A CNC machine tool health monitoring method, system and program product based on multi-source data fusion

Abstract

The present invention discloses a CNC machine tool health monitoring method, system and program product based on multi-source data fusion, which belongs to the technical field of CNC machine tool health monitoring. The sensor module collects multi-source data such as spindle vibration and guide rail offset, and after preprocessing by the data acquisition module, the error judgment module determines the abnormal machine tool. The core lies in the health analysis module, which calculates the amplitude difference and offset difference of each machine tool based on the preset standard error threshold, and performs basic grouping according to the principle of three times the standard deviation, and divides the machine tool into a stable group, a normal group and a fluctuation group. In the subsequent processing process, data is continuously collected and the fluctuation value is calculated, and the machine tool state is judged according to the fluctuation value and the grouping is dynamically adjusted to achieve real-time tracking of the health status of the machine tool. Finally, the CNC machine tool management module marks the machine tool according to the grouping results after dynamic adjustment and formulates corresponding maintenance strategies, thereby improving the stability, processing accuracy and maintenance efficiency of the machine tool.

Description

Technical Field

[0001] The present invention belongs to the technical field of CNC machine tool health monitoring, and relates to machine tool health monitoring and status assessment technology. Specifically, it is a CNC machine tool health monitoring method, system and program product based on multi-source data fusion, which is used to realize comprehensive monitoring of the health status of CNC machine tools, dynamic adjustment and formulation of precise maintenance strategies, and improve equipment stability and processing accuracy. Background Art

[0002] CNC machine tools are machine tools that use computer control systems to achieve automated processing. Their main feature is that they can perform high-precision and high-efficiency processing operations. They are suitable for the production of complex parts. Through pre-programmed instructions, they control the various moving parts of the machine tool to move along a specific path and speed to complete the processing of the workpiece. As the core equipment of modern manufacturing, the health of CNC machine tools is directly related to processing accuracy, production efficiency and economic benefits. With the increasing demand for high-precision, high-efficiency and high-reliability processing in the manufacturing industry, the health monitoring of CNC machine tools has also received increasing attention.

[0003] Existing CNC machine tool health monitoring is mostly based on fixed cycles through standard thresholds, which uniformly monitor the key components of CNC machine tools (such as spindles and guide rails, etc.), and usually use a single vibration sensor or displacement sensor for data collection, and judge the equipment status through preset standard thresholds. However, this single data source monitoring method is difficult to fully reflect the actual operating status of the machine tool, and is prone to misjudgment or missed judgment. At the same time, the fixed cycle detection method cannot adapt to the actual usage intensity and working condition differences of different machine tools, resulting in the inability to reasonably allocate maintenance resources.

[0004] In recent years, with the rapid development of sensor technology, data acquisition technology, and data analysis technology, condition-based maintenance (CBM) has gradually become a research hotspot in the field of CNC machine tool maintenance. The core idea of ​​CBM is to monitor the operating status of machine tools in real time, use various sensors to collect machine tool vibration, temperature, current, acoustic emission and other signals, and combine data analysis technology to process and analyze the collected data, so as to evaluate the health of machine tools, predict potential failures, and formulate corresponding maintenance strategies. Compared with traditional regular maintenance and post-maintenance, CBM can more effectively improve the reliability of machine tools, reduce maintenance costs, and improve production efficiency.

[0005] However, in the actual production process, due to the large number of CNC machine tools in the factory and the huge order volume, the health detection cycle span is long, the potential problems of CNC machine tools are difficult to find, and it is impossible to carry out targeted health detection management of the CNC machine tools under its jurisdiction, which has problems of low practicality and functionality. Secondly, there are differences in time domain characteristics between different sensor data, and how to effectively fuse and analyze these heterogeneous data is still a challenge. Third, there is a lack of intelligent group management mechanism, and it is impossible to perform dynamic adjustment and targeted maintenance according to the actual status of the equipment. At the same time, many systems use fixed thresholds to judge the health status of machine tools. However, the operating status of machine tools will be affected by many factors, such as processing workpieces, cutting parameters, ambient temperature, etc. Fixed thresholds are difficult to adapt to changes in the operating status of machine tools, which can easily lead to misjudgment. In addition, the health status of machine tools is a dynamically changing process. Most existing systems lack dynamic adjustment mechanisms, cannot reflect changes in machine tool status in a timely manner, and are difficult to achieve accurate health assessment.

[0006] In summary, the existing CNC machine tool health monitoring technology still has many shortcomings in terms of the comprehensiveness of data collection, the integration of multi-source data, the dynamics of intelligent grouping management, and the real-time nature of accurate health assessment. Therefore, how to develop a CNC machine tool health monitoring technology that can fully integrate multi-source data, have dynamic adjustment capabilities, and achieve accurate early warning is a technical problem that needs to be solved urgently. Summary of the invention

[0007] 1. Purpose of the invention

[0008] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes a CNC machine tool health monitoring method, system and program product based on multi-source data fusion, which solves the above technical problems by improving the detection method and processing method, specifically by real-time collection of multi-source data such as spindle vibration and guide rail offset, dynamic grouping and fluctuation value analysis combined with standard error thresholds, and formulating intelligent maintenance strategies for different health states, so as to achieve comprehensive monitoring and accurate evaluation of the health state of CNC machine tools, effectively improve the real-time monitoring of equipment operation status, the accuracy of evaluation results and the utilization rate of maintenance resources, and significantly improve the reliability and processing accuracy of CNC machine tools.

