Method and system for predicting running state of metal processing machine tool

Abstract

The invention relates to the technical field of intelligent manufacturing, in particular to a metal processing machine tool operation state prediction method and system. The method comprises the following steps: acquiring design data of a metal processing machine tool, and performing geometric structure design feature extraction according to the design data of the metal processing machine tool so as to obtain geometric structure design data; constructing a metal processing machine tool model according to the geometric structure design data so as to obtain the metal processing machine tool model; carbon tool steel cutter feature extraction and hard alloy cutter feature extraction are carried out according to the structural data of the metal processing machine tool, so that carbon tool steel cutter data and hard alloy cutter data are obtained; and according to the carbon tool steel cutter data and the hard alloy cutter data, cutter abnormal operation state analysis is carried out, so that cutter abnormal operation state data are obtained. The machine tool fault early warning and maintenance efficiency is improved based on the intelligent manufacturing technology.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing technology, and in particular to a method and system for predicting the operating status of a metal processing machine tool. Background Art

[0002] Traditional methods for predicting the state of metal processing machine tools usually rely on static models or manual experience, lacking real-time performance and accuracy. These methods mostly rely on regular detection or manual inspections to diagnose machine tool faults, making it difficult to provide early warnings for potential faults or to intervene in time before a fault occurs, resulting in the inability to effectively prevent the occurrence of faults, affecting production efficiency and processing accuracy. Existing sensor data analysis methods usually rely only on a single type of sensor data, such as vibration, temperature or current signals, while ignoring the fusion and comprehensive analysis of multi-sensor data. This makes the existing system susceptible to factors such as sensor errors and signal noise when dealing with complex working conditions, resulting in poor accuracy and robustness of the prediction results. For metal processing machine tools with a variety of different operating modes and working condition changes, a single data source cannot fully reflect its true operating status and it is difficult to effectively identify various abnormal conditions. Summary of the invention

[0003] Based on this, it is necessary for the present invention to provide a method and system for predicting the operating status of a metal processing machine tool to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above object, a method for predicting the operating status of a metal processing machine tool comprises the following steps:

[0005] Step S1: acquiring metal processing machine tool design data, and extracting geometric structure design features according to the metal processing machine tool design data, thereby obtaining geometric structure design data; constructing a metal processing machine tool model according to the geometric structure design data, thereby obtaining a metal processing machine tool model;

[0006] Step S2: extracting features of carbon tool steel tools and carbide tools according to the metal processing machine tool structure data, thereby obtaining carbon tool steel tool data and carbide tool data; analyzing tool abnormal operation status according to the carbon tool steel tool data and carbide tool data, thereby obtaining tool abnormal operation status data;

[0007] Step S3: extracting servo motor features and stepper motor features according to the metal processing machine tool structure data, thereby obtaining servo motor data and stepper motor data; performing motor overload analysis according to the servo motor data and stepper motor data, thereby obtaining motor overload data;

[0008] Step S4: performing abnormal state fusion of the machining machine tool according to the motor overload data and the abnormal operation state data of the tool, thereby obtaining abnormal state data of the machining machine tool; performing cutting simulation according to the metal machining machine tool model, thereby obtaining cutting simulation data of the metal machining machine tool;

[0009] Step S5: Perform abnormal state prediction on the metal processing machine tool cutting simulation data according to the abnormal state data of the processing machine tool, so as to obtain the abnormal state prediction data of the metal processing machine tool, and upload it to the metal processing machine tool management platform to execute the abnormal state prediction task.

[0010] The present invention helps to establish an accurate machine tool model by acquiring machine tool design data and extracting geometric structure features, which provides a basis for subsequent analysis. The construction of the model can fully reflect the geometric characteristics of the machine tool, ensuring that the subsequent state prediction is more accurate and comprehensive. Then, by extracting the features of carbon tool steel and carbide tools and analyzing their abnormal operating states, the performance changes of the tools can be monitored in real time, tool wear or failure can be identified in advance, and processing errors or shutdowns caused by tool abnormalities can be avoided. In addition, the feature extraction and overload analysis of servo motors and stepper motors can help detect the load state of the motor, identify motor overload or performance degradation problems early, and thus avoid the impact of motor failure on the entire processing process. Further, by integrating motor overload and tool abnormal state data, the abnormal state of the machine tool is comprehensively analyzed, and potential problems of the machine tool can be identified more comprehensively and accurately, and early warning information can be provided in time. Cutting simulation combined with the metal processing machine tool model can simulate the operating state of the machine tool under different working conditions, providing more objective data support for abnormal prediction. Ultimately, by uploading abnormal status prediction data to the management platform in real time, remote monitoring and management can be achieved, so that machine tool failures can be warned in advance and intervened in time, thereby improving production efficiency, reducing failure rates and ensuring processing accuracy.

[0011] Optionally, step S1 specifically includes:

[0012] Step S11: acquiring metal processing machine tool design data, and performing geometric structure design feature extraction according to the metal processing machine tool design data, thereby obtaining geometric structure design data;

[0013] Step S12: extracting the geometric structure features of the bed and the spindle structure features according to the geometric structure design data, thereby obtaining the geometric structure data of the bed and the spindle structure data;

[0014] Step S13: assembling the machine tool model with the bed geometry data according to the spindle structure data, thereby obtaining the machine tool model;

[0015] Step S14: performing virtual simulation on the machine tool model to obtain virtual simulation data of the machine tool model;

[0016] Step S15: Optimizing the collision of metal processing parts on the machine tool model according to the virtual simulation data of the machine tool model, thereby obtaining a metal processing machine tool model.

[0017] The present invention solves the shortcomings of traditional machine tool state prediction methods in terms of real-time performance, accuracy and multi-data fusion by comprehensively analyzing and modeling the design data of metal processing machine tools. By extracting the geometric structure features in the machine tool design data, not only can the basic design information of the machine tool be accurately obtained, but also a solid foundation can be provided for subsequent model construction and analysis. Combined with the feature extraction of the bed geometry and the spindle structure, the key component characteristics of the machine tool are further refined, thereby ensuring that the model is more in line with the actual structure of the machine tool. This multi-level feature extraction method helps to gain an in-depth understanding of the overall design of the machine tool and the structural features of each component, thereby providing a detailed basis for the subsequent assembly of the machine tool model. By assembling the machine tool model with the spindle structure data and the bed geometry data, the reasonable connection between the various components of the machine tool is ensured, thereby improving the accuracy and stability of the model. The subsequent virtual simulation step can obtain more realistic machine tool operation data by simulating the actual operating environment, helping to further verify the rationality and performance of the machine tool model. By optimizing the collision of metal processing parts in the virtual simulation data, the collision problem can be effectively identified and avoided, thereby improving the safety and stability of the machine tool in actual operation. Ultimately, the obtained metal processing machine tool model was not only verified in the virtual environment, but also provided a more reliable prediction basis in actual production, reducing the occurrence of potential failures and improving the operating efficiency and processing accuracy of the machine tool.

