motor state parameters collected by a sensor group of a motor deployed on a CNC machine tool, wherein the motor state parameters include vibration amplitude values, current values, and voltage values; the data transmission module is used to transmit the time queue of motor state parameters to the data processing layer; The data processing layer is used to group the time queue of the motor state parameters according to the parameter sample dimension to obtain a time queue of vibration amplitude values, a time queue of current values, and a time queue of voltage values; The data analysis layer is used to perform feature extraction and correlation analysis on the time queue of the vibration amplitude value, the time queue of the current value, and the time queue of the voltage value to obtain a motor state detection result, wherein the motor state detection result is used to indicate whether the motor state is abnormal; The application layer is configured to generate an alarm notification in response to the motor state detection result indicating that the motor state is abnormal.

[0007] In the above-mentioned intelligent operation and maintenance management system for CNC machine tools, the data analysis layer includes: a motor state parameter timing encoding module, which is used to perform timing sequence encoding on the time queue of the vibration amplitude value, the time queue of the current value and the time queue of the voltage value respectively to obtain the vibration amplitude timing correlation feature vector, the current timing correlation feature vector and the voltage timing correlation feature vector; a motor working power factor timing correlation representation generation module, which is used to input the current timing correlation feature vector and the voltage timing correlation feature vector into the motor working power timing analysis module to obtain the motor working power factor timing correlation representation vector; a motor power factor-vibration coordinated response analysis module, which is used to perform motor power factor-vibration timing feature coordinated response analysis on the motor working power factor timing correlation representation vector and the vibration amplitude timing correlation feature vector to obtain the motor power factor-vibration coordinated response feature; a motor state detection result generation module, which is used to generate the motor state parameter timing correlation representation vector and the vibration amplitude timing correlation feature vector to obtain the motor power factor-vibration coordinated response feature. The motor state detection result is obtained based on the motor power factor-vibration coordinated response feature; wherein the motor power factor-vibration coordinated response analysis module includes: a vibration amplitude time series correlation feature refinement unit, used to perform feature refinement on the vibration amplitude time series correlation feature vector to obtain a sequence of vibration amplitude local time series correlation feature vectors; a motor work power factor-vibration correlation unit, used to calculate the correlation matrix between the motor work power factor time series correlation representation vector and each vibration amplitude local time series correlation feature vector in the sequence of the vibration amplitude local time series correlation feature vector to obtain a sequence of motor work power factor-vibration correlation matrices; a motor power factor-vibration local-global joint response unit, used to perform local and global joint response evaluation on the sequence of the motor work power factor-vibration correlation matrix to obtain a motor power factor-vibration response interaction expression matrix as the motor power factor-vibration coordinated response feature.

[0008] In the above-mentioned intelligent operation and maintenance management system for CNC machine tools, the motor state parameter timing encoding module is used to: input the time queue of the vibration amplitude value, the time queue of the current value and the time queue of the voltage value into the timing sequence encoder based on the Bi-LSTM model respectively to obtain the vibration amplitude timing association feature vector, the current timing association feature vector and the voltage timing association feature vector.

[0009] In the above-mentioned intelligent operation and maintenance management system for CNC machine tools, the motor working power factor-vibration correlation unit includes: a motor working power factor time series correlation representation vector nonlinear transformation subunit, which is used to perform a nonlinear transformation on the motor working power factor time series correlation representation vector to obtain the motor working power factor time series correlation representation vector after nonlinear transformation; a motor working power factor-vibration correlation matrix sequence calculation subunit, which is used to calculate the correlation matrix between the motor working power factor time series correlation representation vector after nonlinear transformation and each vibration amplitude local time series correlation eigenvector in the sequence of the vibration amplitude local time series correlation eigenvector to obtain the sequence of the motor working power factor-vibration correlation matrix.

[0010] In the above-mentioned intelligent operation and maintenance management system for CNC machine tools, the motor power factor-vibration local-global joint response unit includes: a motor work power factor-vibration correlation characteristic interaction energy factor calculation subunit, which is used to calculate the characteristic interaction energy factor of each motor work power factor-vibration correlation matrix in the sequence of the motor work power factor-vibration correlation matrix to obtain a sequence of motor work power factor-vibration correlation characteristic interaction energy factors; a motor work power factor-vibration correlation characteristic interaction energy factor normalization subunit, which is used to input the sequence of the motor work power factor-vibration correlation characteristic interaction energy factors into a Sigmoid function to obtain a sequence of normalized motor work power factor-vibration correlation characteristic interaction energy factors; and a motor power factor-vibration response interaction expression matrix generation subunit, which is used to use the sequence of the normalized motor work power factor-vibration correlation characteristic interaction energy factors as a sequence of weights to calculate the position-weighted sum of the sequence of the motor work power factor-vibration correlation matrix to obtain the motor power factor-vibration response interaction expression matrix.

[0011] In the above-mentioned intelligent operation and maintenance management system for CNC machine tools, the motor working power factor-vibration correlation characteristic interaction energy factor calculation subunit includes: a motor working power factor-vibration correlation kurtosis value calculation secondary subunit, which is used to calculate the kurtosis value of each motor working power factor-vibration correlation matrix in the sequence of the motor working power factor-vibration correlation matrix to obtain a sequence of motor working power factor-vibration correlation kurtosis values; a motor working power factor-vibration interaction energy factor calculation secondary subunit, which is used to calculate the characteristic interaction energy factor of each motor working power factor-vibration correlation matrix based on the each motor working power factor-vibration correlation kurtosis value in the sequence of the motor working power factor-vibration correlation kurtosis value to obtain a sequence of the motor working power factor-vibration correlation characteristic interaction energy factors.

[0012] In the above-mentioned intelligent operation and maintenance management system for CNC machine tools, the motor working power factor-vibration related kurtosis value calculation secondary subunit is used to: respectively calculate the mean and variance of the motor working power factor-vibration related matrix to obtain the motor working power factor-vibration related mean and the motor working power factor-vibration related variance; calculate the positional difference between the motor working power factor-vibration related matrix and the motor working power factor-vibration related mean to obtain the motor working power factor-vibration related difference matrix; calculate the fourth power of each eigenvalue in the motor working power factor-vibration related difference matrix to obtain the motor working power factor-vibration related modulation difference matrix; calculate the expected value of the motor working power factor-vibration related modulation difference matrix to obtain the motor working power factor-vibration related expected value; divide the motor working power factor-vibration related expected value by the square of the motor working power factor-vibration related variance to obtain the motor working power factor-vibration related kurtosis value.