[0009] (II) Technical solution

[0010] In order to achieve the purpose of the invention and solve the technical problems, the present invention adopts the following technical solutions:

[0011] The first invention object of the present invention is to provide a CNC machine tool health monitoring method based on multi-source data fusion, which is used to monitor and dynamically manage the health status of multiple CNC machine tools in an industrial production environment in real time, predict potential faults and formulate targeted maintenance strategies to improve the machining accuracy and operation stability of CNC machine tools. The health monitoring method includes at least the following steps when implemented:

[0012] SS1. Data collection

[0013] Collect multi-source data through sensors on all subordinate machine tools, generate analog signals, and convert the analog signals into digital signals, wherein the multi-source data includes spindle vibration data and guide rail offset data;

[0014] SS2. Error determination

[0015] The spindle amplitude standard error threshold and the guide rail offset standard error threshold are set based on the historical operation data of the machine tool, and the spindle vibration data and guide rail offset data of all the subordinate machine tools are judged according to the two standard error thresholds. The initial health status of each machine tool is judged by calculating the difference between the spindle amplitude of each machine tool and its standard error threshold and the difference between the guide rail offset and its standard error threshold, and the abnormal machine tools with any indicator greater than their standard error threshold and the normal machine tools with all indicators within their standard error threshold range are screened out.

[0016] SS3. Initial grouping

[0017] Based on the error judgment results, all normal machine tools under its jurisdiction are grouped into basic groups. In the basic grouping process, the mean and standard deviation of the two indicators, spindle amplitude difference and guide rail offset difference, of all normal machine tools under its jurisdiction are calculated according to the principle of three times standard deviation. The spindle amplitude and guide rail offset are divided into stable interval, normal interval and fluctuation interval accordingly. Based on the interval distribution of the two indicators, all normal machine tools under its jurisdiction are divided into stable group, normal group and fluctuation group.

[0018] SS4. Dynamic grouping adjustment

[0019] In the subsequent parts processing process, the spindle vibration data and guideway offset data of the machine tools in each group are continuously collected and recorded, and the spindle amplitude fluctuation value and guideway offset fluctuation value of each machine tool are calculated. Based on the fluctuation values ​​of the two indicators and combined with the median absolute deviation method, the health status of the machine tools in each group is re-evaluated and the grouping is dynamically adjusted. When the fluctuation value of the machine tool in the stable group exceeds its stable threshold range, it is downgraded to the normal group or the fluctuation group. When the fluctuation value of the machine tool in the normal group exceeds or falls below the normal range, it is upgraded to the stable group or downgraded to the fluctuation group respectively. When the fluctuation value of the machine tool in the fluctuation group returns to the normal range, it is upgraded to the normal group and downgraded to the abnormal machine tool when it is below the normal range.

[0020] SS5. Maintenance strategy formulation

[0021] After multiple dynamic adjustment groupings, differentiated maintenance strategies are formulated for the machine tools in each group based on the final grouping results. Longer maintenance cycles are set for the machine tools in the stable group, moderate maintenance cycles are set for the machine tools in the normal group, and short-term and high-frequency maintenance cycles are set for the machine tools in the fluctuating group. Real-time reminders are given for detailed inspections and repairs of abnormal machine tools.

[0022] The second invention object of the present invention is to provide a CNC machine tool health monitoring system based on multi-source data fusion, the system comprising at least a sensor module, a data acquisition module, an error judgment module, a health analysis module and a CNC machine tool management module, wherein:

[0023] The sensor module includes at least a vibration sensor and a linear displacement sensor, which are used to record the spindle vibration data and guide rail offset data of all the machine tools under its jurisdiction and generate analog signals;

[0024] The data acquisition module is communicatively connected to the sensor module, and is used to receive analog signals from the sensor module, and convert the analog signals into digital signals and transmit them to the error determination module and the health analysis module;

[0025] The error judgment module is communicatively connected to the data acquisition module, sets a spindle amplitude standard error threshold and a guide rail offset standard error threshold based on the historical operation data of the machine tool, and performs error judgment on the spindle vibration data and guide rail offset data of the subordinate machine tools according to the two standard error thresholds, and judges the preliminary health status of each machine tool by calculating the difference between the spindle amplitude of each machine tool and its standard error threshold and the difference between the guide rail offset and its standard error threshold, and screens out abnormal machine tools with any indicator greater than its standard error threshold and normal machine tools with both indicators within their standard error threshold ranges;

[0026] The health analysis module is communicatively connected to the data acquisition module and the error judgment module, and based on the spindle vibration data and guide rail offset data of each machine tool, combined with the standard error threshold, the mean and standard deviation of the two indicators, the spindle amplitude difference and the guide rail offset difference of all the normal machine tools under its jurisdiction are calculated respectively, and the principle of three times the standard deviation and the interval distribution of the two indicators are combined to perform basic grouping of all the machine tools under its jurisdiction and divide them into a stable group, a normal group and a fluctuation group, and the spindle vibration data and guide rail offset of the machine tools in each group in the subsequent processing process are statistically analyzed to determine the fluctuation value and adjust the machine tool grouping in combination with the median absolute deviation method, when the fluctuation value of the machine tool in the stable group exceeds its stable threshold range, it is downgraded to the normal group or the fluctuation group, when the fluctuation value of the machine tool in the normal group exceeds or is lower than the normal range, it is upgraded to the stable group or downgraded to the fluctuation group, and when the fluctuation value of the machine tool in the fluctuation group returns to the normal range, it is upgraded to the normal group and downgraded to an abnormal machine tool when it is lower than the normal range;

[0027] The CNC machine tool management module is communicatively connected to the health analysis module. Based on the grouping results of the health analysis module, after multiple regroupings, each grouped machine tool is marked according to the final grouping results and a differentiated maintenance strategy is determined. A longer maintenance cycle is set for the stable group machine tools, a moderate maintenance cycle is set for the normal group machine tools, and a short-term and high-frequency maintenance cycle is set for the fluctuating group machine tools. Machine tools with abnormal markings are reminded in real time to conduct detailed inspections and repairs.