[0018] Optionally, step S15 is specifically:

[0019] Step S151: performing collision area identification according to the virtual simulation data of the machine tool model, thereby obtaining collision area data of the machine tool;

[0020] Step S152: extracting metal fixture collision features from the machine tool collision area data, thereby obtaining metal fixture collision data;

[0021] Step S153: adjusting the safety clearance according to the metal fixture collision data, thereby obtaining the metal machine tool safety clearance data;

[0022] Step S154: Optimizing the machine tool model according to the metal machine tool safety clearance data, thereby obtaining a metal processing machine tool model.

[0023] The present invention can effectively identify the collision area of ​​the machine tool during the virtual simulation process through collision area identification, so as to discover potential collision risks in advance and avoid mechanical damage or safety accidents in actual operation. This identification process not only improves the safety of machine tool operation, but also provides important data support for subsequent collision feature extraction. By extracting the collision features of the metal fixture, the specific cause of the collision can be accurately located, the key issues affecting the operation of the machine tool can be identified, and a scientific basis can be provided for further safety clearance adjustment. This adjustment step helps to optimize the relative position between the various components of the machine tool, ensure that there will be no too small safety clearance during the processing process, thereby effectively avoiding the collision between the fixture and the machine tool components, and ensuring the stability and safety of the processing. Finally, by optimizing the machine tool model, the structure and operating parameters of the machine tool can be adjusted, so that the machine tool is more in line with the expected processing requirements in actual work, thereby improving the processing accuracy, work efficiency and reliability of the machine tool. Therefore, the present invention improves the overall performance of the metal processing machine tool through comprehensive collision identification, feature extraction, safety clearance adjustment and model optimization, ensures the safe and stable operation of the machine tool under various working conditions, greatly improves the production efficiency, and reduces the probability of failure.

[0024] Optionally, step S2 specifically includes:

[0025] Step S21: extracting features of carbon tool steel tools and hard alloy tool tools according to the metal processing machine tool structure data, thereby obtaining carbon tool steel tool data and hard alloy tool data;

[0026] Step S22: performing cutting force anomaly analysis on the carbon tool steel tool data, thereby obtaining cutting force anomaly data of the carbon tool steel tool;

[0027] Step S23: performing noise anomaly analysis on the cemented carbide tool data, thereby obtaining cemented carbide tool noise anomaly data;

[0028] Step S24: Integrate the abnormal operation status of the tool according to the abnormal cutting force data of the carbon tool steel tool and the abnormal noise data of the cemented carbide tool, so as to obtain the abnormal operation status data of the tool.

[0029] The present invention improves the operation accuracy and safety of the metal processing machine tool by performing comprehensive abnormal analysis and state integration on the tool of the metal processing machine tool. By extracting the features of carbon tool steel tools and cemented carbide tools, the data of various types of tools can be accurately obtained, laying the foundation for subsequent abnormal analysis. The abnormal analysis of the cutting force of carbon tool steel tools can timely identify abnormal conditions such as excessive or insufficient cutting force, avoid the reduction of processing accuracy or damage to the tool due to abnormal tool state, thereby ensuring the long-term stability of the processing quality and equipment. At the same time, the abnormal noise analysis of cemented carbide tools can detect the abnormal noise caused by tool wear, defects or poor working conditions, and further help to find potential sources of faults. These abnormal data can be integrated to fully reflect the abnormal operation state of the tool, thereby providing a more comprehensive assessment of the health status of the tool. Through this comprehensive abnormal operation state detection method, potential problems of the tool can be discovered in time, intervention can be carried out in advance, failures can be avoided, unnecessary downtime can be reduced, and production efficiency and processing accuracy can be improved.

[0030] Optionally, step S22 is specifically:

[0031] Step S221: performing cutting simulation on the carbon tool steel tool data, thereby obtaining the carbon tool steel tool cutting data;

[0032] Step S222: performing Fourier transform on the carbon tool steel tool cutting data to obtain the carbon tool steel tool cutting frequency domain data;

[0033] Step S223: performing noise frequency statistics according to the cutting frequency domain data of the carbon tool steel tool, so as to obtain high-frequency noise data and low-frequency noise data;

[0034] Step S224: performing fluctuation statistics according to the low-frequency noise data, thereby obtaining low-frequency noise high fluctuation data;

[0035] Step S225: performing an aggravation evaluation on the low-frequency noise high fluctuation data to obtain the low-frequency noise high fluctuation aggravation data;

[0036] Step S226: Perform cutting force anomaly integration of carbon tool steel cutting tools according to the low-frequency noise high fluctuation aggravation data and the high-frequency noise data, so as to obtain cutting force anomaly data of carbon tool steel cutting tools.

[0037] The present invention can deeply understand the cutting process of the tool by performing cutting simulation on the data of carbon tool steel tools, and provide accurate basic data for subsequent data analysis. Fourier transform is used to convert cutting data into frequency domain data to help extract potential frequency components, which is crucial for identifying small changes in tool operation. Further, by performing noise frequency statistics on the cutting frequency domain data, high-frequency and low-frequency noise data can be accurately divided, which helps to identify abnormal conditions caused by tool wear or poor contact. The fluctuation statistical analysis of low-frequency noise high-fluctuation data can reveal potential problems of the tool in the cutting process, such as fatigue or damage of the tool material. And by evaluating the aggravation of low-frequency noise high-fluctuation data, the severity of tool anomalies can be quantified, providing a basis for subsequent intervention. Finally, by integrating low-frequency noise high-fluctuation data with high-frequency noise data, the abnormal state of the cutting force of carbon tool steel tools can be comprehensively evaluated to ensure the accuracy and stability of the tool in operation. The monitoring accuracy and reliability of the tool state are improved, anomalies can be detected and intervened in time, the occurrence of failures can be avoided, and production efficiency and processing quality are improved.