[0013] In the above-mentioned intelligent operation and maintenance management system for CNC machine tools, the motor working power factor-vibration interaction energy factor calculation secondary subunit is used to: calculate the sum of the motor working power factor-vibration correlation variance and the hyperparameter to obtain the motor working power factor-vibration correlation first energy factor; calculate the square of the difference between the motor working power factor-vibration correlation kurtosis value and the motor working power factor-vibration correlation mean to obtain the motor working power factor-vibration correlation difference value; add the value obtained by multiplying the motor working power factor-vibration correlation variance by a constant two and the value obtained by multiplying the hyperparameter by a constant two to the motor working power factor-vibration correlation difference value to obtain the motor working power factor-vibration correlation second energy factor; divide the motor working power factor-vibration correlation first energy factor by the motor working power factor-vibration correlation second energy factor to obtain the motor working power factor-vibration correlation characteristic interaction energy factor corresponding to the motor working power factor-vibration correlation matrix.

[0014] In the above-mentioned intelligent operation and maintenance management system for CNC machine tools, the motor state detection result generation module is used to: input the motor power factor-vibration response interaction expression matrix into the classifier-based motor state detection module to obtain the motor state detection result.

[0015] According to another aspect of the present application, a method for intelligent operation and maintenance management of a CNC machine tool is provided, comprising: A time queue for collecting motor state parameters by a sensor group deployed on a motor of a CNC machine tool, wherein the motor state parameters include vibration amplitude value, current value and voltage value; Grouping the time queue of the motor state parameters according to the parameter sample dimension to obtain a time queue of vibration amplitude values, a time queue of current values, and a time queue of voltage values; performing feature extraction and correlation analysis on the time queue of the vibration amplitude value, the time queue of the current value, and the time queue of the voltage value to obtain a motor state detection result, wherein the motor state detection result is used to indicate whether the motor state is abnormal; In response to the motor state detection result indicating that the motor state is abnormal, an alarm notification is generated.

[0016] The present application has at least the following technical effects: Compared with the existing technology, the CNC machine tool intelligent operation and maintenance management method and system provided by the present application adopts artificial intelligence-based data analysis and processing technology to perform time series analysis on the time series of vibration amplitude values, current values, and voltage values. In this way, the motor status detection results are intelligently obtained based on the characteristics obtained by time series correlation of current and voltage and the time series collaborative response characteristics between the vibration amplitude time series characteristics. More information about the equipment operating status can be captured, especially those subtle changes that may indicate early failures. It can also automatically extract features from the data and identify local time series collaborative patterns between the data, thereby discovering potential problems early and improving the intelligent operation and maintenance management and predictive maintenance capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work. In the drawings: Figure 1 This is a block diagram of the intelligent operation and maintenance management system for CNC machine tools according to an embodiment of the present application.

[0018] Figure 2 This is a block diagram of the data analysis layer in the intelligent operation and maintenance management system of CNC machine tools according to an embodiment of the present application.

[0019] Figure 3 Schematic diagram of data flow in the data analysis layer of the intelligent operation and maintenance management system for CNC machine tools according to an embodiment of the present application.

[0020] Figure 4 4 is a block diagram of a motor dynamic factor-vibration coordinated response analysis module in an intelligent operation and maintenance management system for CNC machine tools according to an embodiment of the present application.

[0021] Figure 5 This is a block diagram of a motor working power factor-vibration correlation unit in the intelligent operation and maintenance management system of a CNC machine tool according to an embodiment of the present application.

[0022] Figure 6It is a block diagram of the motor dynamic factor-vibration local and global joint response unit in the intelligent operation and maintenance management system of CNC machine tools according to an embodiment of the present application.

[0023] Figure 7 Flowchart of the intelligent operation and maintenance management method for CNC machine tools according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described herein.

[0025] As core equipment in modern manufacturing, CNC machine tools play a vital role in a variety of fields, including automotive, aerospace, and precision instrument manufacturing. With the advancement of industrial automation, CNC machine tools are crucial for improving production efficiency and ensuring machining accuracy. Their operating status affects product quality and production efficiency, making effective operation and maintenance management crucial to ensuring smooth production.

[0026] However, traditional CNC machine tool operation and maintenance management systems have limitations in data collection. They usually rely on limited sensor data or manual input, which limits the comprehensive understanding of the machine tool's operating status. For example, these systems may only monitor basic parameters such as speed and temperature, while ignoring more subtle indicators such as vibration, sound and current fluctuations, which may indicate early signs of failure. In addition, traditional systems may rely too much on simple threshold judgments or statistical analysis in data analysis, lacking the ability to identify deep patterns and anomalies in the data, which limits the capture of complex equipment failure modes and early warning signals. For example, the initial wear of a motor may not immediately lead to a significant temperature increase or speed drop, but if small vibration changes or current fluctuations can be monitored, it is possible to detect and solve the problem at an early stage.

[0027] Based on this, this application proposes an intelligent operation and maintenance management system for CNC machine tools. Figure 1 FIG is a block diagram of an intelligent operation and maintenance management system for CNC machine tools according to an embodiment of the present application. Figure 1As shown, in the CNC machine tool intelligent operation and maintenance management system 100, it includes: a data acquisition layer 110, a data processing layer 120, a data analysis layer 130 and an application layer 140; the data acquisition layer 110 includes a data acquisition module and a data transmission module, the data acquisition module is used to receive a time queue of motor state parameters collected by a sensor group of a motor deployed on the CNC machine tool, wherein the motor state parameters include vibration amplitude value, current value and voltage value; the data transmission module is used to transmit the time queue of the motor state parameters to the data processing layer; the data processing layer 120 is used to group the time queue of the motor state parameters according to the parameter sample dimension to obtain a time queue of vibration amplitude value, a time queue of current value and a time queue of voltage value; the data analysis layer 130 is used to perform feature extraction and association analysis on the time queue of vibration amplitude value, the time queue of current value and the time queue of voltage value to obtain a motor state detection result, and the motor state detection result is used to indicate whether the motor state is abnormal; the application layer 140 is used to generate an alarm notification in response to the motor state detection result indicating that the motor state is abnormal.