[0028] The third inventive object of the present invention is to provide a computer program product, including computer instructions, for executing the above-mentioned CNC machine tool health monitoring method based on multi-source data fusion.

[0029] (III) Technical Effect

[0030] Compared with the prior art, the CNC machine tool health monitoring method, system and program product based on multi-source data fusion of the present invention have the following beneficial and significant technical effects:

[0031] (1) The present invention performs the first error analysis on the subordinate CNC machine tools, divides the equipment whose error fluctuation is below the threshold into groups based on the degree of proximity to the error threshold, and adjusts the machine tools in different groups based on the error fluctuation in the group during subsequent part processing, so as to identify stable, normal and fluctuating CNC machine tools, and realize the identification and management of CNC machine tools with different health levels;

[0032] (2) The present invention adjusts the machine tool grouping through the initial error analysis and the subsequent intra-group error fluctuations, which helps to stabilize the grouping of subordinate machine tools, accurately identify CNC machine tools with potential risks, and set different comprehensive maintenance cycles based on machine tools in different groups, which helps to take targeted preventive maintenance measures, avoid unexpected downtime and major failures, and enhance the practicality and functionality of the system;

[0033] (3) The present invention timely adjusts the grouping of CNC machine tools based on the conditions of CNC machine tools during parts processing, and can timely identify and respond to changes in equipment status, thus avoiding the management lag problem caused by fixed grouping. At the same time, through the analysis of intra-group error fluctuations, the real-time health status of the equipment can be identified, thereby improving the accuracy of CNC machine tool health judgment;

[0034] (4) The present invention realizes comprehensive monitoring and management of the health status of CNC machine tools by integrating sensor modules, data acquisition modules, error judgment modules, health analysis modules and CNC machine tool management modules. The system improves the stability and processing accuracy of CNC machine tools through real-time data acquisition, error judgment, dynamic grouping adjustment and precise maintenance strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0036] Figure 1 This is a flow chart of a CNC machine tool health monitoring method based on multi-source data fusion according to an embodiment of the present invention;

[0037] Figure 2 The figure is a block diagram of a CNC machine tool health monitoring system based on multi-source data fusion according to an embodiment of the present invention. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical scheme and advantages of the implementation of the present invention clearer, the technical scheme in the embodiment of the present invention will be described in more detail below in conjunction with the drawings in the embodiment of the present invention. The described embodiments are part of the embodiments of the present invention, not all of the embodiments, and the described embodiments are exemplary and are intended to be used to explain the present invention, and cannot be understood as limiting 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.

[0039] The present invention aims to provide a CNC machine tool health monitoring method, system and program product based on multi-source data fusion. By real-time collection of multi-source data such as spindle vibration and guide rail offset, dynamic grouping and fluctuation value analysis are performed in combination with standard error thresholds, and intelligent maintenance strategies are formulated for different health states. Comprehensive monitoring and accurate evaluation of the health state of CNC machine tools can be achieved, effectively improving the real-time monitoring of equipment operation status, the accuracy of evaluation results and the utilization rate of maintenance resources.

[0040] Embodiment 1:

[0041] As a specific example, Figure 1 As shown, the CNC machine tool health monitoring method based on multi-source data fusion provided by the embodiment of the present invention mainly includes the following steps when implemented:

[0042] SS1. Data collection

[0043] Collect multi-source data through sensors on all machine tools under its jurisdiction, generate analog signals, and convert analog signals into digital signals. Multi-source data includes spindle vibration data and guide rail offset data;

[0044] SS2. Error determination

[0045] The spindle amplitude standard error threshold and the guide rail offset standard error threshold are set based on the historical operation data of the machine tool, and the spindle vibration data and guide rail offset data of all the subordinate machine tools are judged according to the two standard error thresholds. The initial health status of each machine tool is judged by calculating the difference between the spindle amplitude of each machine tool and its standard error threshold and the difference between the guide rail offset and its standard error threshold, and the abnormal machine tools with any indicator greater than their standard error threshold and the normal machine tools with all indicators within their standard error threshold range are screened out.

[0046] SS3. Initial grouping

[0047] Based on the error judgment results, all normal machine tools under its jurisdiction are grouped into basic groups. In the basic grouping process, the mean and standard deviation of the two indicators, spindle amplitude difference and guide rail offset difference, of all normal machine tools under its jurisdiction are calculated according to the principle of three times standard deviation. The spindle amplitude and guide rail offset are divided into stable interval, normal interval and fluctuation interval accordingly. Based on the interval distribution of the two indicators, all normal machine tools under its jurisdiction are divided into stable group, normal group and fluctuation group.

[0048] SS4. Dynamic grouping adjustment

[0049] In the subsequent parts processing process, the spindle vibration data and guideway offset data of the machine tools in each group are continuously collected and recorded, and the spindle amplitude fluctuation value and guideway offset fluctuation value of each machine tool are calculated. Based on the fluctuation values ​​of the two indicators and combined with the median absolute deviation method, the health status of the machine tools in each group is re-evaluated and the grouping is dynamically adjusted. When the fluctuation value of the machine tool in the stable group exceeds its stable threshold range, it is downgraded to the normal group or the fluctuation group. When the fluctuation value of the machine tool in the normal group exceeds or falls below the normal range, it is upgraded to the stable group or downgraded to the fluctuation group respectively. When the fluctuation value of the machine tool in the fluctuation group returns to the normal range, it is upgraded to the normal group and downgraded to the abnormal machine tool when it is below the normal range.

[0050] SS5. Maintenance strategy formulation

[0051] After multiple dynamic adjustments to the groups, differentiated maintenance strategies are formulated for the machine tools in each group based on the final grouping results. Longer maintenance cycles are set for the machine tools in the stable group, moderate maintenance cycles are set for the machine tools in the normal group, and short-term and high-frequency maintenance cycles are set for the machine tools in the fluctuating group. Abnormal machine tools are reminded in real time for detailed inspection and repair.