[0038] Optionally, step S23 is specifically:

[0039] Step S231: extracting noise sensing features from cemented carbide tool data to obtain noise sensing data;

[0040] Step S232: performing filtering and noise reduction processing according to the noise sensing data, thereby obtaining noise sensing filtering and noise reduction data;

[0041] Step S233: performing power spectrum drawing on the noise sensor filtering noise reduction data, thereby obtaining the noise sensor filtering noise reduction power spectrum;

[0042] Step S234: performing density calculation according to the noise sensing filter noise reduction power spectrum, thereby obtaining the noise sensing filter noise reduction power spectrum density data;

[0043] Step S235: Calculating the spectral centroid of the noise sensor filter noise reduction power spectrum according to the noise sensor filter noise reduction power spectrum density data, thereby obtaining spectral centroid data;

[0044] Step S236: performing offset identification on the spectrum center of gravity data, thereby obtaining spectrum center of gravity offset data;

[0045] Step S237: performing tool abnormality time identification on the cemented carbide tool data according to the spectrum center of gravity offset data, thereby obtaining the cemented carbide tool noise abnormality data.

[0046] The present invention effectively improves the precision and accuracy of cemented carbide tool state monitoring through multi-step noise sensor data processing. By extracting noise sensor features from cemented carbide tool data, the noise characteristics of the tool during operation can be obtained, providing important data support for subsequent analysis. The filtering and noise reduction processing of noise sensor data effectively removes noise interference, improves the accuracy and clarity of the data, and makes abnormal signals more prominent. The drawing of the power spectrum further reveals the frequency components in the noise signal and helps identify potential abnormal sources. Density calculation is performed on the noise sensor filtering and noise reduction data to further quantitatively analyze the characteristics of the noise signal and help establish a more accurate model. The spectrum center of gravity calculation provides a more detailed noise analysis tool by analyzing the frequency distribution center of the noise signal, which can be used to identify changes in the tool operation state. When the spectrum center of gravity offset occurs, the system can promptly identify the abnormal phenomenon and indicate a tool problem or fault. By identifying the tool abnormal time on the spectrum center of gravity offset data, the abnormal state of the cemented carbide tool can be accurately identified, and a timely signal can be provided for fault warning.

[0047] Optionally, step S3 specifically includes:

[0048] Step S31: extracting servo motor features and stepper motor features according to the metal processing machine tool structure data, thereby obtaining servo motor data and stepper motor data;

[0049] Step S32: performing transient torque analysis on the servo motor data to obtain transient torque data of the servo motor;

[0050] Step S33: performing winding insulation abnormality analysis on the stepper motor data, thereby obtaining the stepper motor winding insulation abnormality data;

[0051] Step S34: performing motor overload integration according to the servo motor transient torque data and the stepper motor winding insulation abnormality data, thereby obtaining the motor overload data.

[0052] The present invention can obtain independent data of the servo motor and stepper motor respectively by extracting features from the data, thus providing a rich information basis for subsequent analysis. The transient torque analysis of the servo motor helps to reveal the torque variation of the motor under different working conditions, timely identify the abnormal fluctuation of the motor load, and improve the accuracy of motor performance monitoring. The abnormal insulation analysis of the stepper motor winding can detect the potential faults of the motor winding, especially the deterioration of the insulating material, which is crucial for the long-term operation stability of the motor. By integrating the transient torque data of the servo motor with the abnormal insulation data of the stepper motor winding, it is possible to fully identify whether the motor is in an overload state, provide early warning signals in time, and avoid potential damage caused by motor overload. Through this comprehensive analysis of multi-dimensional data, the present invention can more accurately predict motor failures, improve the reliability and production efficiency of metal processing machine tools, reduce the risk of failures, and ensure the stability of processing accuracy.

[0053] Optionally, step S32 is specifically:

[0054] Step S321: collecting torque signals of the servo motor data to obtain torque signal data;

[0055] Step S322: Calculate the instantaneous change rate according to the torque signal data, so as to obtain the instantaneous change rate of the torque signal;

[0056] Step S323: performing change rate statistics according to the instantaneous change rate of the torque signal, so as to obtain high instantaneous change rate torque signal data;

[0057] Step S324: identifying a mutation point of the torque signal data according to the high instantaneous change rate torque signal data, thereby obtaining torque signal mutation point data;

[0058] Step S325: Visualize the transient fluctuation of the servo motor according to the torque signal mutation point data, so as to obtain the transient torque data of the servo motor.

[0059] The present invention can obtain detailed torque data generated by the servo motor during operation through the torque signal acquisition step, providing a reliable basis for subsequent fault diagnosis and performance evaluation. Then, through the calculation of the instantaneous rate of change, the instantaneous fluctuation of the torque signal can be quickly captured, providing a more sensitive monitoring means for identifying potential abnormal changes. The statistics of the torque signal data with high instantaneous rate of change are helpful to screen out those periods with more significant abnormal fluctuations, providing key data for further analysis. By identifying the mutation points of these high instantaneous rate of change data, the mutation points in the torque signal can be accurately located, revealing the instability or fault signs of the motor during operation, so as to predict the abnormal conditions of the motor in advance. Finally, through the visualization of the transient torque data of the servo motor, the trend and pattern of the torque fluctuation can be intuitively displayed, providing clearer information for fault analysis and diagnosis. These steps effectively combine the dynamic characteristics of the torque signal, provide more accurate and real-time data support for the state monitoring of the servo motor, improve the fault prediction ability of the metal processing machine tool, and ensure the efficient and safe operation of the machine tool.

[0060] Optionally, step S33 is specifically:

[0061] Step S331: collecting insulation resistance data of the stepper motor, thereby obtaining insulation resistance data of the stepper motor;

[0062] Step S332: performing temperature statistics according to the stepper motor insulation resistance data, thereby obtaining high-temperature stepper motor insulation resistance data;

[0063] Step S333: constructing an insulation resistance-temperature relationship model according to the stepper motor insulation resistance data and the high-temperature stepper motor insulation resistance data, thereby obtaining an insulation resistance-temperature relationship model;

[0064] Step S334: performing insulation material aging simulation on the insulation resistance-temperature relationship model to obtain insulation material aging data;

[0065] Step S335: obtaining the stepper motor operating voltage data;

[0066] Step S336: performing breakdown voltage statistics according to the insulation material aging data, thereby obtaining insulation material aging breakdown voltage data;

[0067] Step S337: Perform voltage insulation failure determination based on the stepper motor operating voltage data and the insulation material aging breakdown voltage data, thereby obtaining the stepper motor winding insulation abnormality data.