[0028] In an embodiment of the present application, the data acquisition module in the data acquisition layer 110 is configured to receive a time series of motor state parameters collected by a sensor group deployed on a CNC machine tool's motor. These motor state parameters include vibration amplitude, current, and voltage. Vibration is a key indicator of the operating status of mechanical equipment. When the machine tool is operating normally, the motor's vibration signal should be stable. However, when the motor experiences problems such as wear or imbalance, the vibration signal will exhibit abnormal variations. These variations can be detected before other parameters such as temperature or speed change. The motor's current directly reflects the motor's load. When the machine tool is operating normally, the current value should fluctuate within a specific range. However, when the motor experiences problems such as seizure or overload, the current value can vary significantly, potentially manifesting as a sudden increase or decrease. By monitoring the current, abnormal motor loads can be detected promptly, providing early warnings. The motor's voltage reflects the stability and quality of the power supply. Voltage fluctuations can cause unstable motor operation and even damage the motor. Excessively high or low voltages can damage the motor's insulation system, leading to motor failure. In general, the time queue data of the motor state parameters of vibration amplitude, current value, and voltage value is the basis for judging the abnormal state of the motor. This data can not only provide comprehensive monitoring of the motor operating status, but also help the operation and maintenance management system to more accurately identify potential faults through time series analysis and multi-dimensional collaborative analysis, thereby achieving early warning and efficient operation and maintenance. Accordingly, considering that data dispersion analysis will lead to increased complexity in data management, in order to effectively integrate and analyze monitoring data from different data sources, in the technical solution of this application, the data transmission module in the data acquisition layer 110 is used to transmit the time queue of the motor state parameters to the data processing layer.

[0029] In an embodiment of the present application, the data processing layer 120 is used to group the time queues of the motor state parameters according to the parameter sample dimension to obtain time queues of vibration amplitude values, current values, and voltage values. The motor state parameters are temporally correlated with each other, and grouping the motor state parameter monitoring data collected by different sensors can make the data structure clearer, facilitate subsequent temporal analysis and correlation processing, and thus better assist the system in fault diagnosis.

[0030] In an embodiment of the present application, the data analysis layer 130 is used to perform feature extraction and correlation analysis on the time queue of the vibration amplitude value, the time queue of the current value, and the time queue of the voltage value to obtain a motor state detection result, and the motor state detection result is used to indicate whether the motor state is abnormal. It should be understood that the motor is a core component of a CNC machine tool, and its operating state directly affects the safety and production efficiency of the equipment. By analyzing the collected time queue data of the vibration amplitude, current value, and voltage value, it is possible to detect whether the motor state is abnormal, which is conducive to timely discovering potential faults and facilitating enterprises to respond quickly, thereby reducing the risk of unexpected downtime, ensuring the continuity and stability of the production line, and reducing economic losses caused by faults.

[0031] In the embodiment of the present application, the application layer 140 is configured to generate an alarm notification in response to the motor state detection result indicating that the motor state is abnormal. Specifically, once an abnormal motor state is detected, the application layer can quickly generate an alarm notification to ensure that relevant operation and maintenance personnel are promptly notified of the abnormal situation and take appropriate measures to address it.

[0032] Accordingly, in the data analysis layer, the technical concept of the present application is to use artificial intelligence-based data analysis and processing technology to perform time series analysis on the time queues of vibration amplitude values, current values, and voltage values ​​obtained by grouping the data processing layer. In this way, the motor status detection results are intelligently obtained based on the characteristics obtained by time series correlation of current and voltage and the time series collaborative response characteristics between the vibration amplitude time series characteristics. More information about the operating status of the equipment can be captured, especially those subtle changes that may indicate early failures. It can also automatically extract features from the data and identify local time series collaborative patterns between the data, so as to discover potential problems early, thereby improving the intelligence of operation and maintenance management and the ability of predictive maintenance.

[0033] Figure 2 This is a block diagram of the data analysis layer in the intelligent operation and maintenance management system of CNC machine tools according to an embodiment of the present application. Figure 3 Schematic diagram of data flow in the data analysis layer of the intelligent operation and maintenance management system for CNC machine tools according to an embodiment of the present application. Figure 2 and Figure 3As shown, the data analysis layer 130 includes: a motor state parameter time series encoding module 131, which is used to perform time series sequence encoding on the time queue of the vibration amplitude value, the time queue of the current value, and the time queue of the voltage value to obtain a vibration amplitude time series correlation feature vector, a current time series correlation feature vector, and a voltage time series correlation feature vector; a motor work power factor time series correlation representation generation module 132, which is used to input the current time series correlation feature vector and the voltage time series correlation feature vector into the motor work power timing analysis module to obtain a motor work power factor time series correlation representation vector; a motor power factor-vibration coordinated response analysis module 133, which is used to perform motor power factor-vibration time series feature coordinated response analysis on the motor work power factor time series correlation representation vector and the vibration amplitude time series correlation feature vector to obtain a motor power factor-vibration coordinated response feature; and a motor state detection result generation module 134, which is used to obtain the motor state detection result based on the motor power factor-vibration coordinated response feature.

[0034] In an embodiment of the present application, the motor state parameter timing encoding module 131 is used to perform timing sequence encoding on the time queue of the vibration amplitude value, the time queue of the current value, and the time queue of the voltage value to obtain a vibration amplitude timing correlation feature vector, a current timing correlation feature vector, and a voltage timing correlation feature vector. Specifically, in an embodiment of the present application, the motor state parameter timing encoding module is used to input the time queue of the vibration amplitude value, the time queue of the current value, and the time queue of the voltage value into a timing sequence encoder based on a Bi-LSTM model to obtain the vibration amplitude timing correlation feature vector, the current timing correlation feature vector, and the voltage timing correlation feature vector. It should be understood that there are different timing correlation characteristics between each data point at different times in the time queue of the vibration amplitude value, the time queue of the current value, and the time queue of the voltage value, such as the changing trend of vibration amplitude and the current value fluctuation or the periodic change of voltage value. Taking into account that the Bi-LSTM model can simultaneously consider the forward and reverse information of the input time queue, thereby more comprehensively understanding the dynamic characteristics of the input time queue, based on this, in order to better capture and refine the timing correlation characteristics between the data parameters between each time span, in the technical solution of the present application, the time queue of the vibration amplitude value, the time queue of the current value and the time queue of the voltage value are respectively input into the timing sequence encoder based on the Bi-LSTM model to obtain the vibration amplitude timing correlation feature vector, the current timing correlation feature vector and the voltage timing correlation feature vector.

