Real-time monitoring method and related device for tool wear and damage based on spindle vibration signal

Abstract

A real-time monitoring method for tool wear and breakage based on spindle vibration signals and related devices include: synchronously collecting spindle vibration and internal CNC system data; performing power spectrum analysis on the basis of vibration data noise reduction to obtain a power spectrum frequency band energy index reflecting tool wear and a dimensionless monitoring index reflecting tool breakage; fusing the indexes to obtain a comprehensive monitoring index reflecting tool status; proposing an effective method for setting tool wear and breakage thresholds; once the comprehensive monitoring index exceeds the failure threshold, the material removal rate increment is used to assist in determining whether the tool has failed, thereby eliminating false alarms generated at the moment of working condition parameter switching and accurately identifying the tool status. The present invention can achieve real-time monitoring of tool wear and tool breakage using only a single external three-axis acceleration sensor and internal CNC system data, and proposes an industrially acceptable tool status monitoring system for single machine tools and flexible production lines, providing a practical solution for industrial applications.

Description

Technical Field

[0001] The invention belongs to the technical field of numerical control machining, and relates to a real-time monitoring method for tool wear and damage based on a spindle vibration signal and a related device. Background Art

[0002] As the "industrial mother machine" for high-end equipment manufacturing, CNC machine tools are the basis for high-precision manufacturing of complex structural parts for large-scale equipment in the fields of aviation, aerospace, navigation, etc. As the final executor of structural part manufacturing by CNC machine tools, cutting tools have a vital impact on the machining accuracy of parts. Cutting tools will inevitably wear out during the cutting process. Excessive wear of the cutting tools will cause the outer diameter of the cutting edge to become smaller, resulting in unqualified part machining dimensions. At the same time, cutting edge wear will increase the VB value of the tool's back face width, resulting in excessive roughness of the machined surface. Once the tool cutting edge is accidentally damaged, it will not only cause a significant increase in machine tool vibration, but also deteriorate the surface quality of the part.

[0003] High-precision CNC machine tools and correct CNC programs are not necessarily capable of producing qualified parts. Tool geometry is also key to ensuring part quality. Real-time understanding of the tool's geometric state is crucial to ensuring part processing quality. Currently, the judgment of tool wear and damage mainly relies on the experience of machine operators. Tool failure is often not detected until some time after failure, greatly increasing the risk of unqualified parts. Considering the potential risks of tools, conservatively replacing tools prematurely will increase tool usage costs and result in a huge waste of resources. With the maturity and development of automated production lines, the market demand for unmanned production lines has become more urgent. Automatic diagnosis and judgment of tool health status is the most critical step in realizing unmanned production line technology.

[0004] Sensors collecting state response signals and NC command data during machine tool processing can be used to identify tool wear and damage. However, extracting useful information reflecting cutting edge state changes from massive amounts of manufacturing data becomes the key to real-time tool state monitoring. The tool state monitoring process can be broadly divided into three stages: data acquisition, feature extraction, and state identification. Patent 202111085598.5 describes a tool wear state prediction method and device based on adversarial transfer learning. The adversarial transfer learning model is trained using source and target domain datasets. While deep learning methods can achieve end-to-end tool state identification, the limited generalization capabilities of deep learning models currently make them difficult to use in real-world industrial settings. Patent 202210778313.4 describes a tool wear monitoring method based on multi-class signal feature fusion. This method identifies the cutting force coefficient to enable tool wear monitoring under complex cutting conditions. Model-data fusion methods can derive indicators such as the milling force coefficient that are sensitive to tool wear and unaffected by operating parameters, but these methods rely on instantaneous milling force models. It is very challenging to obtain the input parameters required to identify the milling force coefficient from processing and manufacturing data. Patent 201610668278.5 invented a method for online monitoring and early warning of the milling cutter status during complex surface processing. The displacement signal is obtained by the secondary frequency domain integration of the acceleration signal in the X and Y directions of the machine tool spindle end measured by a three-axis acceleration sensor to realize tool status monitoring. Patent No. 202010735535.9 invented a tool breakage identification method based on vibration monitoring. By extracting the root mean square value of the vibration signal, online identification of tool breakage in the aircraft structural parts processing process is realized. However, the above method is easily affected by the interference of working condition parameters, resulting in a large number of false alarms, causing frequent shutdown of the machine tool, and affecting the normal production rhythm. Summary of the Invention

[0005] The present invention aims to provide a real-time monitoring method and device for tool wear and breakage based on spindle vibration signals. This method, utilizing a single acceleration sensor and internal data from a numerical control system, enables tool breakage and damage monitoring under time-varying cutting conditions. The proposed tool wear and breakage monitoring method and device are applicable not only to single machine tools but also to flexible production lines.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] The real-time monitoring method of tool wear and damage based on spindle vibration signals includes:

[0008] Collect CNC machine tool spindle vibration data and CNC system internal data;

[0009] Perform noise reduction on the collected data;

[0010] By performing power spectrum analysis on the spindle vibration data after noise reduction, the power spectrum frequency band energy index reflecting tool wear and the dimensionless monitoring index reflecting tool breakage are obtained;

[0011] By integrating the indicators, we can obtain comprehensive monitoring indicators reflecting tool wear and tool breakage;

[0012] When the comprehensive monitoring indicators exceed the failure threshold, the material removal rate increment is used to assist in determining whether the tool has failed, thereby eliminating false alarms generated at the moment of working condition parameter switching and accurately identifying the tool status.