[0052] Through the above steps, this method realizes comprehensive health monitoring and management of CNC machine tools, effectively improves the real-time monitoring capability of equipment operation status, the accuracy of evaluation results and the utilization efficiency of maintenance resources, and is suitable for cluster management and intelligent health maintenance needs of multiple devices in modern manufacturing environments.

[0053] Embodiment 2:

[0054] As another specific example, Figure 2 As shown, the CNC machine tool health monitoring system based on multi-source data fusion provided by the embodiment of the present invention includes a sensor module, a data acquisition module, an error judgment module, a health analysis module, and a CNC machine tool management module.

[0055] The sensor module in the present invention includes a vibration sensor and a linear displacement sensor, which are used to record the vibration data of the spindle of the CNC machine tool and the offset of the guide rail to generate an analog signal.

[0056] The data acquisition module in the present invention is communicatively connected to the sensor module, and is used for receiving the analog signal of the sensor module, converting the received analog signal into a digital signal through an analog-to-digital converter, and transmitting the converted digital signal to the error judgment module and the health analysis module through Ethernet.

[0057] The error judgment module in the present invention is communicatively connected to the data acquisition module, and sets the spindle amplitude standard error threshold and the guide rail offset standard error threshold based on the historical operation data of the machine tool, and performs error judgment on the spindle vibration data and the guide rail offset data of the subordinate machine tools according to the two standard error thresholds. By calculating the difference between the spindle amplitude of each machine tool and its standard error threshold and the difference between the guide rail offset and its standard error threshold, the initial health status of each machine tool is determined, and abnormal machine tools with any indicator greater than their standard error threshold and normal machine tools with both indicators within their standard error threshold range are screened out.

[0058] The health analysis module in the present invention is communicatively connected to the data acquisition module and the error judgment module. Based on the spindle vibration data and guide rail offset data of each machine tool and combined with the standard error threshold, the mean and standard deviation of the two indicators, the spindle amplitude difference and the guide rail offset difference, of all the normal machine tools under its jurisdiction are calculated respectively. In combination with the principle of three times the standard deviation and the interval distribution of the two indicators, all the machine tools under its jurisdiction are basically grouped and divided into a stable group, a normal group and a fluctuation group. The spindle vibration data and the guide rail offset of the machine tools in each group during the subsequent processing are statistically analyzed to determine the fluctuation value and adjust the machine tool grouping in combination with the median absolute deviation method. When the fluctuation value of the machine tool in the stable group exceeds its stable threshold range, it is downgraded to the normal group or the fluctuation group. When the fluctuation value of the machine tool in the normal group exceeds or is lower than the normal range, it is upgraded to the stable group or downgraded to the fluctuation group respectively. When the fluctuation value of the machine tool in the fluctuation group returns to the normal range, it is upgraded to the normal group and downgraded to an abnormal machine tool when it is lower than the normal range.

[0059] The CNC machine tool management module in the present invention is communicatively connected to the health analysis module. Based on the grouping results of the health analysis module, after multiple regroupings, each grouped machine tool is marked according to the final grouping results and a differentiated maintenance strategy is determined. A longer maintenance cycle is set for the stable group machine tools, a moderate maintenance cycle is set for the normal group machine tools, and a short-term and high-frequency maintenance cycle is set for the fluctuating group machine tools. Machine tools with abnormal marks are reminded in real time to conduct detailed inspection and maintenance.

[0060] Embodiment 3:

[0061] Based on the above embodiment 2, as a further optimization of its content, this embodiment 3 provides a preferred implementation mode. When the error determination module in the present invention performs error determination, it includes:

[0062] Based on the spindle amplitude standard error threshold and guideway offset standard error threshold of the current CNC machine tool, the spindle amplitude and guideway offset of each subordinate machine tool are determined:

[0063] when or Z > i Z or or D > i D When it is on, it means that the relevant machine tool has an abnormality and needs to be marked for maintenance;

[0064] when or Z ≤ i Z and or D ≤ i D , it means that there is no abnormality in the related machine tool.

[0065] It should be noted that the spindle amplitude standard error threshold and guide rail offset standard error threshold of the CNC machine tool need to be set based on the CNC machine tool model and the historical machining part errors under different amplitudes or guide rail offsets.

[0066] Embodiment 4:

[0067] Based on the above-mentioned embodiment 2, this embodiment 4 provides a further preferred implementation mode. When the health analysis module in the present invention performs basic grouping and dynamic grouping adjustment, it includes:

[0068] Establish a database of CNC machine tools under your jurisdiction (e.g. through MySQL ), and number all normal machine tools, and create files in the CNC machine tool database, including the spindle amplitude of the machine tool at that time or Z and rail offset or D , based on the spindle amplitude of the subordinate machine tools or Z and rail offset or D The corresponding axis amplitude standard error threshold i Z and rail offset standard error threshold i D The difference between the two indicators, as well as the mean and standard deviation of the two difference indicators, combined with the principle of three times the standard deviation and the interval distribution of the two indicators, all the machine tools under its jurisdiction are basically grouped and divided into stable group, normal group and fluctuating group;

[0069] Based on the basic grouping of subordinate machine tools, calculate the spindle amplitude of machine tools in different groups during subsequent part processing or Z and rail offset or DThe fluctuation value of the machine tools in the stable group is used to adjust the machine tool grouping based on the fluctuation value and the median absolute deviation method after each part processing is completed. When the fluctuation value of the machine tools in the stable group exceeds the stable threshold range, they are downgraded to the normal group or the fluctuation group. When the fluctuation value of the machine tools in the normal group exceeds or falls below the normal range, they are upgraded to the stable group or downgraded to the fluctuation group respectively. When the fluctuation value of the machine tools in the fluctuation group returns to the normal range, they are upgraded to the normal group. When it is below the normal range, they are downgraded and marked as abnormal machine tools.