[0035] In an embodiment of the present application, the motor working power factor timing correlation representation generation module 132 is used to input the current timing correlation feature vector and the voltage timing correlation feature vector into the motor working power timing analysis module to obtain the motor working power factor timing correlation representation vector. Specifically, in an embodiment of the present application, the motor working power factor timing correlation representation generation module is used to: calculate the position point multiplication between the current timing correlation feature vector and the voltage timing correlation feature vector to obtain the motor working power factor timing correlation representation vector. It should be understood that during the operation of the machine tool motor, there is a close timing correlation relationship between the voltage and the current. For example, under the same load conditions, if the voltage decreases, the current may increase accordingly to maintain the power output required by the motor. Therefore, in the technical solution of the present application, the current timing correlation feature vector and the voltage timing correlation feature vector are input into the motor working power timing analysis module to capture and mine the timing dynamic correlation relationship between the voltage and the current, and obtain the motor working power factor timing correlation representation vector.

[0036] In an embodiment of the present application, the motor power factor-vibration coordinated response analysis module 133 is used to perform a motor power factor-vibration timing characteristic coordinated response analysis on the motor working power factor timing correlation representation vector and the vibration amplitude timing correlation feature vector to obtain the motor power factor-vibration coordinated response feature. Accordingly, considering that the motor working power factor timing correlation representation vector expresses the dynamic characteristics of the motor that change with time during operation, that is, the influence of the current and voltage changes of the motor at different time points on the motor power output. The vibration amplitude timing correlation feature vector expresses the vibration characteristics of the motor that change with time during operation. Moreover, these two vectors have different degrees of correlation within the local time. Therefore, in order to analyze the temporal relationship between the motor's dynamic characteristics and vibration conditions in a fine-grained manner, thereby being able to comprehensively and fully identify the correlation between the motor's dynamic changes and vibration amplitude during operation, in the technical solution of the present application, the motor's working dynamic factor temporal correlation representation vector and the vibration amplitude temporal correlation feature vector are subjected to a motor dynamic factor-vibration temporal feature collaborative response analysis to obtain a motor dynamic factor-vibration response interaction expression matrix as the motor dynamic factor-vibration collaborative response feature. In other words, the motor dynamic factor-vibration temporal feature collaborative response analysis is a feature processing technology that combines feature refinement and feature relationship measurement, and is used to meticulously analyze and fuse various information in the feature vector, accurately identify and understand the relationship between different features, and thus provide more reliable data support for subsequent anomaly detection.

[0037] Specifically, first, the vibration amplitude time-series correlation feature vector is subjected to feature refinement processing to obtain a sequence of local vibration amplitude time-series correlation feature vectors. It should be understood that feature refinement processing can more accurately analyze the changes in vibration amplitude over a short period of time (i.e., within a local time window), provide more detailed insight into the local subtle patterns of vibration amplitude changes over time, and enhance the ability to capture local features. Then, the motor working power factor time-series correlation representation vector is subjected to nonlinear transformation processing to enrich the complex time-series representation of the motor dynamic factors, thereby obtaining a nonlinearly transformed motor working power factor time-series correlation representation vector. Next, a correlation matrix is ​​calculated between each local vibration amplitude feature and the nonlinear transformation feature to obtain a sequence of motor working power factor-vibration correlation matrices. In other words, the correlation matrix can comprehensively reflect the temporal interaction between the motor working power factor and the vibration amplitude. This method allows for a more intuitive visualization of the relationship between the two, facilitating analysis of how dynamic changes during motor operation affect the vibration amplitude. Subsequently, the characteristic interaction energy factor of each correlation matrix is ​​calculated to obtain a sequence of motor working power factor-vibration correlation feature interaction energy factors. In particular, the characteristic interaction energy factors measure the energy of the interaction between the motor's operating dynamics and vibration amplitude. In other words, by calculating these factors, the strength and importance of the interaction between the two can be specifically quantified, thereby better understanding the impact of dynamic changes on vibration under different motor operating states. Furthermore, each characteristic interaction energy factor is normalized using a sigmoid function to a value range within the (0, 1) interval. This is converted into an attention weight to intuitively demonstrate the contribution of each feature to the final detection result, while highlighting those features that are more important for motor state detection. Finally, the sequence of normalized factors is used as a sequence of weights to weight and sum the sequence of the motor's operating dynamics-vibration correlation matrix to obtain the motor dynamics-vibration response interaction matrix. Specifically, the response information of the local features of the motor dynamics and vibration amplitude is combined to provide a comprehensive response representation at the global level, improving the model's grasp of the overall characteristic response pattern.

[0038] Specifically, Figure 4 FIG is a block diagram of a motor dynamic factor-vibration coordinated response analysis module in an intelligent operation and maintenance management system for CNC machine tools according to an embodiment of the present application. Figure 4As shown, the motor power factor-vibration coordinated response analysis module 133 includes: a vibration amplitude time series correlation feature refinement unit 1331, which is used to perform feature refinement on the vibration amplitude time series correlation feature vector to obtain a sequence of vibration amplitude local time series correlation feature vectors; a motor work power factor-vibration correlation unit 1332, which is used to calculate the correlation matrix between the motor work power factor time series correlation representation vector and each vibration amplitude local time series correlation feature vector in the sequence of vibration amplitude local time series correlation feature vectors to obtain a sequence of motor work power factor-vibration correlation matrices; a motor power factor-vibration local-global joint response unit 1333, which is used to perform local and global joint response evaluation on the sequence of motor work power factor-vibration correlation matrices to obtain a motor power factor-vibration response interaction expression matrix as the motor power factor-vibration coordinated response feature.

[0039] More specifically, Figure 5 FIG. 1 is a block diagram of a motor working power factor-vibration correlation unit in an intelligent operation and maintenance management system for CNC machine tools according to an embodiment of the present application. Figure 5 As shown, the motor working power factor-vibration correlation unit 1332 includes: a motor working power factor time series correlation representation vector nonlinear transformation subunit 13321, used to perform a nonlinear transformation on the motor working power factor time series correlation representation vector to obtain the motor working power factor time series correlation representation vector after nonlinear transformation; a motor working power factor-vibration correlation matrix sequence calculation subunit 13322, used to calculate the correlation matrix between the motor working power factor time series correlation representation vector after nonlinear transformation and each vibration amplitude local time series correlation eigenvector in the sequence of the vibration amplitude local time series correlation eigenvector to obtain the sequence of the motor working power factor-vibration correlation matrix.

[0040] In an embodiment of the present application, specifically, the nonlinear transformation subunit of the motor working power factor timing correlation representation vector is used to: multiply the motor working power factor timing correlation representation vector by the motor working power factor transformation matrix, and then add the obtained motor working power factor transformation vector and the motor working power factor bias vector according to position to obtain the nonlinearly transformed motor working power factor timing correlation representation vector.