[0013] Furthermore, the tool name and program name inside the CNC system, as well as real-time spindle speed, feed speed, and XYZ axis coordinate data are collected through the OPC UA protocol;

[0014] The spindle's XYZ vibration data is collected in real time through a three-axis acceleration sensor installed on the spindle's side wall;

[0015] Furthermore, noise reduction is performed on the original data by replacing outliers, removing the mean and trend terms. The outliers are replaced with the mean of the data, so that the instantaneous amplitude fluctuations of the processed data are stabilized.

[0016] Furthermore, the power spectrum band energy index is:

[0017] The power spectrum of the vibration signal in the three directions of the spindle XYZ after preprocessing is analyzed and the power spectrum is calculated by direct method as follows (1):

[0018]

[0019] The power spectrum characteristic frequency amplitude is selected as the monitoring indicator, and the mathematical definition of the power spectrum characteristic frequency amplitude indicator is given as follows (2):

[0020] F=Amp psd (f) (2)

[0021] Where Amp(f) is the amplitude corresponding to the characteristic frequency f;

[0022] Define the tooth passing frequency as f p , its size is determined by the formula f p =(n / 60)·N t Calculated, n is the spindle speed, N t is the number of tool teeth; select the tool tooth passing frequency f in the power spectrum component respectively p and triple frequency 3f p Amplitude is used as a monitoring indicator;

[0023] By analyzing the relative distribution of the power spectrum frequency components of the spindle vibration signal at different wear stages, a monitoring indicator reflecting the tool wear change, the power spectrum band energy index, is proposed; 0-7f p The cumulative sum of the amplitudes of the frequency components within the frequency band gives the power spectrum band energy index, which is mathematically defined as follows:

[0024]

[0025] Further, dimensionless monitoring indicators:

[0026] According to the amplitude and distribution of the power spectrum frequency components before and after tool breakage, two dimensionless monitoring indicators that are sensitive to tool breakage but not affected by working parameters are defined.

[0027] According to the relative change in the amplitude of the frequency components of the power spectrum of the vibration signal before and after tool breakage, the mathematical definition of the dimensionless index - amplitude ratio is given as follows (4):

[0028]

[0029] Among them, Amp psd (f p ) is the fundamental frequency amplitude of the blade tooth passing frequency, Amp psd (2f p ) is the amplitude of twice the frequency of the blade tooth passing through;

[0030] According to the relative changes in the frequency components and distribution of the power spectrum of the vibration signal before and after tool breakage, the mathematical definition of the dimensionless index - energy ratio is given as follows (5):

[0031]

[0032] in, is the low-frequency band energy of the vibration signal power spectrum, is the high frequency band energy of the knife vibration signal power spectrum.

[0033] Furthermore, the indicators are integrated to obtain comprehensive monitoring indicators reflecting tool wear and tool breakage:

[0034] The mathematical definition of normalization of different indicators is as follows (6):

[0035]

[0036] Among them, K normal is the amplitude of each monitoring index when the tool is normal, K i is the monitoring index before normalization of tool wear at different stages, K t It is the monitoring index of different degradation stages of the tool after normalization;

[0037] After the indicators are normalized, the mathematical definition of the fusion between different indicators is given as follows (7):

[0038]

[0039] Among them, K synthetic is a comprehensive monitoring index for tool breakage or a comprehensive monitoring index for tool breakage; n is the weight of each monitoring indicator; K tn It is the monitoring index used for feature fusion after normalization, including indicators between different directions and different types; through the normalization processing and dimensionless index fusion through formula (6)-(7), the comprehensive monitoring index reflecting tool wear and tool breakage is obtained.