[0068] The present invention can accurately obtain the insulation status of the motor by collecting the insulation resistance of the stepper motor, and provide basic data for subsequent fault diagnosis and health monitoring. The temperature statistics based on the insulation resistance data of the stepper motor further reveal the change trend of the insulation resistance of the motor under high temperature conditions, which can help identify the risk of insulation failure caused by temperature increase. By constructing an insulation resistance temperature relationship model, the insulation performance of the motor at different temperatures is comprehensively considered, and the change of insulation resistance under different environmental conditions can be more accurately predicted, providing a scientific basis for fault warning. The insulation material aging simulation step helps predict the aging process of the motor insulation material by considering the influence of time and temperature on the motor insulation material, and provides forward-looking guidance for equipment maintenance. After obtaining the working voltage data of the stepper motor, the breakdown voltage statistics can be further combined with the insulation material aging data to help determine whether the motor is close to the critical point of insulation failure. Finally, by combining the working voltage with the insulation material aging breakdown voltage data, the motor winding insulation abnormality can be determined in real time, thereby realizing early identification and warning of faults. The real-time and accuracy of the motor fault prediction of the metal processing machine tool are effectively improved, the risk of fault occurrence is reduced, and the stable operation and processing accuracy of the equipment are guaranteed.

[0069] Optionally, the present specification further provides a metal processing machine tool operating state prediction system, which is used to execute the metal processing machine tool operating state prediction method as described above, and the metal processing machine tool operating state prediction system includes:

[0070] A metal processing machine tool model building module is used to obtain metal processing machine tool design data, and extract geometric structure design features based on the metal processing machine tool design data, so as to obtain geometric structure design data; and build a metal processing machine tool model based on the geometric structure design data, so as to obtain a metal processing machine tool model;

[0071] The tool abnormal operation state analysis module is used to extract the features of carbon tool steel tools and carbide tools according to the structure data of metal processing machine tools, so as to obtain the data of carbon tool steel tools and carbide tools; analyze the abnormal operation state of tools according to the data of carbon tool steel tools and carbide tools, so as to obtain the data of abnormal operation state of tools;

[0072] The motor overload analysis module is used to extract servo motor features and stepper motor features according to the metal processing machine tool structure data, so as to obtain servo motor data and stepper motor data; and to perform motor overload analysis according to the servo motor data and stepper motor data, so as to obtain motor overload data;

[0073] The cutting simulation module is used to fuse the abnormal state of the machining machine tool according to the motor overload data and the abnormal operation state data of the tool, so as to obtain the abnormal state data of the machining machine tool; and to perform cutting simulation according to the metal machining machine tool model, so as to obtain the cutting simulation data of the metal machining machine tool;

[0074] The abnormal state prediction module of metal processing machine tools is used to predict the abnormal state of metal processing machine tool cutting simulation data according to the abnormal state data of the processing machine tools, so as to obtain the abnormal state prediction data of the metal processing machine tools and upload it to the metal processing machine tool management platform to perform the abnormal state prediction task.

[0075] The metal processing machine tool operation status prediction system of the present invention can implement any metal processing machine tool operation status prediction method of the present invention, and is used to combine the operation and signal transmission medium between various modules to complete the metal processing machine tool operation status prediction method. The internal modules of the system cooperate with each other, thereby improving the machine tool fault warning and maintenance efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

[0077] Figure 1 It is a schematic diagram of the steps of the method for predicting the operating state of a metal processing machine tool according to the present invention;

[0078] Figure 2 Detailed step flow diagram of step S1 in the present invention;

[0079] Figure 3 Detailed step flow diagram of step S15 in the present invention;

[0080] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0081] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0082] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0083] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0084] To achieve this, please refer to Figures 1 to 3 The present invention provides a method for predicting the operating state of a metal processing machine tool, the method comprising the following steps:

[0085] Step S1: acquiring metal processing machine tool design data, and extracting geometric structure design features according to the metal processing machine tool design data, thereby obtaining geometric structure design data; constructing a metal processing machine tool model according to the geometric structure design data, thereby obtaining a metal processing machine tool model;

[0086] In this embodiment, design data is extracted from the design database of the metal processing machine tool, involving information such as the structural parameters, component specifications, and motion modes of the machine tool. These design data are usually presented in the form of CAD drawings, 3D models, or other design documents. Using special CAD software (such as AutoCAD or SolidWorks), the design data is parsed to extract the geometric features of each component of the machine tool, including the geometric shapes and dimensions of the machine tool bed, spindle, tool holder, workbench, etc. For each component, its length, width, height, aperture, axial offset and other parameters are extracted. When extracting geometric structure design features, it is necessary to ensure that the accuracy of all parameters meets the design requirements. For example, for the diameter of the spindle and the width of the workbench, it is necessary to accurately extract from the design drawing and use appropriate dimensioning tools to obtain the values. Then, based on these geometric features, a 3D model of the metal processing machine tool is generated using a computer-aided design (CAD) tool to ensure that the relative position, size, and connection method of each component meet the design requirements. This model can be used for subsequent machine tool assembly and simulation.

[0087] Step S2: extracting features of carbon tool steel tools and carbide tools according to the metal processing machine tool structure data, thereby obtaining carbon tool steel tool data and carbide tool data; analyzing tool abnormal operation status according to the carbon tool steel tool data and carbide tool data, thereby obtaining tool abnormal operation status data;

[0088] In this embodiment, design features related to the tool are extracted based on the structural data of the metal processing machine tool. These tool features include tool material, shape, size, surface coating, blade angle, tool life, etc. For carbon tool steel tools and carbide tools, their relevant geometric properties are extracted respectively. The specific operation is to obtain the specific parameters of each tool, such as the outer diameter, blade length, number of blades, etc. of the tool by reading the CAD design file of the tool or the standardized tool data table. In the analysis of the abnormal operation status of the tool, it is necessary to monitor the use of the tool in real time. By connecting with the data interface of the metal processing machine tool, the operation data of the tool, including cutting speed, feed rate, cutting force, etc., is obtained. Hardware devices such as vibration sensors, force sensors or temperature sensors are used to monitor the status of the tool in real time and compare it with the preset abnormal state threshold. For example, when the tool temperature exceeds 100°C or the vibration amplitude exceeds 5mm / s, the tool is considered to be in an abnormal state. If an abnormal situation occurs, the abnormal operation status data of the tool is recorded and generated immediately.