[0041] In an embodiment of the present application, specifically, the motor working power factor-vibration correlation matrix sequence calculation subunit is used to: vector multiply the transpose vector of the motor working power factor timing correlation representation vector after the nonlinear transformation and each vibration amplitude local timing correlation eigenvector in the sequence of the vibration amplitude local timing correlation eigenvector to obtain the sequence of the motor working power factor-vibration correlation matrix.

[0042] More specifically, Figure 6FIG is a block diagram of a motor dynamic factor-vibration local-global joint response unit in an intelligent operation and maintenance management system for CNC machine tools according to an embodiment of the present application. Figure 6 As shown, the motor power factor-vibration local-global joint response unit 1333 includes: a motor work power factor-vibration correlation characteristic interaction energy factor calculation subunit 13331, used to calculate the characteristic interaction energy factor of each motor work power factor-vibration correlation matrix in the sequence of the motor work power factor-vibration correlation matrix to obtain a sequence of motor work power factor-vibration correlation characteristic interaction energy factors; a motor work power factor-vibration correlation characteristic interaction energy factor normalization subunit 13332, used to input the sequence of the motor work power factor-vibration correlation characteristic interaction energy factors into a sigmoid function to obtain a sequence of normalized motor work power factor-vibration correlation characteristic interaction energy factors; and a motor power factor-vibration response interaction expression matrix generation subunit 13333, used to calculate the position-weighted sum of the sequence of the motor work power factor-vibration correlation matrix using the sequence of the normalized motor work power factor-vibration correlation characteristic interaction energy factors as a sequence of weights to obtain the motor work power factor-vibration response interaction expression matrix.

[0043] More specifically, in an embodiment of the present application, the motor working power factor-vibration correlation characteristic interaction energy factor calculation subunit includes: a motor working power factor-vibration correlation kurtosis value calculation secondary subunit, which is used to calculate the kurtosis value of each motor working power factor-vibration correlation matrix in the sequence of the motor working power factor-vibration correlation matrix to obtain a sequence of motor working power factor-vibration correlation kurtosis values; a motor working power factor-vibration interaction energy factor calculation secondary subunit, which is used to calculate the characteristic interaction energy factor of each motor working power factor-vibration correlation matrix based on each motor working power factor-vibration correlation kurtosis value in the sequence of the motor working power factor-vibration correlation kurtosis value to obtain a sequence of the motor working power factor-vibration correlation characteristic interaction energy factors.

[0044] More specifically, in an embodiment of the present application, the motor work power factor-vibration associated kurtosis value calculation secondary subunit is used to: respectively calculate the mean and variance of the motor work power factor-vibration associated matrix to obtain the motor work power factor-vibration associated mean and the motor work power factor-vibration associated variance; calculate the positional difference between the motor work power factor-vibration associated matrix and the motor work power factor-vibration associated mean to obtain the motor work power factor-vibration associated difference matrix; calculate the fourth power of each eigenvalue in the motor work power factor-vibration associated difference matrix to obtain the motor work power factor-vibration associated modulation difference matrix; calculate the expected value of the motor work power factor-vibration associated modulation difference matrix to obtain the motor work power factor-vibration associated expected value; divide the motor work power factor-vibration associated expected value by the square of the motor work power factor-vibration associated variance to obtain the motor work power factor-vibration associated kurtosis value.

[0045] More specifically, in an embodiment of the present application, the motor work power factor-vibration interaction energy factor calculation secondary subunit is used to: calculate the sum of the motor work power factor-vibration correlation variance and the hyperparameter to obtain the motor work power factor-vibration correlation first energy factor; calculate the square of the difference between the motor work power factor-vibration correlation kurtosis value and the motor work power factor-vibration correlation mean to obtain the motor work power factor-vibration correlation difference value; add the value obtained by multiplying the motor work power factor-vibration correlation variance by a constant two and the value obtained by multiplying the hyperparameter by a constant two to the motor work power factor-vibration correlation difference value to obtain the motor work power factor-vibration correlation second energy factor; divide the motor work power factor-vibration correlation first energy factor by the motor work power factor-vibration correlation second energy factor to obtain the motor work power factor-vibration correlation characteristic interaction energy factor corresponding to the motor work power factor-vibration correlation matrix.

[0046] In the embodiment of the present application, specifically, the motor power factor-vibration coordinated response analysis module is used to perform a coordinated response analysis between the timing characteristics of the motor working power factor timing correlation representation vector and the vibration amplitude timing correlation feature vector, which can be expressed as follows: in, is the vibration amplitude time series correlation eigenvector, For feature refinement operation, are the first, second, ..., and third in the sequence of the local time series correlation feature vectors of the vibration amplitude. ,..., The local temporal correlation eigenvector of the vibration amplitude, is the number of eigenvectors in the sequence of the local time series associated eigenvectors of the vibration amplitude, The motor working power is represented by a timing-related vector, and are the motor working power factor transformation matrix and the motor working power factor bias vector respectively, is the temporal correlation vector representing the motor working power after the nonlinear transformation, yes The transposed vector of represents vector multiplication, For the The motor working power factor-vibration correlation matrix between the local time series correlation feature vector of the vibration amplitude and the motor working power factor time series correlation representation vector after the nonlinear transformation, yes The eigenvalues ​​at each position in , yes The mean of To calculate the expected value of the matrix, yes The variance of for The kurtosis value of is a hyperparameter, yes The corresponding motor working power factor-vibration correlation characteristic interaction energy factor, yes function, yes The corresponding normalized motor working power factor-vibration correlation characteristic interaction energy factor, It is the point product of position. is the motor dynamics-vibration response interaction expression matrix.

[0047] In an embodiment of the present application, the motor state detection result generation module 134 is used to obtain the motor state detection result based on the motor power factor-vibration coordinated response feature. Specifically, in an embodiment of the present application, the motor state detection result generation module is used to: input the motor power factor-vibration response interaction expression matrix into the motor state detection module based on the classifier to obtain the motor state detection result. That is, the motor power factor-vibration coordinated response feature obtained by performing a coordinated response analysis between timing features using the motor working power factor timing correlation representation vector and the vibration amplitude timing correlation feature vector is classified and processed, so as to intelligently obtain the motor state detection result, which can capture more information about the operating status of the equipment, especially those subtle changes that may indicate early failures. It can also automatically extract features from the data and identify local timing coordinated patterns between each data, so as to discover potential problems early, thereby improving the intelligence of operation and maintenance management and the ability of predictive maintenance.