[0040] Furthermore, when the comprehensive monitoring indicators exceed the failure threshold, the material removal rate increment is used to assist in determining whether the tool has failed, thereby eliminating false alarms caused by the switching of working parameters and accurately identifying the tool status:

[0041] The tool wear threshold is set based on the part processing quality as a constraint. The processing quality refers to the allowable dimensional deviation of the key dimensions of the part and the allowable surface roughness deviation of the part processing surface. Through trial cutting experiments, a correlation database between the part processing quality and the comprehensive tool wear index at different tool wear stages is established. The tool wear threshold is dynamically set according to the different quality requirements of the processed parts to maximize the use of tool life. The mathematical definition of the tool wear failure threshold is given as shown in the following formula (8):

[0042] T wear =min{K tolerance ,K roughness} (8)

[0043] Among them, T wear K is the tool wear monitoring threshold; tolerance K is the maximum amplitude of the comprehensive monitoring index within the tolerance range; roughness The maximum amplitude of the comprehensive monitoring index within the allowable range of surface roughness;

[0044] The tool breakage threshold is set based on the Gaussian statistical distribution method for identifying abnormal data. The mathematical definition of the tool breakage monitoring floating threshold is shown in the following formula (9):

[0045]

[0046] Among them, T breakage is the tool breakage monitoring threshold; μ is the mean of the historical tool breakage comprehensive monitoring index; σ is the standard deviation of the historical tool breakage comprehensive monitoring index; N is the number of historical tool breakage comprehensive monitoring indexes; x(j) is the historical tool breakage comprehensive monitoring index;

[0047] The material removal rate MRR, which measures the change of working condition parameters, is defined as follows (10):

[0048]

[0049] Among them, v f is the machine feed speed; a p is the axial cutting depth of the tool; a e is the radial cutting width of the tool;

[0050] The feed per tooth of the tool can also be used to identify and monitor the changes in working parameters under fixed cutting width and cutting depth. The feed per tooth that measures the changes in working parameters is defined as follows (11):

[0051]

[0052] When the comprehensive monitoring index exceeds the threshold, it is first determined whether the index amplitude fluctuation is caused by the change of the working condition parameters. If the increase in the monitoring index amplitude is caused by the change of the working condition parameters, no alarm information will be issued, and the tool status monitoring system will continue to monitor the tool status. If the comprehensive monitoring index reaches the failure threshold while the working condition parameters remain constant, it is determined that the tool has failed.

[0053]

[0054] Among them, Normal means the tool is in a normal state; Fault means the tool is in a failure state, including excessive tool wear and abnormal tool damage; Q is the material removal rate change threshold; K synthetic It is a comprehensive monitoring indicator for tool wear or tool breakage; T threshold It is the failure threshold of tool wear or tool breakage.

[0055] Furthermore, a real-time monitoring system for tool wear and damage based on spindle vibration signals includes:

[0056] Data acquisition module, used to collect spindle vibration and internal data of CNC system;

[0057] Preprocessing module, used to perform noise reduction on the collected data;

[0058] The indicator design module is used to perform power spectrum analysis on the spindle vibration data after noise reduction to obtain power spectrum band energy indicators and dimensionless monitoring indicators that reflect tool wear changes;

[0059] The fusion module is used to fuse the indicators to obtain comprehensive monitoring indicators reflecting tool wear and tool breakage;

[0060] The comprehensive judgment module uses the material removal rate increment to assist in determining whether the tool has failed when the comprehensive monitoring indicators exceed the failure threshold, thereby eliminating false alarms generated at the moment of working condition parameter switching and accurately identifying the tool status.

[0061] Furthermore, a computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the real-time monitoring method for tool wear and damage based on the spindle vibration signal are implemented.

[0062] Furthermore, a computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a method for real-time monitoring of tool wear and damage based on a spindle vibration signal.

[0063] Compared with the prior art, the present invention has the following technical effects:

[0064] This invention utilizes a low-cost, easily installed three-axis acceleration sensor to collect spindle vibration signals and develops a method for monitoring tool wear and breakage under time-varying cutting conditions. The developed tool condition monitoring system can not only monitor the tool condition of a single CNC machine tool, but can also be quickly integrated into flexible production lines to enable real-time monitoring and remote access of multiple CNC machines, providing a practical solution for monitoring tool wear and breakage on-site.

[0065] This invention enables real-time monitoring of tool wear and breakage using only data from a single external three-axis acceleration sensor and internal CNC system data. This system provides an industrially acceptable tool condition monitoring system for single machine tools and flexible production lines. The hardware sensors used in this monitoring system are low-cost and easy to install.

[0066] Based on the frequency components of the spindle vibration signal's power spectrum, this paper proposes comprehensive indicators for monitoring tool wear and tool breakage. These indicators include characteristic frequency amplitude, characteristic frequency band energy, and characteristic frequency band amplitude and energy ratios. These indicators are sensitive to tool wear and breakage but less sensitive to operating parameters, significantly reducing false alarms.

[0067] This paper proposes a tool wear failure threshold method based on part machining quality constraints, which can maximize tool life and reduce tool costs. It also proposes a tool breakage floating threshold method based on Gaussian statistical distribution, which greatly simplifies the design of the tool breakage failure threshold.