[0089] Step S3: extracting servo motor features and stepper motor features according to the metal processing machine tool structure data, thereby obtaining servo motor data and stepper motor data; performing motor overload analysis according to the servo motor data and stepper motor data, thereby obtaining motor overload data;

[0090] In this embodiment, the relevant features of the motor are extracted from the structural data of the metal processing machine tool, including the model, rated power, voltage, speed, torque, etc. of the servo motor and stepper motor. The relevant parameters are extracted by viewing the machine tool design document or the technical specification of the motor. For example, the rated power of the servo motor is 1.5kW and the maximum speed is 3000rpm. Then, the operating data of the motor, including current, voltage, speed and temperature, etc., are obtained in real time through the control system of the motor. The machine overload analysis mainly relies on the comparison of the real-time operating data of the motor with the preset overload threshold. The overload threshold is set according to the rated power of the motor. If the current of the motor exceeds 120% of its rated current, the motor is considered to be in an overloaded state. The current sensor and voltage sensor are used to monitor the motor state in real time. If an overload occurs, the system will trigger an alarm mechanism and record the motor overload data.

[0091] Step S4: performing abnormal state fusion of the machining machine tool according to the motor overload data and the abnormal operation state data of the tool, thereby obtaining abnormal state data of the machining machine tool; performing cutting simulation according to the metal machining machine tool model, thereby obtaining cutting simulation data of the metal machining machine tool;

[0092] In this embodiment, the motor overload data is combined with the tool abnormal operation state data, and a weighted fusion algorithm is used to fuse the abnormal state of the processing machine tool. The specific operation is to integrate the motor overload condition and the tool abnormal state data through the data fusion platform, and assign different weights to determine the overall abnormal state of the processing machine tool. For example, if the weight of the motor overload state is 0.6 and the weight of the tool abnormal state is 0.4, the final machine tool abnormal state data is obtained by the weighted average method. When using a metal processing machine tool model for cutting simulation, it is first necessary to input the geometric model of the machine tool and the operating parameters of the tool, and select appropriate cutting speed, feed rate, cutting depth and other parameters. The cutting process is simulated by cutting simulation software (such as DEFORM or ANSYS), and the cutting force, heat distribution, stress strain and other data of the machine tool in actual processing are calculated. These data will be used to evaluate the working status of the machine tool, especially tool wear, workpiece deformation and other conditions.

[0093] Step S5: Perform abnormal state prediction on the metal processing machine tool cutting simulation data according to the abnormal state data of the processing machine tool, so as to obtain the abnormal state prediction data of the metal processing machine tool, and upload it to the metal processing machine tool management platform to execute the abnormal state prediction task.

[0094] In this embodiment, it is necessary to integrate the abnormal state data of the processing machine tool with the cutting simulation data to form a training set for the training of the machine learning model. By collecting historical operation data, including motor overload, abnormal operation state of the tool and cutting simulation data, a database for predicting abnormal state is constructed. Select appropriate machine learning algorithms, such as support vector machine (SVM) or decision tree, to train the model so that it can learn the abnormal state patterns that occur under different working conditions. In the prediction process, a clear threshold must be set to ensure the accuracy of the prediction. For example, when the tool wear exceeds a certain value (such as 80%), the system determines it as an abnormal state and issues a warning in advance based on the working condition data. At the same time, for the motor overload state, the overload threshold can be set to 120% of the rated current. Once this threshold is exceeded, an alarm is triggered immediately. By analyzing the abnormal occurrence time point, abnormal type, abnormal severity and other information output by the prediction model, the risks faced by the machine tool are further determined. After the prediction, the system uploads the results to the metal processing machine tool management platform by encryption. In this process, all data need to be encrypted to ensure the security and integrity during the transmission process to prevent data from being tampered with or leaked. The uploaded data includes the predicted abnormal status, the corresponding predicted time point and the recommended maintenance measures, so that the management platform can perform subsequent monitoring, maintenance and decision-making tasks.

[0095] Optionally, step S1 specifically includes:

[0096] Step S11: acquiring metal processing machine tool design data, and performing geometric structure design feature extraction according to the metal processing machine tool design data, thereby obtaining geometric structure design data;

[0097] In the present embodiment, the complete design data of the metal processing machine tool is obtained from the design database or CAD system, including but not limited to the geometric parameters and size information of each part such as the bed, spindle, and tool rest. Then, the geometric design features of the machine tool are extracted using CAD software (such as AutoCAD, SolidWorks, etc.). These features include the key geometric parameters such as the length, width, height, thickness of the bed and the diameter, length, and installation position of the spindle. For each design feature, a strict dimensional tolerance standard is set, such as a bed thickness tolerance of ± 0.5 mm and a spindle diameter tolerance of ± 0.1 mm. Through these data, the geometric design data of the metal processing machine tool is constructed, and a basis is provided for subsequent machine tool model construction.

[0098] Step S12: extracting the geometric structure features of the bed and the spindle structure features according to the geometric structure design data, thereby obtaining the geometric structure data of the bed and the spindle structure data;

[0099] In this embodiment, the geometric structure feature extraction of the bed mainly includes the support structure, thickness distribution, reinforcement position, etc. of the bed, and these data are specifically analyzed by using 3D modeling software (such as SolidWorks or CATIA). The bearing capacity, stability and other parameters of the bed also need to be calculated by finite element analysis (FEA) to ensure that they meet the load requirements when the machine tool is working. For the spindle, its structural data is mainly extracted, including the size of the spindle, the position of the mounting hole, the precision requirements of the spindle hole, etc. All these parameters should be extracted according to standardized design requirements, such as the roundness tolerance requirement of the spindle hole ≤0.05mm. Through this step, the geometric structure data of the bed and the spindle are obtained to provide data support for subsequent machine tool assembly and virtual simulation.

[0100] Step S13: assembling the machine tool model with the bed geometry data according to the spindle structure data, thereby obtaining the machine tool model;

[0101] In this embodiment, the bed and spindle structure data are imported into a three-dimensional modeling platform (such as CATIA, Siemens NX, etc.). According to the design requirements, the positional relationship and assembly method of each component are adjusted to ensure that the interface accuracy requirements between the bed and the spindle meet the design specifications. For example, the connection aperture error between the spindle and the bed cannot exceed ±0.02mm, and the position deviation requirement of the mounting hole is ≤0.1mm. During assembly, the correct matching of each part is ensured by using constraint relationships, and the contact surface between the spindle and the bed must meet certain surface roughness requirements, usually Ra≤0.8μm. After completing the assembly, a complete metal processing machine tool model is obtained, which provides accurate geometric structure information for subsequent virtual simulation.

