A method, device, equipment and medium for monitoring the health of ultra-precision fly-cutting machine tools

By extracting the vibration data of ultra-precision fly-cut machine tools and processing the multivariate Gaussian distribution model, effective identification and health monitoring of fly-cut actions is achieved, the problem of identification and monitoring gap in the existing technology is solved, and the processing quality is improved and efficiency.

CN114815740BActive Publication Date: 2024-11-26硕橙(厦门)科技有限公司 +1
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
CN202210414628.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-04-20
Publication Date
2024-11-26
Estimated Expiration
2042-04-20

AI Technical Summary

Technical Problem

Compared with the actual requirements of the actual production requirements of the existing technology, the effective identification of the fly-cut action of the Chinese Super Precision Flycut Machine Tool has a large gap between the effective identification of the fly-cut action and the monitoring of the health degree from the actual production requirements, resulting in poor process quality and low processing efficiency.

Method used

By obtaining the vibration data of the fly-cut machine tool, calling the trained fly-cut action recognition model for data preprocessing, generating feature data, and using the multivariate Gaussian distribution health model to process the feature data, generating health scores, and outputting early warning signals , to identify fly-cut action points and abnormal working conditions in real time.

Benefits of technology

Effectively identify the fly-cut action points of ultra-precision fly-cut machine tools, improve processing quality and efficiency, solve the problem of the gap between fly-cut action recognition and health monitoring in the existing technology, and ensure the processing quality and efficiency of the machine tools.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114815740B_ABST
    Figure CN114815740B_ABST
Patent Text Reader

Abstract

The present invention provides a method, device, equipment and medium for monitoring the health of ultra-precision fly-cutting machine tools, including: obtaining vibration data of the fly-cutting machine tool to be monitored; calling a trained fly-cutting action recognition model to pre-process the vibration data to generate corresponding feature data, and constructing a fly-cutting action recognition model by marking the action of multiple groups of historical vibration data of the fly-cutting machine tools, and extracting the feature quantities of the feed sequence and the output sequence from the data; calling a trained health model to process the feature data to generate a health score, and obtaining multiple groups of historical feature data under various process parameters of the fly-cutting machine tools on the basis of the recognition of the fly-cutting action points, and training a health model based on multivariate Gaussian distribution according to the extracted three-dimensional features of the feed vibration value, the output vibration value and the data zero-crossing rate; and outputting a warning signal according to the health score. It is intended to solve the problems of poor process quality and low processing efficiency of ultra-precision fly-cutting machine tools on the market.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of precision machining technology, and in particular to a method, device, equipment and medium for monitoring the health of an ultra-precision fly-cutting machine tool. Background Art

[0002] Ultra-precision machining technology is an important supporting technology for modern high-tech warfare, the foundation for the development of modern high-tech industries and science and technology, and the development direction of modern manufacturing science.

[0003] As far as ultra-precision machine tools are concerned, ultra-precision machine tools, as important equipment for realizing ultra-precision machining, are the most intuitive embodiment of a country's ultra-precision machining technology level. Among them, ultra-precision fly-cutting machine tool processing technology is a mechanical processing technology for processing high-precision optical devices. During the machine tool processing process, the fly-cutting action has a significant impact on the quality of the mechanical processing technology. The effective identification and health monitoring of the fly-cutting action of ultra-precision fly-cutting machines in the existing technology are still quite different from the actual production requirements, which makes the process quality of ultra-precision fly-cutting machines on the market poor and the processing efficiency low.

[0004] In view of this, this application is filed. Summary of the invention

[0005] In view of this, the purpose of the present invention is to provide a method, device, equipment and medium for monitoring the health of ultra-precision fly-cutting machines, which can effectively solve the problems of effective identification and health monitoring of the fly-cutting action of ultra-precision fly-cutting machines in the prior art. Compared with the actual production requirements, there is still a considerable gap, which makes the ultra-precision fly-cutting machines on the market have poor process quality and low processing efficiency.