[0068] This invention proposes a method for diagnosing tool breakage and damage by combining comprehensive indicators with operating parameters for simultaneous monitoring and fusion. This improves the accuracy of tool condition monitoring methods. This method eliminates false alarms caused by transient changes in operating parameters, reducing unnecessary downtime. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 This is a schematic diagram of the hardware composition of the tool wear and damage monitoring system;

[0070] Figure 2 This is a comparison chart of the spindle vibration signal abnormal value before and after replacement;

[0071] Figure 3 is the power spectrum of the main shaft vibration signal;

[0072] Figure 4 Characteristic frequency amplitude indicators for tool wear monitoring, (a) tool pass frequency fundamental frequency amplitude ratio change curve; (b) tool pass frequency 3 times frequency amplitude ratio change curve;

[0073] Figure 5 The power spectrum band energy index change curve used for tool wear monitoring;

[0074] Figure 6 Schematic diagram of amplitude ratio for tool breakage monitoring;

[0075] Figure 7 Schematic diagram of energy ratio for tool breakage monitoring;

[0076] Figure 8 Comparison results of amplitude ratio, energy ratio and root mean square index for tool breakage monitoring

[0077] Figure 9 Schematic diagram of the change of cutting edge envelope after tool wear and its influence on the machining size of parts;

[0078] Figure 10 Schematic diagram of tool failure determination based on simultaneous diagnosis of MRR and comprehensive monitoring indicators;

[0079] Figure 11 This is the tool breakage monitoring result diagram;

[0080] Figure 12 This is the functional diagram of the proposed tool condition monitoring system;

[0081] Figure 13 This is the proposed scheme diagram of the tool condition monitoring system for a single CNC machine tool. DETAILED DESCRIPTION

[0082] The present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0083] A real-time monitoring method and system for tool wear and damage based on spindle vibration signals includes the following steps:

[0084] Step 1: Synchronous acquisition of spindle vibration signals and internal data of the CNC system

[0085] On the DMU 50 machine tool, OPC UA is used to collect internal CNC system data, including the tool name and program name of the executing NC program, as well as real-time spindle speed, feed rate, and X, Y, and Z axis coordinates. A single three-axis accelerometer mounted on the spindle sidewall collects spindle vibration data in the X, Y, and Z directions. The vibration generated by the tool-holder-spindle process system is captured in real time by the accelerometer located close to the cutting area. Figure 1 This figure shows the hardware and data transmission diagram of the tool condition monitoring system proposed in this invention. To associate the manufacturing data generated during each tool's machining process with the tool name and other information, and to match the tool's monitoring thresholds with the machining scenario, the raw data collected by the sensor and the monitoring system's output data are named after the tool and program names.

[0086] Step 2: Data Preprocessing

[0087] The raw vibration signals collected by sensors contain a significant amount of noise, including measurement errors. This interference can lead to inaccurate monitoring results and false alarms. Therefore, the raw data is processed using outlier replacement, removing the mean and trend terms. The method for identifying outlier data points can be determined based on a Gaussian statistical distribution, with outlier data replaced by the mean of the entire data set. Figure 2 This is a comparison chart of the results before and after replacing the data outlier points. After outlier data processing, false alarms caused by sensor measurement errors can be reduced.

[0088] Step 3: Tool wear and tear monitoring indicators

[0089] 3.1 Power spectrum analysis of vibration signals

[0090] The power spectrum (PSD) of a signal reflects the distribution of power energy of each frequency component of a random signal. By performing power spectrum analysis on the pre-processed vibration signal in the three directions of the spindle X, Y, and Z, the direct method for calculating the power spectrum is given as follows (1):

[0091]

[0092] Figure 3 It is the frequency component of the power spectrum of the main shaft vibration signal, and the frequency range is 0-300Hz.

[0093] 3.2 Tool wear monitoring indicators based on vibration signal power spectrum

[0094] The power spectrum characteristic frequency amplitude is selected as the monitoring indicator, and the mathematical definition of the power spectrum characteristic frequency amplitude indicator is given as follows (2):

[0095] F=Amp psd (f) (2)

[0096] Where Amp(f) is the amplitude corresponding to the characteristic frequency f.

[0097] Define the tooth passing frequency as f p , its size can be obtained by the formula f p =(n / 60)·N t Calculated, n is the spindle speed, N t is the number of tool teeth. Select the tool tooth passing frequency f in the power spectrum component respectively p and triple frequency 3f p Amplitude is used as a monitoring indicator. Figure 4 The power spectrum of the spindle vibration signal during the machining of rectangular slot parts shows the fundamental frequency and 3rd frequency amplitude change curve of the cutter tooth passing frequency. (a) is the cutter tooth passing frequency f p Amplitude change curve; (a) is the triple frequency 3f of the knife tooth passing frequency p Amplitude change curve. Figure 4 It can be seen that in the machining process of complex parts, the proposed characteristic frequency amplitude index can well reflect the tool wear degradation process.