[0102] Step S14: performing virtual simulation on the machine tool model to obtain virtual simulation data of the machine tool model;

[0103] In this embodiment, virtual simulation is performed by computer-aided engineering (CAE) software (such as ANSYS, Abaqus, etc.) to simulate the mechanical behavior encountered by the machine tool during the processing, such as cutting force, vibration, thermal deformation, etc. During the simulation process, different working conditions need to be set, such as cutting speed, feed rate, etc. By performing load analysis on the machine tool model, the mechanical properties such as stress and strain in actual processing are simulated. For example, when performing cutting simulation, the feed rate of the tool can be set to 0.5mm / s and the cutting depth is 3mm. These values ​​are determined according to the actual processing capacity of the machine tool and the requirements of the processing task. The simulation results include the stress distribution, displacement deformation and temperature rise of each component of the machine tool, providing data support for subsequent optimization.

[0104] Step S15: Optimizing the collision of metal processing parts on the machine tool model according to the virtual simulation data of the machine tool model, thereby obtaining a metal processing machine tool model.

[0105] In this embodiment, simulation software is used to perform collision analysis on various parts of the machine tool, especially the relative movement of parts such as the spindle, tool holder, and bed. By setting reasonable collision judgment criteria, such as the minimum safe distance between parts should not be less than 0.5mm, it is ensured that the parts will not interfere or collide during the processing. According to the collision detection results, the machine tool model is adjusted to optimize the interference problems in the design. For example, if the spindle collides with the bed, it is necessary to adjust the position of the spindle or change the design angle of the tool holder. The optimization process should continue until the interference between all parts is eliminated and meets the design requirements. Finally, an optimized metal processing machine tool model is obtained, which not only meets the structural requirements, but also has high operating stability and is suitable for actual processing applications.

[0106] Optionally, step S15 is specifically:

[0107] Step S151: performing collision area identification according to the virtual simulation data of the machine tool model, thereby obtaining collision area data of the machine tool;

[0108] In this embodiment, the completed virtual simulation data of the machine tool model is imported, and the collision area of ​​the machine tool is identified using three-dimensional collision detection software (such as ANSYS, Siemens NX or CATIA). By analyzing the relative position, motion trajectory and processing process between the various components of the machine tool, the potential collision area is identified. During simulation, it is necessary to set the working space and motion range of each machine tool component to ensure that it meets the design specifications. For example, the maximum movement range of the tool holder during the processing is set to 200mm, and the maximum movable range of the bed is set to 150mm. The standard for collision detection is that when the distance between the two components is less than 0.5mm, it is determined to be a collision area. Through this operation, all collision areas are identified and collision area data is generated. The obtained collision area data includes information such as the coordinates of the collision point, the collision surface and the collision intensity.

[0109] Step S152: extracting metal fixture collision features from the machine tool collision area data, thereby obtaining metal fixture collision data;

[0110] In this embodiment, by analyzing the collision area data, especially the detailed annotation of the collision area related to the metal fixture, the characteristics of the collision between the metal fixture and other parts of the machine tool (such as tools, workpieces, spindles, etc.) are extracted. This process is performed using 3D modeling software combined with a collision recognition module. The specific operation is to set the maximum angle range of contact between the fixture and the workpiece. For example, when the contact angle between the fixture and the workpiece exceeds 30°, it is considered to have a collision. Further, based on the collision characteristics, the shape, size, installation position and other parameters of the fixture are extracted, and the minimum safe distance from each part of the machine tool is calculated. For example, the minimum distance between the fixture and the bed is 2 mm, and the maximum collision angle of the tool is 45°. Through this step, the metal fixture collision data is obtained, including the contact surface of the fixture, the collision intensity and the frequency of its occurrence.

[0111] Step S153: adjusting the safety clearance according to the metal fixture collision data, thereby obtaining the metal machine tool safety clearance data;

[0112] In this embodiment, the safety clearance is adjusted according to the metal fixture collision data. An optimization algorithm (such as a genetic algorithm or a particle swarm optimization algorithm) is used to automatically adjust the safety clearance between the machine tool and the metal fixture. The minimum safety clearance between the fixture and the workpiece is set to 2mm, and the minimum distance between the fixture and the bed is required to be 1.5mm. The collision data is analyzed to ensure that all fixture collision areas are within the set safety clearance range. For example, when a fixture collides with a tool, the position of the fixture is adjusted according to the simulation data to ensure that the minimum clearance between it and the tool is greater than 2mm. At the same time, during the adjustment process, the shape, size and working range of the fixture need to be considered to avoid any contact between the fixture and the machine tool components. Through these operations, optimized metal machine tool safety clearance data is obtained.

[0113] Step S154: Optimizing the machine tool model according to the metal machine tool safety clearance data, thereby obtaining a metal processing machine tool model.

[0114] In this embodiment, a three-dimensional modeling software (such as AutoCAD, CATIA or Siemens NX) is used to adjust the position of the machine tool components. According to the safety clearance data, the structure of each component of the machine tool is redesigned to ensure that there is enough clearance between each component to avoid any interference. The specific operation is: according to the safety clearance calculated above, adjust the spacing between the fixture and the spindle to ensure that the distance between the fixture and the tool is 2mm, the distance between the fixture and the bed is 1.5mm, and the contact angle between the tool and the workpiece does not exceed 30°. During the optimization process, new virtual simulations are also required to ensure that the optimized design does not collide and meets the load requirements during the processing. After these operations, a metal processing machine tool model that meets the safety clearance requirements is obtained, which has high stability and processing accuracy.

[0115] Optionally, step S2 specifically includes:

[0116] Step S21: extracting features of carbon tool steel tools and hard alloy tool tools according to the metal processing machine tool structure data, thereby obtaining carbon tool steel tool data and hard alloy tool data;

[0117] In this embodiment, according to the design drawings of the machine tool and the tool material characteristic data, the physical characteristics such as geometric dimensions, shape, hardness, and wear resistance of carbon tool steel tools and carbide tools are extracted respectively. For carbon tool steel tools, the characteristics such as the length, width, and angle of the cutting edge are extracted, and the performance characteristics of the tool are obtained by combining the hardness value (for example, the hardness is 55HRC) and the wear resistance data of the material. For carbide tools, the data such as its overall dimensions, material hardness (for example, the hardness is 90HRA), and the use temperature range are extracted, and its heat resistance and fatigue resistance are analyzed in combination with the alloy composition (such as containing tungsten, cobalt, etc.). The feature extraction process needs to further improve the working state characteristics of the tool through the sensor data of the processing machine tool and the real-time monitoring data of the mechanical parts. Finally, the carbon tool steel tool data and the carbide tool data are stored separately and used as the basic data for subsequent analysis.