[0006] The present invention discloses a method for monitoring the health of an ultra-precision fly-cutting machine tool, comprising:

[0007] Obtain vibration data of the fly-cutting machine tool to be monitored;

[0008] The trained fly-cutting action recognition model is called to perform data preprocessing on the vibration data to generate corresponding feature data, wherein the fly-cutting action recognition model is constructed by performing action marking on multiple groups of historical vibration data of fly-cutting machines and extracting feature quantities of feed sequence and exit sequence therefrom;

[0009] The trained health model is called to process the characteristic data to generate a health score, wherein, based on the identification of the fly-cutting action point, multiple groups of historical characteristic data under various process parameters of the fly-cutting machine tool are obtained, and the health model based on the multivariate Gaussian distribution is trained according to the three-dimensional features of the feed vibration value, the exit vibration value, and the data zero-crossing rate;

[0010] According to the health score, a warning signal is output.

[0011] Preferably, before calling the trained fly-cut action recognition model to perform data preprocessing on the vibration data, the method further includes:

[0012] Select and filter a plurality of historical vibration data, generate a plurality of fly-cut samples with clear and complete data sequences, and mark the starting points of the feed and exit sequences of the fly-cut samples;

[0013] Perform zero value calculation on the collected multiple fly-cut samples to generate fly-cut action recognition model parameters;

[0014] A fly-cut action recognition model library is constructed according to the fly-cut action recognition model parameters to generate a fly-cut action recognition model.

[0015] Preferably, before calling the trained health model to process the characteristic data, the method further includes:

[0016] Calling a multivariate Gaussian model to calculate the acquired multiple sets of historical feature data of the fly-cutting machine tool, constructing a fly-cutting feature scoring model library, and generating a fly-cutting feature scoring model, wherein the acquired multiple sets of historical feature data are used as input parameters of the multivariate Gaussian model;

[0017] A multivariate Gaussian model is used to calculate the historical feature data of multiple groups of fly-cutting machine tools, build a fly-cutting action sequence benchmark library, and generate a fly-cutting action sequence scoring model;

[0018] A healthiness model is generated according to the fly-cut feature scoring model and the fly-cut action sequence scoring model.

[0019] Preferably, the trained fly-cut action recognition model is called to perform data preprocessing on the vibration data to generate corresponding feature data, specifically:

[0020] Screening and processing the vibration data to generate corresponding fly-cut samples, and marking the start and end points of the sequence;

[0021] Processing the fly-cut samples to zero values ​​to generate corresponding replacement data sequences with zero values;

[0022] Performing a zero value comparison process on the replacement data sequence to generate a tool-in sequence and a tool-out sequence;

[0023] The maximum values ​​of the tool-in sequence and the tool-out sequence are identified, and corresponding feature data are generated.

[0024] Preferably, the trained health model is called to process the characteristic data to generate a health score, specifically:

[0025] Calling the fly-cut feature scoring model to perform probability calculation on the feature data to generate a corresponding feature score;

[0026] Calling the flying cut action sequence scoring model to perform comparison processing on the feature data to generate a corresponding action sequence score;

[0027] The feature score and the action sequence score are compared to generate a health score.

[0028] The present invention also provides a health monitoring device for an ultra-precision fly-cutting machine tool, comprising:

[0029] A data acquisition unit, used for acquiring vibration data of the fly-cutting machine tool to be monitored;

[0030] An action recognition unit is used to call a trained fly-cut action recognition model to perform data preprocessing on the vibration data to generate corresponding feature data, wherein the fly-cut action recognition model is constructed by performing action marking on multiple groups of historical vibration data and based on feature quantities of a tool-in sequence and a tool-out sequence extracted therefrom;

[0031] The motion analysis unit is used to call the trained health model to process the feature data and generate a health score, wherein, based on the identification of the fly-cutting action point, multiple groups of historical feature data under various process parameters of the fly-cutting machine tool are obtained, and the health model based on the multivariate Gaussian distribution is trained according to the three-dimensional features of the feed vibration value, the exit vibration value, and the data zero-crossing rate;

[0032] The early warning unit is used to output an early warning signal according to the health score.

[0033] The present invention also provides an ultra-precision fly-cutting machine tool health monitoring device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, it implements an ultra-precision fly-cutting machine tool health monitoring method as described in any one of the above.

[0034] The present invention also provides a readable storage medium storing a computer program, wherein the computer program can be executed by a processor of a device where the storage medium is located to implement a method for monitoring the health of an ultra-precision fly-cutting machine tool as described in any one of the above items.