[0098] By analyzing the relative distribution of the power spectrum frequency components of the spindle vibration signal at different wear stages, a monitoring index reflecting the tool wear change - the power spectrum band energy index is proposed. p The cumulative sum of the amplitudes of the frequency components within the frequency band. The mathematical definition of the power spectrum band energy index is given as follows (3):

[0099]

[0100] Figure 5 This is the change curve of the power spectrum band energy index as the tool wear degree increases during the processing of the groove cavity parts. If a single transient abnormal value appears in the monitoring curve, it can be determined as tool wear failure by exceeding the monitoring threshold twice within 5 seconds.

[0101] 3.3 Tool breakage monitoring indicators based on vibration signal power spectrum

[0102] To address the challenge of tool breakage monitoring under time-varying cutting conditions, a tool breakage monitoring index based on dimensionless monitoring indicators is proposed. Based on the amplitude and distribution of the power spectrum frequency components before and after tool breakage, two dimensionless monitoring indicators that are sensitive to tool breakage and unaffected by operating condition parameters are defined.

[0103] According to the relative change in the amplitude of the frequency components of the power spectrum of the vibration signal before and after tool breakage, the mathematical definition of the dimensionless index - amplitude ratio is given as follows (4):

[0104]

[0105] Among them, Amp psd (f p ) is the fundamental frequency amplitude of the blade tooth passing frequency, Amp psd (2f p ) is the amplitude of twice the blade tooth passing frequency. Figure 6 This is a graph of the amplitude ratio index used for tool breakage monitoring.

[0106] According to the relative changes in the frequency components and distribution of the power spectrum of the vibration signal before and after tool breakage, the mathematical definition of the dimensionless index - energy ratio is given as follows (5):

[0107]

[0108] in, is the low-frequency band energy of the vibration signal power spectrum, is the high-frequency energy of the tool vibration signal's power spectrum. The boundary between high and low frequencies in the vibration signal is determined based on the data sampling rate and machine tool dynamics. For a sampling rate of 5120 Hz, the 10th frequency of the tool tooth's transit frequency is selected as the high- and low-frequency boundary. Figure 7 This is a graph of energy ratio indicators used for tool breakage monitoring. Figure 8 This is the amplitude change curve of the amplitude ratio and energy ratio indicators proposed in the present invention before and after tool breakage. It can be seen that the monitoring indicators are very sensitive to the occurrence of tool breakage.

[0109] 3.4、Indicator normalization processing

[0110] In order to achieve further dimensionality reduction and fusion of tool condition monitoring indicators, the mathematical definition of normalization of different indicators is given as follows (6):

[0111]

[0112] Among them, K normal K is the amplitude of each monitoring indicator when the tool is normal, obtained by trial cutting with a normal tool. In order to make the amplitude of the monitoring indicator under normal working conditions well representative, the number of monitoring indicators is usually selected to be 60 or more data points; i K is the monitoring index before normalization at different wear stages of the tool; t It is the monitoring index of different degradation stages of the tool after normalization.

[0113] 3.5 Indicator Fusion

[0114] After the indicators are normalized, the mathematical definition of the fusion between different indicators is given as follows (7):

[0115]

[0116] Among them, K synthetic is a comprehensive monitoring index for tool breakage or a comprehensive monitoring index for tool breakage; n is the weight of each monitoring indicator, which can be determined according to the sensitivity of different indicators to the tool wear and damage status; K tn It is a monitoring indicator used for feature fusion after normalization, including indicators of different directions and types.

[0117] By normalizing and fusing dimensionless indicators through equations (6) and (7), comprehensive monitoring indicators reflecting tool wear and tool breakage can be obtained. In addition, the feature fusion method can also be used with the help of other intelligent methods, such as principal component analysis and clustering.

[0118] Step 4: Tool wear and breakage threshold setting

[0119] The fourth step is specifically as follows:

[0120] 4.1. Tool wear threshold setting

[0121] The tool wear threshold is set based on the part processing quality as a constraint. Processing quality refers to the allowable dimensional deviation of the key dimensions of the part and the allowable surface roughness deviation of the part processing surface. Through trial cutting experiments, a database of correlations between the part processing quality and comprehensive tool wear indicators at different tool wear stages is established. According to the different quality requirements of the processed parts, the tool wear threshold can be dynamically set to maximize the use of tool life. The mathematical definition of the tool wear failure threshold is given as shown in the following formula (8):

[0122] T wear =min{K tolerance ,K roughness} (8)

[0123] Among them, T wear K is the tool wear monitoring threshold; tolerance K is the maximum amplitude of the comprehensive monitoring index within the tolerance range; roughness It is the maximum amplitude of the comprehensive monitoring index within the allowable range of surface roughness. Figure 9 Schematic diagram of the cutting edge envelope before and after tool wear and its impact on part machining accuracy.