[0118] Step S22: performing cutting force anomaly analysis on the carbon tool steel tool data, thereby obtaining cutting force anomaly data of the carbon tool steel tool;

[0119] In this embodiment, a force sensor installed on a machine tool is used to collect cutting force data generated during the cutting process. These data include the size, direction, and frequency of the cutting force. Then, the cutting force data is compared with the cutting force value within the normal working range. The normal cutting force range needs to be preset according to the workpiece material, tool geometry, and processing parameters (such as cutting speed, feed speed, depth, etc.). For example, the maximum cutting force of a carbon tool steel tool is set to 200N and the minimum cutting force is set to 50N. On this basis, the actual collected cutting force data is analyzed for abnormalities. If the cutting force exceeds the set threshold (for example, the cutting force exceeds 250N), it is marked as abnormal cutting force, and the time point when the abnormality occurs and related data are recorded. Abnormal data include but are not limited to overload, excessive impact force, or abnormal cutting force fluctuations, and finally the abnormal cutting force data of carbon tool steel tools is obtained for subsequent tool status diagnosis and optimization.

[0120] Step S23: performing noise anomaly analysis on the cemented carbide tool data, thereby obtaining cemented carbide tool noise anomaly data;

[0121] In this embodiment, a sound sensor or a vibration sensor installed on a processing machine tool is used to collect noise data generated when the tool is working. The normal working range of the noise is set, which is usually defined by experiments or historical data. For example, the noise value of a carbide tool under normal operating conditions should be controlled between 50-70dB. If the noise generated by the tool exceeds the set threshold (for example, more than 80dB), it is marked as noise abnormality. Noise analysis requires data comparison and matching in combination with parameters such as the tool speed, cutting depth, and feed rate. By comparing the collected real-time noise data with the preset standard threshold, the abnormal noise source is identified, and the abnormal carbide tool noise data is obtained. These abnormal data can be used as a basis for abnormal tool wear, installation problems or other faults.

[0122] Step S24: Integrate the abnormal operation status of the tool according to the abnormal cutting force data of the carbon tool steel tool and the abnormal noise data of the cemented carbide tool, so as to obtain the abnormal operation status data of the tool.

[0123] In this embodiment, the abnormal cutting force data is compared with the abnormal noise data to find the correlation between the two. For example, when the cutting force exceeds the normal range, whether the noise value also changes abnormally. During the integration process, it is necessary to set a suitable abnormal threshold. For example, when the abnormal cutting force value is greater than the set threshold (250N) and the noise value is greater than the set threshold (80dB), the tool is considered to be in an abnormal state. During the integration process, the working environment and working characteristics of different types of tools are considered to ensure that the abnormal state of each tool can be accurately reflected. Ultimately, the integrated data includes the abnormal type of each tool (such as overload, wear, damage, etc.), the time of occurrence of the abnormality, the severity and the cause of the failure. This data will be used for subsequent fault diagnosis, early warning and tool management optimization.

[0124] Optionally, step S22 is specifically:

[0125] Step S221: performing cutting simulation on the carbon tool steel tool data, thereby obtaining the carbon tool steel tool cutting data;

[0126] In this embodiment, it is necessary to use the obtained carbon tool steel tool data for cutting simulation. In order to perform cutting simulation, the geometric model of the tool and the workpiece is input using the CAD / CAM system of the CNC machine tool, and the processing parameters such as cutting speed, feed rate, cutting depth, etc. are set at the same time. Select appropriate cutting simulation software (for example, ANSYS or Simulink, etc.) and input these parameters, and simulate through finite element analysis (FEA) or discrete element method (DEM). During the simulation process, it is necessary to calculate the heat conduction, cutting force, material removal rate, cutting vibration, etc. of the tool. According to the simulation results, the cutting data of the carbon tool steel tool can be obtained, including the distribution of cutting force, tool stress, cutting heat, tool wear prediction, etc. These cutting data will provide basic data for subsequent frequency domain analysis.

[0127] Step S222: performing Fourier transform on the carbon tool steel tool cutting data to obtain the carbon tool steel tool cutting frequency domain data;

[0128] In this embodiment, the time domain signal data obtained from the cutting simulation is obtained, including the time series data of the cutting force, the tool vibration data, etc. These time domain signals are input into the Fourier transform algorithm (for example, using the fast Fourier transform FFT algorithm), and the time domain data is converted into frequency domain data through Fourier transform. In this process, a suitable sampling frequency (for example, 1000Hz) and number of sampling points (for example, 1024 points) are set to ensure the accuracy and frequency resolution of the data. The frequency domain data after Fourier transformation can reveal the energy distribution at different frequencies and help analyze the mechanical behavior of different frequency bands during the cutting process. Finally, the obtained frequency domain data of carbon tool steel tool cutting includes the amplitude and phase information of each frequency band, as well as the corresponding frequency range, providing data support for subsequent noise analysis.

[0129] Step S223: performing noise frequency statistics according to the cutting frequency domain data of the carbon tool steel tool, so as to obtain high-frequency noise data and low-frequency noise data;

[0130] In this embodiment, different frequency components in the frequency domain data are analyzed to identify high-frequency and low-frequency noise. In this process, a frequency threshold is set, for example, the low-frequency range is set to 0-500Hz, and the high-frequency range is set to above 500Hz. The frequency domain data is divided into two parts, low-frequency and high-frequency, and the power spectral density (PSD) of each frequency band is calculated respectively, and the noise energy of each frequency band is recorded. For example, in the low-frequency range, if the power spectral density of the frequency interval 0-500Hz is large, it is marked as low-frequency noise; similarly, the part with a frequency greater than 500Hz is marked as high-frequency noise. By counting the noise data in different frequency ranges, high-frequency noise data and low-frequency noise data are finally obtained, which will be further used for fluctuation statistics and abnormality assessment.

[0131] Step S224: performing fluctuation statistics according to the low-frequency noise data, thereby obtaining low-frequency noise high fluctuation data;

[0132] In this embodiment, a statistical analysis tool (such as the NumPy library in Matlab or Python) is used to calculate the fluctuation of the low-frequency noise data. The fluctuation amplitude of the noise signal is evaluated by calculating the standard deviation, variance and other fluctuation indicators of the low-frequency noise data. For example, if the low-frequency noise data is in the frequency range of 0-500Hz, the standard deviation of the time series data of this frequency band is calculated and compared with the set fluctuation threshold (for example, when the fluctuation standard deviation exceeds 5dB, it is considered to be high fluctuation). In this process, a suitable threshold is set to mark the fluctuation of the low-frequency noise. If the fluctuation amplitude of the low-frequency noise is too large (such as exceeding 10dB), it is considered that the noise has a large fluctuation and is marked as low-frequency noise high fluctuation data.