[0048] Here, in the preferred example, the motor working power factor time series correlation representation vector and the vibration amplitude time series correlation feature vector represent the motor current-motor voltage time series cooperative correlation feature and the vibration amplitude time series correlation feature, respectively. Therefore, when the motor working power factor time series correlation representation vector and the vibration amplitude time series correlation feature vector are input into the local-global collaborative response analysis module based on feature decoupling, the motor working power factor time series correlation representation vector and the vibration amplitude time series correlation feature vector due to the source domain time series distribution difference and encoding method difference will be transmitted in the local-global collaborative response analysis process based on feature decoupling, so that the motor power factor-vibration response interaction expression matrix has a diversified set expression distribution of cross-modal aggregate response features. Therefore, it is expected to improve the balance between the regression mapping accuracy and completeness when the motor power factor-vibration response interaction expression matrix is ​​input into the motor state detection module based on the classifier for classification regression, thereby improving the accuracy of the obtained motor state detection result.

[0049] In a preferred example, inputting the motor power factor-vibration response interaction expression matrix into a classifier-based motor state detection module to obtain a motor state detection result includes: Expanding the motor power factor-vibration response interaction expression matrix into a motor power factor-vibration response interaction expression vector; Performing L random sampling on the motor power factor-vibration response interaction expression vector to obtain L motor power factor-vibration response interaction random perturbation eigenvalues; Arrange the L motor power factor-vibration response interaction random disturbance eigenvalues ​​in order from large to small to obtain a motor power factor-vibration response interaction random disturbance ordered vector; Calculating the inner product of the motor power factor-vibration response interaction expression vector to obtain the motor power factor-vibration response interaction energy modulation eigenvalue; Performing semantic intensity scaling on the motor power factor-vibration response interaction random perturbation ordered vector based on the motor power factor-vibration response interaction energy modulation characteristic value to obtain a semantically enhanced motor power factor-vibration response interaction random perturbation ordered vector; Calculating the product of the random perturbation ordered vector of the semantic reinforcement motor power factor-vibration response interaction and the transposed vector of the motor power factor-vibration response interaction expression vector to obtain the motor power factor-vibration response interaction best example prompt orthogonal modulation matrix; Calculating the matrix product between the motor power factor-vibration response interaction best example prompt orthogonal modulation matrix and the motor power factor-vibration response interaction expression vector to obtain an optimized motor power factor-vibration response interaction expression vector; The optimized motor dynamic factor-vibration response interaction expression vector is input into the classifier-based motor state detection module to obtain the motor state detection result.

[0050] The above optimization process is expressed as follows: in, represents a normal distribution, represents the Gaussian density probability function, and represent the mean and variance of the motor dynamics-vibration response interaction expression vector, respectively, represents a random variable that follows a normal distribution, represents the number of random samplings, Indicates the The random disturbance eigenvalue of the interaction between the motor dynamic factor and the vibration response, represents the motor dynamics-vibration response interaction expression vector, The first expression vector representing the motor dynamic factor-vibration response interaction eigenvalues, It means to calculate the average value of the vector. represents the length of the motor dynamics-vibration response interaction expression vector, Indicates that the random disturbance eigenvalues ​​of the motor dynamic factor-vibration response interaction are arranged in order from large to small. represents the ordered vector of random disturbances caused by the interaction between motor dynamics and vibration response, represents the motor dynamic factor-vibration response interaction energy modulation eigenvalue, represents the square of the vector's two-norm, Indicates point multiplication by position, represents matrix multiplication, represents the transpose of a vector, The best example of the interaction between motor dynamics and vibration response is the orthogonal modulation matrix, Represents the optimized motor dynamics-vibration response interaction expression vector.

[0051] Accordingly, the motor force factor-vibration response interaction expression vector, obtained by expanding the motor force factor-vibration response interaction expression matrix, is randomly sampled L times to obtain a set of new eigenvalues ​​based on the original features but containing slight variations. This introduces randomness to explore different aspects of the feature space and, by simulating the noise or variation in real data, helps the model better generalize to unseen data. Next, these motor force factor-vibration response interaction random perturbation eigenvalues ​​are sorted from large to small to form an ordered motor force factor-vibration response interaction random perturbation vector. This helps identify which perturbation directions are likely to be most important for the current problem. This sorting not only helps the algorithm focus on the feature differences that are most capable of distinguishing different categories, but also helps reveal hidden data structures or patterns. Subsequently, the inner product of the motor power factor-vibration response interaction expression vector is calculated to obtain the motor power factor-vibration response interaction energy modulation eigenvalue. The motor power factor-vibration response interaction energy modulation eigenvalue reflects the overall strength or "energy" of the original feature vector and provides a benchmark for measuring the importance of the feature. Using this motor power factor-vibration response interaction energy modulation eigenvalue as a reference point, the previously formed motor power factor-vibration response interaction random perturbation ordered vector is semantically scaled to construct a semantically enhanced motor power factor-vibration response interaction random perturbation ordered vector. In this way, those perturbations that appear to be more important in the original feature space are amplified, thereby enhancing the model's sensitivity to changes in key features, while also retaining relatively small but potentially useful perturbation information.

[0052] Next, the product of the randomly perturbed ordered vector of semantically reinforced motor-motor-response interactions and the transpose of the motor-motor-response interaction expression vector is calculated to obtain the motor-motor-response interaction best-example cue orthogonal modulation matrix. This matrix aims to identify how features interact from two different perspectives. The resulting matrix can be considered a conversion tool for capturing and representing complex linear relationships between features. Finally, the final feature structure modulation is achieved by applying the constructed motor-motor-response interaction best-example cue orthogonal modulation matrix to the original motor-motor-response interaction expression vector. This makes the features more consistent with the objective function, reduces redundant information, and maximizes the retention of useful information, thereby improving learning efficiency and prediction performance.

[0053] In summary, the feature manifold optimization process for the motor dynamics-vibration response interaction matrix has been elucidated. This process extracts optimal prompt examples based on fine-grained target retrieval with random perturbations. This optimal prompt example then guides adversarial robust learning to complete the structure of the high-dimensional manifold of feature vectors within the feature space, improving the detail and structure of the feature representation. This achieves a balanced performance between mapping accuracy and mapping completeness in the classification and regression process based on the diverse feature distribution of the motor dynamics-vibration response interaction matrix, improving the accuracy of the motor state detection results obtained by the classifier-based motor state detection module based on the motor dynamics-vibration response interaction matrix. This allows for capturing more information about the equipment's operating status, particularly subtle changes that may indicate early failures. Furthermore, the process automatically extracts features from the data and identifies local temporal coordination patterns between data points, enabling early detection of potential problems and enhancing the intelligent and predictive maintenance capabilities of operations and maintenance management.