[0124] 4.2. Tool wear threshold setting

[0125] The tool breakage process is unpredictable and is a sudden phenomenon. Once the cutting edge is damaged, the amplitude of the comprehensive tool breakage monitoring index will quickly deviate from the normal state, which is very different from the index amplitude of the tool in the normal state. A floating monitoring threshold setting method based on Gaussian statistical distribution is proposed to minimize the complexity of setting the tool breakage threshold. The mathematical definition of the tool breakage monitoring floating threshold is given as follows (9):

[0126]

[0127] Among them, T breakage is the tool breakage monitoring threshold; μ is the mean of the historical tool breakage comprehensive monitoring index; σ is the standard deviation of the historical tool breakage comprehensive monitoring index; N is the number of historical tool breakage comprehensive monitoring indicators; x(j) is the historical tool breakage comprehensive monitoring index.

[0128] Step 5: Identification of operating parameter changes

[0129] During machining, when a tool removes material from different locations on a part, considering that the spindle load may change suddenly at corners and steps, the feed rate is often reduced in the G code to keep the spindle load fluctuating within a small range. The material removal rate (MRR), which measures the change in working condition parameters, is defined as follows (10):

[0130] MRR(t)=v f (t)·a p (t)·a e (t) (10)

[0131] Among them, v f is the machine feed speed; a p is the axial cutting depth of the tool; a e is the radial cutting width of the tool.

[0132] In most machining situations, the axial and radial cutting depths of the tool remain basically constant. The feed per tooth of the tool can also be used to identify and monitor changes in working parameters. The feed per tooth, which measures changes in working parameters, is defined as follows:

[0133]

[0134] Step 5: Method for determining tool wear, breakage and failure

[0135] During the process of removing blank material due to the relative motion between the tool and the workpiece, changes in cutting parameters often cause fluctuations in the amplitude of monitoring indicators, while the tool is in a normal state. In this case, false alarms are easily generated. A method for distinguishing tool breakage and tool breakage failure is proposed by synchronously monitoring comprehensive monitoring indicators and working condition parameters.

[0136] When the comprehensive monitoring index exceeds the threshold, the system first determines whether the fluctuation in the index amplitude is caused by a change in the operating parameters. If the increase in the monitoring index amplitude is caused by a change in the operating parameters, no alarm message will be issued, and the tool condition monitoring system will continue to monitor. If the comprehensive monitoring index reaches the failure threshold while the operating parameters remain constant, the tool is determined to have failed.

[0137]

[0138] Among them, Normal means the tool is in a normal state; Fault means the tool is in a failure state, including excessive tool wear and abnormal tool damage; Q is the material removal rate change threshold; K synthetic K is a comprehensive monitoring indicator for tool wear or tool breakage; threshold It is the failure threshold of tool wear or tool breakage. Figure 10 Schematic diagram of tool wear failure judgment, (a) is the process of tool failure judgment; (b) is the curve diagram of tool failure judgment. Figure 11 This example shows the results of a tool breakage monitoring case. At the moment of a change in operating conditions, the tool's comprehensive monitoring indicators may exhibit slight amplitude fluctuations under normal conditions, potentially creating a false alarm. However, the proposed method, when identifying a material removal rate increment greater than a threshold, comprehensively assesses whether the amplitude fluctuations in the indicators are due to a change in operating conditions, rather than tool breakage. Furthermore, comprehensive monitoring indicators can quickly identify tool breakage at the moment of breakage, significantly reducing damage to parts caused by tool breakage. Figure 12 This is a flow chart of the tool wear and damage status monitoring proposed by the present invention based on the vibration acceleration sensor. Based on this hardware, other process monitoring functions can also be further expanded, such as cutting vibration monitoring and part surface roughness monitoring.

[0139] Step 6: Tool Condition Monitoring System for Flexible Production Lines

[0140] Spindle vibration data collected by an external three-axis accelerometer and internal data from the CNC system can be used to monitor tool wear and breakage. The tool wear and breakage monitoring algorithm proposed in this invention is packaged and embedded in tool condition monitoring software developed using LabVIEW. The tool condition monitoring system hardware for a single machine tool consists of a three-axis accelerometer, an industrial control unit with built-in (IEPE / ICP) data acquisition circuitry, and a touchscreen display. Real-time monitoring of tool wear and breakage is achieved through the embedded monitoring algorithm industrial software. Figure 13This diagram shows the tool condition monitoring system scheme for a single CNC machine tool. Furthermore, in a flexible production line, each machine tool's data acquisition and industrial control unit can locally collect, process, and display data. Simultaneously, the monitoring data generated by each machine tool can be transmitted to a central server via a switch, allowing for remote display of tool wear and breakage data.