[0070] More specifically, when the health analysis module performs basic grouping of all the machine tools under its jurisdiction, the specific steps are as follows:

[0071] First, based on the spindle vibration data and guide rail offset data of different machine tools, combined with the spindle amplitude standard error threshold and the guide rail offset standard error threshold, the spindle amplitude difference and guide rail offset difference of each machine tool are calculated respectively. The algorithm formula is:

[0072]

[0073] in, Representative number is i The amplitude of the machine tool spindle The standard error threshold of the axis amplitude i Z The difference, Representative number is i Machine guide rail offset Standard deviation threshold from rail offset i D The difference, i =1,2,…, n , n The number of all machine tools under its jurisdiction;

[0074] Secondly, the mean and standard deviation of the spindle amplitude difference and the mean and standard deviation of the guide rail offset difference of all the subordinate machine tools are calculated respectively. The algorithm formula is:

[0075]

[0076]

[0077] in, They represent the mean of the spindle amplitude difference and the guideway offset difference of all the machine tools under their jurisdiction. They represent the standard deviation of the spindle amplitude difference and the standard deviation of the guideway offset difference of all the machine tools under their jurisdiction respectively;

[0078] Next, set the stable interval, normal interval, and fluctuation interval of the spindle amplitude and the stable interval, normal interval, and fluctuation interval of the guide rail offset, and determine the interval distribution of the spindle amplitude and guide rail offset of each machine tool under its jurisdiction: i Machine tools, when When , it means that the main axis amplitude is in a stable range; when When , it means that the main axis amplitude is in the normal range; when When , it means that the main axis amplitude is in the fluctuation range; when When , it means that the guide rail offset is in a stable range; when When , it means that the guide rail offset is in the normal range; when When , it means that the guide rail offset is in the fluctuation range;

[0079] It should be noted that the spindle amplitude difference obtained by subtracting the spindle amplitude from the standard threshold and the offset difference obtained by subtracting the guideway offset from the standard threshold represent the distance between the current normal CNC machine tool spindle amplitude and the guideway offset and the abnormal threshold. That is, the larger the difference, the farther the distance to the threshold, and the more stable the CNC machine tool. The difference distribution interval is obtained based on the principle of three times the standard deviation. or when When , it means that the spindle amplitude and guide rail offset change very little, and can be maintained in a relatively stable state, with little impact on the machining accuracy. or When the CNC machine tool spindle and guide rail are running normally, there are certain fluctuations, but the fluctuation range is within the acceptable normal range and will not affect the processing quality.

[0080] or , it means that the difference is very close to the standard error threshold, which is more likely to cause failure and pose potential risks.

[0081] Finally, a stable group, a normal group, and a fluctuation group are established in the CNC machine tool database. The amplitude of the machine tool spindle and the guide rail offset are combined to divide each machine tool file in the CNC machine tool database:

[0082] For the number i If the machine tool and , it is classified into the stable group; if and , it is classified into the normal group; if and , it is classified into the fluctuation group.

[0083] Furthermore, after each part processing is completed, the health analysis module dynamically adjusts the grouping of subordinate machine tools based on the fluctuation value and the median absolute deviation method. The specific steps are as follows:

[0084] First, in the subsequent part processing process, the spindle amplitude and guide rail offset of the machine tools in each group are recorded according to the stable group, normal group and fluctuation group, and the spindle amplitude fluctuation value and guide rail offset fluctuation value of each machine tool in the group are calculated. The algorithm formula is:

[0085]

[0086] in, A j Representative number is j The spindle amplitude of the machine tool during this processing Spindle amplitude compared to the previous machining The fluctuation value of ΔB j Representative number is j The guide rail offset of the machine tool during this processing Guideway offset from the previous machining Fluctuation value of

[0087] Secondly, the status of the machine tools in the group is determined based on the fluctuation value. A j <0 and ΔB j <0, then the number will be j The machine tool is determined to be an upgraded machine tool; if A j ≥0 or ΔB j ≥0, the median absolute deviation method is used to determine whether the machine tool is a downgraded machine tool, including:

[0088] Collect all spindle amplitude fluctuation values ​​and guideway offset fluctuation values ​​less than 0 in the group, and calculate the median value respectively , ,in M A All the main axis amplitude fluctuation values ​​less than 0 in the group The median value of a is the number of spindle amplitude fluctuation values ​​less than 0, M B For all rail offset fluctuation values ​​less than 0 in the group The median value of b is the number of rail offset fluctuation values ​​less than 0;

[0089] Calculate absolute deviation , ,in For the group k A spindle amplitude fluctuation value less than 0 The absolute deviation of For the group k A rail offset fluctuation value less than 0 The absolute deviation of

[0090] calculate , ,in m A , m B The absolute deviation The median value of ,in Based on the median absolute deviation m A , m B The judgment threshold of k is the preset coefficient;

[0091] Determine each fluctuation value greater than or equal to 0 in the group. If or , then the number will be j The machine tools are judged as downgraded machine tools, among which, Corresponding to A j , ΔB j The absolute deviation.

[0092] It should be noted that the preset coefficient k The value of is 2 or 3, which can be adjusted according to the specific situation. k The value of is adjusted to balance the detection sensitivity and false alarm rate. k The value will increase false positives, and larger k The median absolute deviation method may miss some real outliers, but it can better reflect the typical fluctuation range of the data and reduce false positives.