[0035] In summary, the present embodiment provides a method, device, equipment and medium for monitoring the health of an ultra-precision fly-cutting machine tool. The fly-cutting action recognition model can identify the fly-cutting action points in the vibration data in real time, extract the feature data and input it into the health model based on the multivariate Gaussian distribution to score the fly-cutting action features. At the same time, the identified action sequence is scored with a preset benchmark action table, and the two scoring items are compared to select the final health score. Through the fly-cutting action recognition model and the health model, the abnormal working conditions of the ultra-precision diamond fly-cutting machine tool processing process can be identified in real time, and the machine tool processing quality and processing efficiency can be effectively guaranteed, thereby solving the effective identification and health monitoring of the fly-cutting action of the ultra-precision fly-cutting machine tool in the prior art. Compared with the actual production requirements, there is still a considerable gap, which makes the ultra-precision fly-cutting machine tools on the market have poor process quality and low processing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a flow chart of a method for monitoring the health of an ultra-precision fly-cutting machine tool provided in an embodiment of the present invention.

[0037] Figure 2 It is a structural schematic diagram of an ultra-precision diamond fly-cutting machine tool provided by an embodiment of the present invention.

[0038] Figure 3 It is a waveform diagram of the ultra-precision fly-cutting machine tool health monitoring method, device, equipment and medium provided by an embodiment of the present invention.

[0039] Figure 4 It is a module schematic diagram of the ultra-precision fly-cutting machine tool health monitoring device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the invention claimed for protection, but merely represents the selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0041] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0042] See also Figures 1 to 3The first embodiment of the present invention provides a method for monitoring the health of an ultra-precision fly-cutting machine tool, which can be performed by an ultra-precision fly-cutting machine tool health monitoring device (hereinafter referred to as the monitoring device), and in particular, by one or more processors in the monitoring device to implement the following steps:

[0043] S101, obtaining vibration data of the fly-cutting machine tool to be monitored;

[0044] Specifically, in this embodiment, the vibration data may be a first vibration sensor signal deployed on the workbench beam, a second vibration sensor signal of the workbench, and a third vibration sensor signal of the ground. It should be noted that in other embodiments, other types of vibration data may also be used, which are not specifically limited here, but all of these solutions are within the protection scope of the present invention.

[0045] S102, calling the trained fly-cutting action recognition model to perform data preprocessing on the vibration data to generate corresponding feature data, wherein the fly-cutting action recognition model is constructed by performing action marking on multiple groups of historical vibration data of fly-cutting machines and according to feature quantities of feed sequence and output sequence extracted therefrom;

[0046] Specifically, in this embodiment, before calling the trained fly-cut action recognition model to perform data preprocessing on the vibration data, the method further includes:

[0047] Select and filter a plurality of historical vibration data, generate a plurality of fly-cut samples with clear and complete data sequences, and mark the starting points of the feed and exit sequences of the fly-cut samples;

[0048] Perform zero value calculation on the collected multiple fly-cut samples to generate fly-cut action recognition model parameters;

[0049] A fly-cut action recognition model library is constructed according to the fly-cut action recognition model parameters to generate a fly-cut action recognition model.

[0050] Specifically, in this embodiment, for the fly-cut action recognition model:

[0051] Screening and processing the vibration data to generate corresponding fly-cut samples, and marking the start and end points of the sequence;

[0052] Processing the fly-cut samples to zero values ​​to generate corresponding replacement data sequences with zero values;

[0053] Performing a zero value comparison process on the replacement data sequence to generate a tool-in sequence and a tool-out sequence;

[0054] The maximum values ​​of the tool-in sequence and the tool-out sequence are identified, and corresponding feature data are generated.

[0055] In this embodiment, the vibration sensor signal of the fly-cutting machine tool is used as the data source, the fly-cutting action is marked on the historical data, and the normal samples with clear and complete fly-cutting vibration data sequence are screened out, and the starting and ending points of the feed and exit vibration data sequence are marked; after pre-processing the data, the number of adjacent zero values ​​of the feed sequence and the exit sequence is extracted, and after statistical analysis, it is used as a parameter to construct a fly-cutting action recognition model. Among them, the data pre-processing includes, for the normal fly-cutting marked samples of a specified time length, extracting a certain proportion of data to establish a dynamic background noise data distribution model, and rescanning the fly-cutting samples within the time length, and setting the data that meets the background noise data distribution model to zero.