[0054] In summary, the CNC machine tool intelligent operation and maintenance management system 100 based on the embodiment of the present application is explained, which uses artificial intelligence-based data analysis and processing technology to perform time series analysis on the time queue of vibration amplitude values, the time queue of current values, and the time queue of voltage values. In this way, the motor status detection results are intelligently obtained based on the characteristics obtained by time series correlation of current and voltage and the time series collaborative response characteristics between the vibration amplitude time series characteristics. More information about the operating status of the equipment can be captured, especially those subtle changes that may indicate early failures. It can also automatically extract features from the data and identify local time series collaborative patterns between the data, so as to discover potential problems early, thereby improving the intelligence of operation and maintenance management and the ability of predictive maintenance.

[0055] As described above, the CNC machine tool intelligent operation and maintenance management system 100 according to the embodiment of the present application can be implemented in various terminal devices, such as a server for intelligent operation and maintenance management of CNC machine tools. In one example, the CNC machine tool intelligent operation and maintenance management system 100 according to the embodiment of the present application can be integrated into the terminal device as a software module and / or hardware module. For example, the CNC machine tool intelligent operation and maintenance management system 100 can be a software module in the operating system of the terminal device, or it can be an application developed for the terminal device; of course, the CNC machine tool intelligent operation and maintenance management system 100 can also be one of the many hardware modules of the terminal device.

[0056] Alternatively, in another example, the CNC machine tool intelligent operation and maintenance management system 100 and the terminal device may also be separate devices, and the CNC machine tool intelligent operation and maintenance management system 100 may be connected to the terminal device via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0057] Figure 7 Flowchart of the intelligent operation and maintenance management method for CNC machine tools according to the embodiment of the present application. Figure 7 As shown, in the intelligent operation and maintenance management method of CNC machine tools, the method includes: S110, a sensor group of a motor deployed on the CNC machine tool collects a time queue of motor state parameters, wherein the motor state parameters include vibration amplitude value, current value and voltage value; S120, data grouping the time queue of the motor state parameters according to the parameter sample dimension to obtain a time queue of vibration amplitude value, a time queue of current value and a time queue of voltage value; S130, feature extraction and association analysis are performed on the time queue of vibration amplitude value, the time queue of current value and the time queue of voltage value to obtain a motor state detection result, and the motor state detection result is used to indicate whether there is an abnormality in the motor state; S140, in response to the motor state detection result indicating that the motor state is abnormal, an alarm notification is generated.

[0058] The specific operations of each step in the above CNC machine tool intelligent operation and maintenance management method have been referenced above. Figures 1 to 6 The description of the intelligent operation and maintenance management system for CNC machine tools has been introduced in detail, and therefore, its repeated description will be omitted.

[0059] In summary, the intelligent operation and maintenance management method for CNC machine tools based on the embodiment of the present application is explained. It uses artificial intelligence-based data analysis and processing technology to perform time series analysis on the time queue of vibration amplitude values, the time queue of current values, and the time queue of voltage values. In this way, the motor status detection results are intelligently obtained based on the characteristics obtained by time series correlation of current and voltage and the time series collaborative response characteristics between the vibration amplitude time series characteristics. More information about the operating status of the equipment can be captured, especially those subtle changes that may indicate early failures. It can also automatically extract features from the data and identify local time series collaborative patterns between the data, so as to discover potential problems early, thereby improving the intelligence of operation and maintenance management and the ability of predictive maintenance.

[0060] The basic principles of the present application have been described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, and effects mentioned in this application are merely illustrative and not restrictive, and it should not be assumed that these advantages, strengths, and effects are required of each embodiment of this application. In addition, the specific details of the above application are merely illustrative and for ease of understanding, and are not restrictive. The above details do not limit this application to being implemented by adopting the above specific details.

Claims

1. Intelligent operation and maintenance management system for CNC machine tools, characterized in that: include: Data collection layer, data processing layer, data analysis layer and application layer; The data acquisition layer includes a data acquisition module and a data transmission module, wherein the data acquisition module is used to receive a time queue of motor state parameters collected by a sensor group of a motor deployed on a CNC machine tool, wherein the motor state parameters include vibration amplitude values, current values, and voltage values; and the data transmission module is used to transmit the time queue of the motor state parameters to the data processing layer; The data processing layer is used to group the time queue of the motor state parameters according to the parameter sample dimension to obtain a time queue of vibration amplitude values, a time queue of current values, and a time queue of voltage values; The data analysis layer is used to perform feature extraction and correlation analysis on the time queue of the vibration amplitude value, the time queue of the current value and the time queue of the voltage value to obtain a motor state detection result, and the motor state detection result is used to indicate whether the motor state is abnormal; The application layer is used to generate an alarm notification in response to the motor state detection result indicating that the motor state is abnormal.

2. The intelligent operation and maintenance management system for CNC machine tools according to claim 1 is characterized in that: The data analysis layer includes: A motor state parameter timing encoding module, used for respectively performing timing sequence encoding on the time queue of the vibration amplitude value, the time queue of the current value and the time queue of the voltage value to obtain a vibration amplitude timing correlation feature vector, a current timing correlation feature vector and a voltage timing correlation feature vector; A motor working power factor timing correlation representation generation module, used for inputting the current timing correlation feature vector and the voltage timing correlation feature vector into the motor working power timing analysis module to obtain a motor working power factor timing correlation representation vector; A motor power factor-vibration coordinated response analysis module is used to perform a motor power factor-vibration timing characteristic coordinated response analysis on the motor working power factor timing correlation representation vector and the vibration amplitude timing correlation feature vector to obtain a motor power factor-vibration coordinated response feature; A motor state detection result generating module, used for obtaining the motor state detection result based on the motor power factor-vibration coordinated response characteristics; The motor dynamic factor-vibration coordinated response analysis module includes: A vibration amplitude time series correlation feature refinement unit, used for performing feature refinement on the vibration amplitude time series correlation feature vector to obtain a sequence of vibration amplitude local time series correlation feature vectors; A motor working power factor-vibration correlation unit, used for calculating the correlation matrix between the motor working power factor time series correlation representation vector and each vibration amplitude local time series correlation feature vector in the sequence of the vibration amplitude local time series correlation feature vector to obtain a sequence of motor working power factor-vibration correlation matrices; The motor power factor-vibration local and global joint response unit is used to perform local and global joint response evaluation on the sequence of the motor working power factor-vibration association matrix to obtain the motor power factor-vibration response interaction expression matrix as the motor power factor-vibration coordinated response feature.