[0093] Furthermore, the machine tools are grouped and adjusted based on the upgraded and downgraded machine tools. The specific steps are as follows:

[0094] For the stable group, when the CNC machine tool is an upgraded machine tool, the current CNC machine tool will not be adjusted. When the CNC machine tool is a downgraded machine tool, the current CNC machine tool file will be migrated to the normal group;

[0095] For the normal group, when the CNC machine tool is an upgraded machine tool, the current CNC machine tool is migrated to the stable group; when the CNC machine tool is a downgraded machine tool, the current CNC machine tool file is migrated to the fluctuating group;

[0096] For the fluctuation group, when the CNC machine tool is an upgraded machine tool, the current CNC machine tool is migrated to the normal group; when the CNC machine tool is a downgraded machine tool, the current machine tool is marked as an abnormal CNC machine tool in the CNC machine tool database.

[0097] It should be noted that, for CNC machine tools marked as abnormal, operators can obtain the abnormal CNC machine tool number by checking the CNC machine tool database and perform maintenance in a timely manner.

[0098] Embodiment 5:

[0099] Based on the above-mentioned embodiment 2, this embodiment 5 provides a further preferred implementation mode. The CNC machine tool management module in the present invention, based on the grouping result of the health analysis module, marks the CNC machine tools in each group after multiple re-groupings, and determines the maintenance strategy. The specific steps are as follows:

[0100] After the subordinate CNC machine tools have been processed for multiple times, the CNC machine tool spindle amplitude and guide rail offset in the CNC machine tool database are collected through the data acquisition module each time the part processing is completed. Based on the adjustment strategy of different CNC machine tool groups, the subordinate CNC machine tool files are adjusted in the CNC machine tool database after each part processing is completed;

[0101] Based on the grouping results after multiple regroupings of the health analysis module, the CNC machine tools in the stable group are marked green, the CNC machine tools in the normal group are marked blue, and the CNC machine tools in the fluctuating group are marked red, and the health maintenance strategy of the CNC machine tools is set.

[0102] It should be noted that after multiple regroupings, the grouping of CNC machine tools gradually becomes stable, and the number of times is generally set to 100 times. The number of times can also be adjusted according to actual use needs.

[0103] The specific content of the CNC machine tool health maintenance strategy is as follows: for CNC machine tools marked in green, the comprehensive maintenance cycle is set to 80 days; for CNC machine tools marked in blue, the comprehensive maintenance cycle is set to 50 days; for CNC machine tools marked in red, the comprehensive maintenance cycle is set to 15 days.

[0104] It should be noted that by grouping CNC machine tools based on stabilization, maintenance cycles are set for CNC machine tools. For equipment in the stable group, the error fluctuation is small and the state is stable. A longer maintenance cycle is set to avoid economic losses caused by frequent maintenance. For equipment in the normal group, there are slight error fluctuations but they are still within the normal range. A moderate maintenance cycle can detect and deal with potential problems in time. For equipment in the fluctuating group, due to the large error fluctuations of the equipment, there may be potential health problems, so frequent inspections and maintenance are required to avoid unexpected downtime.

[0105] Through the above embodiments, the purpose of the present invention is fully and effectively achieved. Those skilled in the art can understand that the present invention includes but is not limited to the contents described in the drawings and the above specific embodiments. Although the present invention has been described with respect to the most practical and preferred embodiments currently considered, it should be understood that the present invention is not limited to the disclosed embodiments, and any modification that does not deviate from the functional and structural principles of the present invention will be included in the scope of the claims.

Claims

1. A CNC machine tool health monitoring method based on multi-source data fusion, used for real-time monitoring and dynamic management of the health status of multiple CNC machine tools in an industrial production environment, characterized in that: The health monitoring method comprises at least the following steps when implemented: SS1. Collect multi-source data through sensors on all machine tools under its jurisdiction, generate analog signals, and convert the analog signals into digital signals, wherein the multi-source data includes spindle vibration data and guide rail offset data; SS2. Set the spindle amplitude standard error threshold and the guideway offset standard error threshold based on the historical operation data of the machine tools, and perform error judgment on the spindle vibration data and guideway offset data of all the subordinate machine tools according to the two standard error thresholds. By calculating the difference between the spindle amplitude of each machine tool and its standard error threshold, and the difference between the guideway offset and its standard error threshold, determine the initial health status of each machine tool, and screen out abnormal machine tools with any indicator greater than its standard error threshold, and normal machine tools with all indicators within the standard error threshold range; SS3. Based on the error judgment results, all normal machine tools under its jurisdiction are grouped into basic groups. In the basic grouping process, the mean and standard deviation of the two indicators, spindle amplitude difference and guideway offset difference, of all normal machine tools under its jurisdiction are calculated according to the principle of three times standard deviation. The spindle amplitude and guideway offset are divided into stable interval, normal interval and fluctuation interval respectively. Based on the interval distribution of the two indicators, all normal machine tools under its jurisdiction are divided into stable group, normal group and fluctuation group. SS4. In the subsequent parts processing process, the spindle vibration data and guideway offset data of the machine tools in each group are continuously collected and recorded, and the spindle amplitude fluctuation value and guideway offset fluctuation value of each machine tool are calculated. Based on the fluctuation values ​​of the two indicators and combined with the median absolute deviation method, the health status of the machine tools in each group is re-evaluated and the grouping is dynamically adjusted. When the fluctuation value of the machine tools in the stable group exceeds its stable threshold range, it is downgraded to the normal group or the fluctuation group. When the fluctuation value of the machine tools in the normal group exceeds or falls below the normal range, it is upgraded to the stable group or downgraded to the fluctuation group respectively. When the fluctuation value of the machine tools in the fluctuation group returns to the normal range, it is upgraded to the normal group and downgraded to the abnormal machine tool when it is below the normal range. SS5. After multiple dynamic adjustments to the grouping, differentiated maintenance strategies are formulated for the machine tools in each group according to the final grouping results. Longer maintenance cycles are set for the stable group machines, moderate maintenance cycles are set for the normal group machines, and short-term high-frequency maintenance cycles are set for the fluctuating group machines. Real-time reminders are given for detailed inspection and repair of abnormal machine tools.