[0056] For example, take the spindle speed, feed depth, and processing material parameter table 1 of the ultra-precision diamond fly cutting machine as an example:

[0057] Table 1

[0058] Spindle speed Feed speed Feed rate Processing materials 360 60 10 KDP

[0059] Among them, a 10.6v / g iepe vibration sensor and a ZonicBook / 618E digital acquisition board can be used, and the sampling rate is configured to be 9e-5s; from the digital acquisition board, the fly-cutting machine tool processing vibration data of several processing cycles are extracted for screening and marking. It should be noted that in other embodiments, other types of vibration sensors and digital acquisition boards can also be used, which are not specifically limited here, but these solutions are all within the protection scope of the present invention.

[0060] S103, calling the trained health model to process the feature data and generate a health score, wherein, based on the identification of the fly-cutting action point, multiple groups of historical feature data under various process parameters of the fly-cutting machine tool are obtained, and the health model based on the multivariate Gaussian distribution is trained according to the three-dimensional features of the feed vibration value, the exit vibration value, and the data zero-crossing rate;

[0061] Specifically, in this embodiment, before calling the trained health model to process the feature data, the method further includes:

[0062] Calling a multivariate Gaussian model to calculate the acquired multiple sets of historical feature data of the fly-cutting machine tool, constructing a fly-cutting feature scoring model library, and generating a fly-cutting feature scoring model, wherein the acquired multiple sets of historical feature data are used as input parameters of the multivariate Gaussian model;

[0063] A multivariate Gaussian model is used to calculate the historical feature data of multiple groups of fly-cutting machine tools, build a fly-cutting action sequence benchmark library, and generate a fly-cutting action sequence scoring model;

[0064] A healthiness model is generated according to the fly-cut feature scoring model and the fly-cut action sequence scoring model.

[0065] Specifically, in this embodiment, for the health model:

[0066] Calling the fly-cut feature scoring model to perform probability calculation on the feature data to generate a corresponding feature score;

[0067] Calling the flying cut action sequence scoring model to perform comparison processing on the feature data to generate a corresponding action sequence score;

[0068] The feature score and the action sequence score are compared to generate a health score.

[0069] In this embodiment, the health model includes a fly-cut feature scoring model and a fly-cut action sequence scoring model. The fly-cut feature scoring model inputs the acquired fly-cut sample features into a multivariate Gaussian model, and the obtained probability value is multiplied by a fixed value to obtain a feature score; in the action sequence scoring of the fly-cut action sequence scoring model, the action sequence obtained by the fly-cut action recognition must meet the order of feed, exit, feed... At the same time, at a certain speed, in each sampling window, the collected feed and exit actions must meet a certain number, and the real-time action sequence is compared with the pre-configured benchmark action table under the process parameters. On the basis of a fixed value, a certain score is deducted for each mismatch, so as to obtain the final action sequence score. Among them, the final health score takes the smaller value of the two scores of the fly-cut feature scoring model and the fly-cut action sequence scoring model.

[0070] For example, taking Table 1 as an example, according to the formula Dynamic data preprocessing based on the univariate Gaussian model is performed. Every 1 second of fly-cut vibration data is obtained, and the 25th to 75th percentile data are taken to establish the above model. The rescanned data is substituted into the model for calculation. If p(x)<0.003, the data is set to zero. The fly-cut action model parameters obtained based on statistical calculations are as shown in Table 2 below:

[0071] Table 2

[0072]

[0073] The fly cutting feature scoring model based on the multivariate Gaussian model is: Substitute the feature data into the model for training to obtain the model parameters.

[0074] The reference action table is automatically generated by the configured parameters, where 2 and 1 are respectively the infeed and outfeed, as described in Table 3 below:

[0075] Table 3

[0076] 1,2,1,2,1,2,1,2,1,2 2,1,2,1,2,1,2,1,2,1

[0077] S104: Outputting a warning signal according to the health score.

[0078] Specifically, in this embodiment, during real-time monitoring, the fly-cutting action recognition model and the fly-cutting machine tool health model are automatically matched from the fly-cutting action recognition model library and the fly-cutting machine tool health model library according to the entered process parameters. If the health score is lower than a certain threshold, an early warning prompt signal is output.