3. The intelligent operation and maintenance management system for CNC machine tools according to claim 2 is characterized in that: The motor state parameter timing encoding module is used to: input the time queue of the vibration amplitude value, the time queue of the current value and the time queue of the voltage value into a timing sequence encoder based on a Bi-LSTM model respectively to obtain the vibration amplitude timing association feature vector, the current timing association feature vector and the voltage timing association feature vector.

4. The intelligent operation and maintenance management system for CNC machine tools according to claim 3 is characterized in that: The motor working power factor-vibration correlation unit comprises: A nonlinear transformation subunit of the motor working power factor timing correlation representation vector, used for performing a nonlinear transformation on the motor working power factor timing correlation representation vector to obtain a nonlinearly transformed motor working power factor timing correlation representation vector; The motor working power factor-vibration correlation matrix sequence calculation subunit is used to calculate the correlation matrix between the motor working power factor time series correlation representation vector after the nonlinear transformation and each vibration amplitude local time series correlation feature vector in the sequence of the vibration amplitude local time series correlation feature vector to obtain the motor working power factor-vibration correlation matrix sequence.

5. The intelligent operation and maintenance management system for CNC machine tools according to claim 4 is characterized in that: The motor power factor-vibration local and global joint response unit comprises: A motor work power factor-vibration correlation characteristic interaction energy factor calculation subunit, used to calculate the characteristic interaction energy factor of each motor work power factor-vibration correlation matrix in the sequence of the motor work power factor-vibration correlation matrix to obtain a sequence of motor work power factor-vibration correlation characteristic interaction energy factors; A motor work power factor-vibration correlation characteristic interaction energy factor normalization subunit, used for inputting the sequence of the motor work power factor-vibration correlation characteristic interaction energy factor into a Sigmoid function to obtain a sequence of normalized motor work power factor-vibration correlation characteristic interaction energy factors; The motor power factor-vibration response interaction expression matrix generating subunit is used to calculate the position-weighted sum of the sequence of the motor work power factor-vibration association matrix using the sequence of the normalized motor work power factor-vibration association characteristic interaction energy factors as the sequence of weights to obtain the motor power factor-vibration response interaction expression matrix.

6. The intelligent operation and maintenance management system for CNC machine tools according to claim 5 is characterized in that: The motor working power factor-vibration correlation characteristic interaction energy factor calculation subunit includes: A motor work power factor-vibration associated kurtosis value calculation secondary subunit is used to calculate the kurtosis value of each motor work power factor-vibration associated matrix in the sequence of the motor work power factor-vibration associated matrix to obtain a sequence of motor work power factor-vibration associated kurtosis values; The motor working power factor-vibration interaction energy factor calculation secondary subunit is used to calculate the characteristic interaction energy factor of each motor working power factor-vibration association matrix based on each motor working power factor-vibration association kurtosis value in the sequence of the motor working power factor-vibration association kurtosis value to obtain the sequence of the motor working power factor-vibration association characteristic interaction energy factors.

7. The intelligent operation and maintenance management system for CNC machine tools according to claim 6 is characterized in that: The motor working power factor-vibration associated kurtosis value calculation secondary subunit is used for: Respectively calculating the mean and variance of the motor working power factor-vibration correlation matrix to obtain the motor working power factor-vibration correlation mean and the motor working power factor-vibration correlation variance; Calculating the position difference between the motor working power factor-vibration correlation matrix and the motor working power factor-vibration correlation mean to obtain the motor working power factor-vibration correlation difference matrix; Calculating the fourth power of each eigenvalue in the motor working power factor-vibration correlation difference matrix to obtain the motor working power factor-vibration correlation modulation difference matrix; Calculating the expected value of the motor working power factor-vibration correlation modulation difference matrix to obtain the motor working power factor-vibration correlation expected value; The motor work power factor-vibration correlation expected value is divided by the square of the motor work power factor-vibration correlation variance to obtain the motor work power factor-vibration correlation kurtosis value.

8. The intelligent operation and maintenance management system for CNC machine tools according to claim 7 is characterized in that: The motor working power factor-vibration interaction energy factor calculation secondary subunit is used for: Calculating the sum of the motor working power factor-vibration correlation variance and the hyperparameter to obtain the motor working power factor-vibration correlation first energy factor; Calculating the square of the difference between the motor work power factor-vibration correlation kurtosis value and the motor work power factor-vibration correlation mean value to obtain the motor work power factor-vibration correlation difference value; The motor work power factor-vibration correlation variance is multiplied by a constant 2, the value obtained by multiplying the hyper parameter by a constant 2, and the motor work power factor-vibration correlation difference value are added to obtain a motor work power factor-vibration correlation second energy factor; The motor work power factor-vibration associated first energy factor is divided by the motor work power factor-vibration associated second energy factor to obtain the motor work power factor-vibration associated characteristic interaction energy factor corresponding to the motor work power factor-vibration association matrix.

9. The intelligent operation and maintenance management system for CNC machine tools according to claim 8, characterized in that: The motor state detection result generating module is used to: input the motor power factor-vibration response interaction expression matrix into the classifier-based motor state detection module to obtain the motor state detection result.

10. Intelligent operation and maintenance management method for CNC machine tools, characterized in that: include: A time queue for collecting motor state parameters by a sensor group of a motor deployed on a CNC machine tool, wherein the motor state parameters include vibration amplitude value, current value and voltage value; The time queue of the motor state parameters is grouped according to the parameter sample dimension to obtain a time queue of vibration amplitude values, a time queue of current values, and a time queue of voltage values; Performing feature extraction and correlation analysis on the time queue of the vibration amplitude value, the time queue of the current value, and the time queue of the voltage value to obtain a motor state detection result, wherein the motor state detection result is used to indicate whether the motor state is abnormal; In response to the motor state detection result indicating that the motor state is abnormal, an alarm notification is generated.

Citation Information

Patent Citations

  • Equipment fault prediction system and method based on digital twin and Internet of Things technology

    CN117991752A

  • Hydraulic engineering management method and system based on digital twinning

    CN118839868A

  • Robot failure symptom detection apparatus and robot failure symptom detection method

    US20240192674A1

  • Production management control system for electronic-grade ammonium fluoride production

    WO2024198266A1

Cited By

  • Gas generator set monitoring system and method

    CN119439856A