2. A CNC machine tool health monitoring system based on multi-source data fusion, comprising a sensor module, a data acquisition module, an error determination module, a health analysis module, and a CNC machine tool management module, characterized in that: The sensor module includes at least a vibration sensor and a linear displacement sensor, which are used to record the spindle vibration data and guide rail offset data of all the machine tools under its jurisdiction and generate analog signals; The data acquisition module is communicatively connected to the sensor module, and is used to receive analog signals from the sensor module, and convert the analog signals into digital signals and transmit them to the error determination module and the health analysis module; The error judgment module is communicatively connected to the data acquisition module, sets a spindle amplitude standard error threshold and a guide rail offset standard error threshold based on the historical operation data of the machine tool, and performs error judgment on the spindle vibration data and guide rail offset data of the subordinate machine tools according to the two standard error thresholds, and judges the preliminary health status of each machine tool by calculating the difference between the spindle amplitude of each machine tool and its standard error threshold and the difference between the guide rail offset and its standard error threshold, and screens out abnormal machine tools with any indicator greater than its standard error threshold and normal machine tools with both indicators within their standard error threshold ranges; The health analysis module is communicatively connected to the data acquisition module and the error judgment module, and based on the spindle vibration data and guide rail offset data of each machine tool, combined with the standard error threshold, the mean and standard deviation of the two indicators, the spindle amplitude difference and the guide rail offset difference of all the normal machine tools under its jurisdiction are calculated respectively, and the principle of three times the standard deviation and the interval distribution of the two indicators are combined to perform basic grouping of all the machine tools under its jurisdiction and divide them into a stable group, a normal group and a fluctuation group, and the spindle vibration data and guide rail offset of the machine tools in each group in the subsequent processing process are statistically analyzed to determine the fluctuation value and adjust the machine tool grouping in combination with the median absolute deviation method, when the fluctuation value of the machine tool in the stable group exceeds its stable threshold range, it is downgraded to the normal group or the fluctuation group, when the fluctuation value of the machine tool in the normal group exceeds or is lower than the normal range, it is upgraded to the stable group or downgraded to the fluctuation group, and when the fluctuation value of the machine tool in the fluctuation group returns to the normal range, it is upgraded to the normal group and downgraded to an abnormal machine tool when it is lower than the normal range; The CNC machine tool management module is communicatively connected to the health analysis module. Based on the grouping results of the health analysis module, after multiple regroupings, each grouped machine tool is marked according to the final grouping results and a differentiated maintenance strategy is determined. A longer maintenance cycle is set for the stable group machine tools, a moderate maintenance cycle is set for the normal group machine tools, and a short-term and high-frequency maintenance cycle is set for the fluctuating group machine tools. Machine tools with abnormal markings are reminded in real time to conduct detailed inspections and repairs.

3. A CNC machine tool health monitoring system based on multi-source data fusion according to claim 2, characterized in that: When performing basic grouping and dynamic grouping adjustment, the health analysis module includes: Establish a CNC machine tool database, number all normal machine tools, and create files in the CNC machine tool database, including the spindle amplitude of the machine tool at that time. η Z and rail offset η D , based on the spindle amplitude η Z and rail offset η D The corresponding axis amplitude standard error threshold θ Z and rail offset standard error threshold θ D The difference between the two indicators, as well as the mean and standard deviation of the two difference indicators, combined with the principle of three times the standard deviation and the interval distribution of the two indicators, all the machine tools under its jurisdiction are basically grouped and divided into stable group, normal group and fluctuating group; Based on the basic grouping of subordinate machine tools, calculate the spindle amplitude of machine tools in different groups during subsequent part processing η Z and rail offset η D The fluctuation value of the machine tools in the stable group is used to adjust the machine tool grouping based on the fluctuation value and the median absolute deviation method after each part processing is completed. When the fluctuation value of the machine tools in the stable group exceeds the stable threshold range, they are downgraded to the normal group or the fluctuation group. When the fluctuation value of the machine tools in the normal group exceeds or falls below the normal range, they are upgraded to the stable group or downgraded to the fluctuation group respectively. When the fluctuation value of the machine tools in the fluctuation group returns to the normal range, they are upgraded to the normal group. When it is below the normal range, they are downgraded and marked as abnormal machine tools.

4. A CNC machine tool health monitoring system based on multi-source data fusion according to claim 3, characterized in that: When the health analysis module performs basic grouping on all the machine tools under its jurisdiction, the specific steps are as follows: First, based on the spindle vibration data and guide rail offset data of different machine tools, combined with the spindle amplitude standard error threshold and the guide rail offset standard error threshold, the spindle amplitude difference and guide rail offset difference of each machine tool are calculated respectively. The algorithm formula is: in, Representative number is i The amplitude of the machine tool spindle The standard error threshold of the axis amplitude θ Z The difference, Representative number is i Machine guide rail offset Standard deviation threshold from rail offset θ D The difference, i =1,2,…, n , n The number of all machine tools under its jurisdiction; Secondly, the mean and standard deviation of the spindle amplitude difference and the mean and standard deviation of the guide rail offset difference of all the subordinate machine tools are calculated respectively. The algorithm formula is: in, They represent the mean of the spindle amplitude difference and the guideway offset difference of all the machine tools under their jurisdiction. They represent the standard deviation of the spindle amplitude difference and the standard deviation of the guideway offset difference of all the machine tools under their jurisdiction respectively; Next, set the stable interval, normal interval, and fluctuation interval of the spindle amplitude and the stable interval, normal interval, and fluctuation interval of the guide rail offset respectively, and determine the interval distribution of the spindle amplitude and guide rail offset of each machine tool under its jurisdiction: For the number i Machine tools, when When , it means that the main axis amplitude is in a stable range; when When , it means that the main axis amplitude is in the normal range; when When , it means that the main axis amplitude is in the fluctuation range; when When , it means that the guide rail offset is in a stable range; when When , it means that the guide rail offset is in the normal range; when When , it means that the guide rail offset is in the fluctuation range; Finally, a stable group, a normal group, and a fluctuation group are established in the CNC machine tool database. The amplitude of the machine tool spindle and the guide rail offset are combined to divide each machine tool file in the CNC machine tool database: For the number i If the machine tool , it is classified into the stable group; if and , it is classified into the normal group; if , it is classified into the fluctuation group.