[0079] See also Figure 4 The second embodiment of the present invention provides a health monitoring device for an ultra-precision fly-cutting machine tool, comprising:

[0080] A data acquisition unit 201 is used to acquire vibration data of the fly-cutting machine tool to be monitored;

[0081] The action recognition unit 202 is used to call the trained fly-cut action recognition model to perform data preprocessing on the vibration data to generate corresponding feature data, wherein the fly-cut action recognition model is constructed by performing action marking on multiple groups of historical vibration data and extracting feature quantities of the feed sequence and the output sequence therefrom;

[0082] The motion analysis unit 203 is used to call the trained health model to process the feature data and generate a health score, wherein, based on the identification of the fly-cutting action point, multiple groups of historical feature data under various process parameters of the fly-cutting machine tool are obtained, and the health model based on the multivariate Gaussian distribution is trained according to the three-dimensional features of the feed vibration value, the exit vibration value, and the data zero-crossing rate;

[0083] The early warning unit 204 is used to output an early warning signal according to the health score.

[0084] The third embodiment of the present invention provides an ultra-precision fly-cutting machine tool health monitoring device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, it implements an ultra-precision fly-cutting machine tool health monitoring method as described in any one of the above.

[0085] The fourth embodiment of the present invention provides a readable storage medium storing a computer program, wherein the computer program can be executed by a processor of a device where the storage medium is located to implement a method for monitoring the health of an ultra-precision fly-cutting machine tool as described in any one of the above.

[0086] Exemplarily, the computer program described in the third and fourth embodiments of the present invention can be divided into one or more modules, and the one or more modules are stored in the memory and executed by the processor to complete the present invention. The one or more modules can be a series of computer program instruction segments that can complete specific functions, and the instruction segments are used to describe the execution process of the computer program in the implementation of an ultra-precision fly-cutting machine tool health monitoring device. For example, the device described in the second embodiment of the present invention.

[0087] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the method for monitoring the health of an ultra-precision fly-cutting machine tool, and uses various interfaces and lines to connect the various parts of the method for monitoring the health of an ultra-precision fly-cutting machine tool.

[0088] The memory can be used to store the computer program and / or module, and the processor realizes various functions of a method for monitoring the health of an ultra-precision fly-cutting machine tool by running or executing the computer program and / or module stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, a text conversion function, etc.), etc.; the data storage area can store data created according to the use of a mobile phone (such as audio data, text message data, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0089] Wherein, if the implemented module is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0090] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0091] The above are only preferred implementations of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention.

Claims

1. A method for monitoring the health of an ultra-precision fly-cutting machine tool, characterized in that: include: Obtain vibration data of the fly-cutting machine tool to be monitored; The trained fly-cutting action recognition model is called to perform data preprocessing on the vibration data to generate corresponding feature data, wherein the fly-cutting action recognition model is constructed by performing action marking on multiple groups of historical vibration data of fly-cutting machines and extracting feature quantities of feed sequence and exit sequence therefrom; The trained health model is called to process the characteristic data to generate a health score, wherein, based on the identification of the fly-cutting action point, multiple groups of historical characteristic data under various process parameters of the fly-cutting machine tool are obtained, and the health model based on the multivariate Gaussian distribution is trained according to the three-dimensional features of the feed vibration value, the exit vibration value, and the data zero-crossing rate; Outputting a warning signal according to the health score; Before calling the trained health model to process the characteristic data, the method further includes: Calling a multivariate Gaussian model to calculate the acquired multiple sets of historical feature data of the fly-cutting machine tool, constructing a fly-cutting feature scoring model library, and generating a fly-cutting feature scoring model, wherein the acquired multiple sets of historical feature data are used as input parameters of the multivariate Gaussian model; A multivariate Gaussian model is used to calculate the historical feature data of multiple groups of fly-cutting machine tools, build a fly-cutting action sequence benchmark library, and generate a fly-cutting action sequence scoring model; Generate a healthiness model according to the fly-cut feature scoring model and the fly-cut action sequence scoring model; The trained health model is called to process the feature data to generate a health score, specifically: Calling the fly-cut feature scoring model to perform probability calculation on the feature data to generate a corresponding feature score; Calling the flying cut action sequence scoring model to perform comparison processing on the feature data to generate a corresponding action sequence score; The feature score and the action sequence score are compared to generate a health score.