[0141] In yet another embodiment of the present invention, a real-time monitoring system for tool wear and damage based on a spindle vibration signal is provided, which can be used to implement the above-mentioned real-time monitoring method for tool wear and damage based on a spindle vibration signal. Specifically, the system includes:

[0142] Data acquisition module, used to collect spindle XYZ direction vibration and internal data of CNC system;

[0143] Preprocessing module, used to perform noise reduction on the collected data;

[0144] An index acquisition module is used to perform power spectrum analysis on the spindle vibration data after noise reduction to obtain power spectrum frequency band energy index and dimensionless monitoring index reflecting tool wear changes;

[0145] The fusion module is used to fuse the indicators to obtain comprehensive monitoring indicators reflecting tool wear and tool breakage;

[0146] The judgment module, when the comprehensive monitoring indicators exceed the failure threshold, uses the material removal rate increment to assist in determining whether the tool has failed, thereby eliminating false alarms generated at the moment of working condition parameter switching and accurately identifying the tool status.

[0147] The module division in the embodiments of the present invention is illustrative and represents only one logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional modules in various embodiments of the present invention may be integrated into a single processor, exist physically as separate modules, or two or more modules may be integrated into a single module. The integrated modules may be implemented in either hardware or software functional modules.

[0148] In another embodiment of the present invention, a computer device is provided, comprising a processor and a memory, wherein the memory is used to store a computer program, wherein the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., which is the computing core and control core of the terminal, and is suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in the computer storage medium to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of a real-time monitoring method for tool wear and damage based on a spindle vibration signal.

[0149] In another embodiment of the present invention, the present invention further provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device for storing programs and data. It is understandable that the computer-readable storage medium here can include both built-in storage media in the computer device and, of course, extended storage media supported by the computer device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space. These instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the above-mentioned embodiment regarding a real-time monitoring method for tool wear and breakage based on a spindle vibration signal.

[0150] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0151] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0152] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0153] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0154] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A real-time monitoring method for tool wear and damage based on spindle vibration signals, characterized in that: include: Collect CNC machine tool spindle vibration data and CNC system internal data; Perform noise reduction on the collected data; By performing power spectrum analysis on the spindle vibration data after noise reduction, the power spectrum frequency band energy index reflecting tool wear and the dimensionless monitoring index reflecting tool breakage are obtained; By integrating the indicators, we can obtain comprehensive monitoring indicators reflecting tool wear and tool breakage; When the comprehensive monitoring index exceeds the failure threshold, the material removal rate increment is used to assist in determining whether the tool has failed, thereby eliminating false alarms caused by the switching of working parameters and accurately identifying the tool status; Power spectrum band energy index: The power spectrum of the vibration signal in the three directions of the spindle XYZ after preprocessing is analyzed and the power spectrum is calculated by direct method as follows (1): The power spectrum characteristic frequency amplitude is selected as the monitoring indicator, and the mathematical definition of the power spectrum characteristic frequency amplitude indicator is given as follows (2): F=Amp psd (f) (2) Among them, Amp psd (f) is the amplitude corresponding to the characteristic frequency f; Define the tooth passing frequency as f p , its size is determined by the formula f p =(n / 60)·N t Calculated, n is the spindle speed, N t is the number of tool teeth; select the tool tooth passing frequency f in the power spectrum component respectively p and triple frequency 3f p Amplitude is used as a monitoring indicator; By analyzing the relative distribution of the power spectrum frequency components of the spindle vibration signal at different wear stages, a monitoring indicator reflecting the tool wear change, the power spectrum band energy index, is proposed; 0-7f p The cumulative sum of the amplitudes of the frequency components within the frequency band gives the power spectrum band energy index, which is mathematically defined as follows: Dimensionless monitoring indicators: According to the amplitude and distribution of the power spectrum frequency components before and after tool breakage, two dimensionless monitoring indicators that are sensitive to tool breakage but not affected by working parameters are defined. According to the relative change in the amplitude of the frequency components of the power spectrum of the vibration signal before and after tool breakage, the mathematical definition of the dimensionless index - amplitude ratio is given as follows (4): Among them, Amp psd (f p ) is the fundamental frequency amplitude of the blade tooth passing frequency, Amp psd (2f p ) is the amplitude of twice the frequency of the blade tooth passing through; According to the relative changes in the frequency components and distribution of the power spectrum of the vibration signal before and after tool breakage, the mathematical definition of the dimensionless index - energy ratio is given as follows (5): in, is the low-frequency band energy of the vibration signal power spectrum, is the high frequency band energy of the knife vibration signal power spectrum.

2. The real-time monitoring method for tool wear and damage based on spindle vibration signals according to claim 1 is characterized in that: The tool name, program name, real-time spindle speed, feed speed, and XYZ axis coordinate data within the CNC system are collected through the OPC UA protocol; The spindle XYZ vibration data is collected in real time through a three-axis acceleration sensor installed on the side wall of the spindle.

3. The real-time monitoring method for tool wear and damage based on spindle vibration signals according to claim 1 is characterized in that: Noise reduction: Replace outliers on the original data and remove mean and trend items.