5. A CNC machine tool health monitoring system based on multi-source data fusion according to claim 4, characterized in that: After each part processing is completed, the health analysis module dynamically adjusts the grouping of subordinate machine tools based on the fluctuation value and the median absolute deviation method. The specific steps are as follows: First, in the subsequent part processing process, the spindle amplitude and guide rail offset of the machine tools in each group are recorded according to the stable group, normal group and fluctuation group, and the spindle amplitude fluctuation value and guide rail offset fluctuation value of each machine tool in the group are calculated. The algorithm formula is: in, ΔA j Representative number is j The spindle amplitude of the machine tool during this processing Spindle amplitude compared to the previous machining The fluctuation value of ΔB j Representative number is j The guide rail offset of the machine tool during this processing Guideway offset from the previous machining Fluctuation value of Secondly, the status of the machine tools in the group is determined based on the fluctuation value. ΔA j <0 and ΔB j <0, then the number will be j The machine tool is determined to be an upgraded machine tool; if ΔA j ≥0 or ΔB j ≥0, the median absolute deviation method is used to determine whether the machine tool is a downgraded machine tool, including: Collect all spindle amplitude fluctuation values ​​and guideway offset fluctuation values ​​less than 0 in the group, and calculate the median value respectively , ,in M A All the main axis amplitude fluctuation values ​​less than 0 in the group The median value of a is the number of spindle amplitude fluctuation values ​​less than 0, M B For all rail offset fluctuation values ​​less than 0 in the group The median value of b is the number of rail offset fluctuation values ​​less than 0; Calculate absolute deviation , ,in For the group k A spindle amplitude fluctuation value less than 0 The absolute deviation of For the group k The rail offset fluctuation value is less than 0 The absolute deviation of calculate , ,in m A , m B The absolute deviation The median value of ,in Based on the median absolute deviation m A , m B The judgment threshold of k is the preset coefficient; Determine each fluctuation value greater than or equal to 0 in the group. If or , then the number will be j The machine tools are judged as downgraded machine tools, among which, Corresponding to ΔA j , ΔB j The absolute deviation.

6. A CNC machine tool health monitoring system based on multi-source data fusion according to claim 5, characterized in that: Combine the upgraded and downgraded machine tools to adjust the machine tools in groups. The specific steps are as follows: For the stable group, if the machine tool in the group is an upgraded machine tool, it will not be adjusted. If it is a downgraded machine tool, its file will be migrated to the normal group; For the normal group, if the machine tool in the group is an upgraded machine tool, it will be moved to the stable group. If it is a downgraded machine tool, its file will be moved to the fluctuating group. For the fluctuation group, when the machine tool in the group is an upgraded machine tool, it will be migrated to the normal group. If it is a downgraded machine tool, it will be marked as an abnormal CNC machine tool in the CNC machine tool database.

7. The CNC machine tool health monitoring system based on multi-source data fusion according to claim 6 is characterized in that: The CNC machine tool management module, based on the grouping results of the health analysis module, marks the machine tools in each group and determines the maintenance strategy after multiple regroupings. The specific steps are as follows: Each time a part is processed, the data acquisition module collects the amplitude of the machine tool spindle and the guide rail offset in the CNC machine tool database. Based on the adjustment strategy of different machine tool groups, the subordinate machine tool files are adjusted in the CNC machine tool database after each part is processed. Based on the grouping results after multiple regroupings by the health analysis module, the machine tools in the stable group are marked green, the machine tools in the normal group are marked blue, and the machine tools in the fluctuating group are marked red, and the health maintenance strategy of the CNC machine tools is set.

8. The CNC machine tool health monitoring system based on multi-source data fusion according to claim 7 is characterized in that: The health maintenance strategy for CNC machine tools is as follows: for green-marked machine tools, a comprehensive maintenance cycle is set to 80 days; for blue-marked machine tools, a comprehensive maintenance cycle is set to 50 days; and for red-marked machine tools, a comprehensive maintenance cycle is set to 15 days.

9. The CNC machine tool health monitoring system based on multi-source data fusion according to claim 2 is characterized in that: The error determination module determines the spindle vibration data and guide rail offset of the subordinate machine tools based on the standard error threshold, and determines the abnormal machine tools. The specific steps are as follows: Based on the spindle amplitude standard error threshold of the current machine tool θ Z and rail offset standard error threshold θ D , the spindle amplitude of each machine tool under its jurisdiction η Z and rail offset η D Make a judgment: when η Z > θ Z or η D > θ D When it is on, it means that the relevant machine tool has an abnormality and needs to be marked for maintenance; when η Z ≤ θ Z and η D ≤ θ D , it means that there is no abnormality in the related machine tool.

10. A computer program product comprising computer instructions, characterized in that: The computer instructions are used to execute the CNC machine tool health monitoring method based on multi-source data fusion as described in claim 1.

Images