2. The method for monitoring the health of an ultra-precision fly-cutting machine tool according to claim 1, characterized in that: Before calling the trained fly-cut action recognition model to perform data preprocessing on the vibration data, the method further includes: Select and filter a plurality of historical vibration data, generate a plurality of fly-cut samples with clear and complete data sequences, and mark the starting points of the feed and exit sequences of the fly-cut samples; Perform zero value calculation on the collected multiple fly-cut samples to generate fly-cut action recognition model parameters; A fly-cut action recognition model library is constructed according to the fly-cut action recognition model parameters to generate a fly-cut action recognition model.

3. The method for monitoring the health of an ultra-precision fly-cutting machine tool according to claim 1, characterized in that: The trained fly-cut action recognition model is called to perform data preprocessing on the vibration data to generate corresponding feature data, specifically: Screening and processing the vibration data to generate corresponding fly-cut samples, and marking the start and end points of the sequence; Processing the fly-cut samples to zero values ​​to generate corresponding replacement data sequences with zero values; Performing a zero value comparison process on the replacement data sequence to generate a tool-in sequence and a tool-out sequence; The maximum values ​​of the tool-in sequence and the tool-out sequence are identified, and corresponding feature data are generated.

4. An ultra-precision fly-cutting machine tool health monitoring device, characterized in that: include: A data acquisition unit, used for acquiring vibration data of the fly-cutting machine tool to be monitored; An action recognition unit is used to call a trained fly-cut action recognition model to perform data preprocessing on the vibration data to generate corresponding feature data, wherein the fly-cut action recognition model is constructed by performing action marking on multiple groups of historical vibration data and based on feature quantities of a tool-in sequence and a tool-out sequence extracted therefrom; The motion analysis unit is used to call the trained health model to process the feature data and generate a health score, wherein, based on the identification of the fly-cutting action point, multiple groups of historical feature data under various process parameters of the fly-cutting machine tool are obtained, and the health model based on the multivariate Gaussian distribution is trained according to the three-dimensional features of the feed vibration value, the exit vibration value, and the data zero-crossing rate; An early warning unit, configured to output an early warning signal according to the health score; Before calling the trained health model to process the characteristic data, the method further includes: Calling a multivariate Gaussian model to calculate the acquired multiple sets of historical feature data of the fly-cutting machine tool, constructing a fly-cutting feature scoring model library, and generating a fly-cutting feature scoring model, wherein the acquired multiple sets of historical feature data are used as input parameters of the multivariate Gaussian model; A multivariate Gaussian model is used to calculate the historical feature data of multiple groups of fly-cutting machine tools, build a fly-cutting action sequence benchmark library, and generate a fly-cutting action sequence scoring model; Generate a healthiness model according to the fly-cut feature scoring model and the fly-cut action sequence scoring model; The trained health model is called to process the feature data to generate a health score, specifically: Calling the fly-cut feature scoring model to perform probability calculation on the feature data to generate a corresponding feature score; Calling the flying cut action sequence scoring model to perform comparison processing on the feature data to generate a corresponding action sequence score; The feature score and the action sequence score are compared to generate a health score.

5. An ultra-precision fly-cutting machine tool health monitoring device, characterized in that: The method comprises a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, a method for monitoring the health of an ultra-precision fly-cutting machine tool as described in any one of claims 1 to 3 is implemented.

6. A readable storage medium, characterized in that: A computer program is stored, and the computer program can be executed by a processor of the device where the storage medium is located to implement a health monitoring method for an ultra-precision fly-cutting machine tool as described in any one of claims 1 to 3.

Citation Information

Patent Citations

  • Monitoring device, monitoring system, monitoring program and recording medium

    CN107924173A

  • Relative vibration identification method based on ultra-precise fly-cutting machining surface shape

    CN111144193A

  • Method, device and system for monitoring state of ultra-precise diamond fly-cutting machine tool and medium

    CN113359610A

  • NUMERICAL CONTROL DEVICE

    DE112019007355T5

  • Abnormality detection device, abnormality detection method, and abnormality detection program

    JP2020042519A