4. The real-time monitoring method for tool wear and damage based on spindle vibration signals according to claim 1 is characterized in that: By integrating the indicators, we can obtain comprehensive monitoring indicators reflecting tool wear and tool breakage: The mathematical definition of normalization of different indicators is as follows (6): Among them, K normal is the amplitude of each monitoring index when the tool is normal, K i is the monitoring index before normalization of tool wear at different stages, K t It is the monitoring index of different degradation stages of the tool after normalization; After the indicators are normalized, the mathematical definition of the fusion between different indicators is given as follows (7): Among them, K synthetic is a comprehensive monitoring index for tool breakage or a comprehensive monitoring index for tool breakage; n is the weight of each monitoring indicator; K tn It is the monitoring index used for feature fusion after normalization, including indicators between different directions and different types; through the normalization processing and dimensionless index fusion through formula (6)-(7), the comprehensive monitoring index reflecting tool wear and tool breakage is obtained.

5. The real-time monitoring method for tool wear and damage based on spindle vibration signals according to claim 1 is characterized in that: When the comprehensive monitoring indicators exceed the failure threshold, the material removal rate increment is used to assist in determining whether the tool has failed, thereby eliminating false alarms caused by the switching of working parameters and accurately identifying the tool status: The tool wear threshold is set based on the part processing quality as a constraint. The processing quality refers to the allowable dimensional deviation of the key dimensions of the part and the allowable surface roughness deviation of the part processing surface. Through trial cutting, a correlation database between the part processing quality and the comprehensive tool wear index at different tool wear stages is established. The tool wear threshold is dynamically set according to the different quality requirements of the processed parts to maximize the use of tool life. The mathematical definition of the tool wear failure threshold is given as shown in the following formula (8): T wear =min{K tolerance ,K roughness }(8) Among them, T wear K is the tool wear monitoring threshold; tolerance K is the maximum amplitude of the comprehensive monitoring index within the tolerance range; roughness The maximum amplitude of the comprehensive monitoring index within the allowable range of surface roughness; The tool breakage threshold is set based on the Gaussian statistical distribution method for identifying abnormal data. The mathematical definition of the tool breakage monitoring floating threshold is shown in the following formula (9): Among them, T breakage is the tool breakage monitoring threshold; μ is the mean of the historical tool breakage comprehensive monitoring index; σ is the standard deviation of the historical tool breakage comprehensive monitoring index; N is the number of historical tool breakage comprehensive monitoring indexes; x(j) is the historical tool breakage comprehensive monitoring index; The material removal rate MRR, which measures the change of working condition parameters, is defined as follows (10): MRR(t)=v f (t)·a p (t)·a e (t)(10)where v f is the machine feed speed; a p is the axial cutting depth of the tool; a e is the radial cutting width of the tool; The feed rate per tooth of the tool can also be used to identify and monitor the changes in working parameters under fixed cutting width and cutting depth. The feed rate per tooth that measures the changes in working parameters is defined as follows (11): When the comprehensive monitoring index exceeds the threshold, it is first determined whether the index amplitude fluctuation is caused by the change of the working condition parameters. If the increase in the monitoring index amplitude is caused by the change of the working condition parameters, no alarm information will be issued, and the tool status monitoring system will continue to monitor the tool status. If the comprehensive monitoring index reaches the failure threshold while the working condition parameters remain constant, it is determined that the tool has failed. Among them, Normal means the tool is in a normal state; Fault means the tool is in a failure state, including excessive tool wear and abnormal tool damage; Q is the material removal rate change threshold; K synthetic It is a comprehensive monitoring indicator for tool wear or tool breakage; T threshold It is the failure threshold of tool wear or tool breakage.

6. A real-time monitoring system for tool wear and damage based on a spindle vibration signal, which runs the real-time monitoring method for tool wear and damage based on a spindle vibration signal according to any one of claims 1 to 5, characterized in that: include: Data acquisition module, used to collect spindle vibration and internal data of CNC system; Preprocessing module, used to perform noise reduction on the collected data; The indicator design module is used to perform power spectrum analysis on the spindle vibration data after noise reduction to obtain power spectrum band energy indicators and dimensionless monitoring indicators that reflect tool wear changes; The fusion module is used to fuse the indicators to obtain comprehensive monitoring indicators reflecting tool wear and tool breakage; The comprehensive judgment module uses the material removal rate increment to assist in determining whether the tool has failed when the comprehensive monitoring indicators exceed the failure threshold, thereby eliminating false alarms generated at the moment of working condition parameter switching and accurately identifying the tool status.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the real-time monitoring method for tool wear and damage based on a spindle vibration signal as described in any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the real-time monitoring method for tool wear and damage based on a spindle vibration signal as claimed in any one of claims 1 to 5 are implemented.

Images