[0133] Step S225: performing an aggravation evaluation on the low-frequency noise high fluctuation data to obtain the low-frequency noise high fluctuation aggravation data;

[0134] In this embodiment, the aggravation of low-frequency noise is evaluated based on the high fluctuation data of low-frequency noise, combined with process parameters such as cutting force data and vibration data. A threshold for aggravation evaluation is set. For example, when the fluctuation amplitude of low-frequency noise exceeds 10dB, the noise is considered to be aggravated. The low-frequency noise data with a fluctuation amplitude greater than this threshold is analyzed, and the time point and amplitude change of the noise aggravation are recorded. For example, if the fluctuation amplitude of low-frequency noise exceeds 10dB within a certain period of time, and is accompanied by an abnormal increase in cutting force (such as a cutting force fluctuation of more than 15%), it is considered that there is a phenomenon of aggravated low-frequency noise during this period of time. Finally, the aggravation evaluation results in the high fluctuation data of low-frequency noise.

[0135] Step S226: Perform cutting force anomaly integration of carbon tool steel cutting tools according to the low-frequency noise high fluctuation aggravation data and the high-frequency noise data, so as to obtain cutting force anomaly data of carbon tool steel cutting tools.

[0136] In this embodiment, the data of high fluctuation intensification of low-frequency noise is compared and analyzed with the data of high-frequency noise. Combined with the cutting force data, when the low-frequency noise and high-frequency noise fluctuate and intensify at the same time (such as the low-frequency noise fluctuation exceeds 10dB and the high-frequency noise is greater than 80dB), it is considered that the cutting force of the tool at this time is abnormal. According to this judgment standard, these data are integrated to form the abnormal cutting force data of carbon tool steel tools. For example, if the fluctuation amplitude of the cutting force exceeds the normal range (such as exceeding 200N) and is accompanied by significant noise fluctuations (both low-frequency and high-frequency noise are abnormal), it is marked as abnormal tool cutting force. These data will be used for subsequent tool status monitoring and fault diagnosis.

[0137] Optionally, step S23 is specifically:

[0138] Step S231: extracting noise sensing features from cemented carbide tool data to obtain noise sensing data;

[0139] In this embodiment, the sound data generated when the tool is working is collected by a noise sensor. A noise sensor suitable for the frequency response range is selected, for example, a microphone sensor with a frequency response range of 20Hz to 20kHz, and it is installed near the tool to ensure that the noise generated during the cutting process is accurately captured. The noise signal data collected by the sensor in real time is stored at a certain sampling frequency (for example, 1kHz) and sampling time (for example, 10 seconds). The original noise signal is transmitted to the processing system through a data acquisition card. The collected signal is analyzed in the time domain using specific data processing software (such as Matlab), and the characteristics of the noise signal, such as instantaneous power, average power, etc., are extracted to provide the original noise sensing data for subsequent data processing and analysis.

[0140] Step S232: performing filtering and noise reduction processing according to the noise sensing data, thereby obtaining noise sensing filtering and noise reduction data;

[0141] In this embodiment, a bandpass filter is used to remove low-frequency noise and high-frequency noise. The operating frequency range of the bandpass filter is selected, for example, the low-frequency lower limit is set to 100Hz, and the high-frequency upper limit is set to 5kHz, so as to ensure that the effective noise signal during the cutting of the carbide tool is retained. By setting the parameters of the filter, a digital filter (for example, a Butterworth filter or a Chebyshev filter) is used to process the original noise signal, and unnecessary frequency components in the noise signal are removed to obtain filtered and denoised data. In a specific implementation, the order (for example, 4th order) and cutoff frequency of the filter need to be adjusted according to the spectral characteristics of the noise signal. The filtered data will be used for subsequent power spectrum analysis to reduce the interference of external noise on the analysis results.

[0142] Step S233: performing power spectrum drawing on the noise sensor filtering noise reduction data, thereby obtaining the noise sensor filtering noise reduction power spectrum;

[0143] In this embodiment, the time domain signal after filtering and noise reduction is converted into a frequency domain signal using a fast Fourier transform (FFT) algorithm. An appropriate window function (e.g., a Hanning window or a rectangular window) is selected to perform windowing on the original signal to reduce spectrum leakage. The windowed data is subjected to frequency domain analysis using the FFT algorithm to obtain the amplitude and phase information of the signal at each frequency point. The power spectrum of the noise signal is plotted to analyze the distribution of each frequency component in the frequency domain. By setting the frequency axis range (e.g., 0-5kHz), the main frequency band of the cutting noise is ensured to be fully covered. The power spectrum diagram shows the power intensity of different frequency components for further noise analysis and classification.

[0144] Step S234: performing density calculation according to the noise sensing filter noise reduction power spectrum, thereby obtaining the noise sensing filter noise reduction power spectrum density data;

[0145] In this embodiment, the power spectrum amplitude value is used to normalize the energy distribution of the frequency domain signal to calculate the power spectrum density (PSD) of each frequency band. Specifically, the power spectrum density is calculated using the formula:

[0146]

[0147] where |FFT(f)| 2 represents the square of the amplitude of each frequency point, and Δf is the frequency resolution, that is, the sampling frequency divided by the number of sampling points. By calculating PSD, the energy distribution of each frequency band can be obtained. At this time, the calculation frequency resolution of the power spectrum density (for example, 1Hz or 10Hz) is set to determine the accuracy of the calculation result. Finally, the noise sensor filter noise reduction power spectrum density data is obtained for subsequent spectrum centroid calculation.

[0148] Step S235: Calculating the spectral centroid of the noise sensor filter noise reduction power spectrum according to the noise sensor filter noise reduction power spectrum density data, thereby obtaining spectral centroid data;

[0149] In this embodiment, the spectrum center of gravity is calculated based on the noise sensor filter noise reduction power spectrum density data. The spectrum center of gravity can reflect the spectrum distribution characteristics of the signal and indicate the "center position" of the power spectrum. In the specific operation, the spectrum center of gravity is calculated using the following formula:

[0150]

[0151] Among them, fi is the i-th frequency point, (f iis the power spectrum density value corresponding to the frequency point. By calculating the weighted average of all frequency points, the center of gravity of the spectrum can be obtained, which represents the main frequency component of the noise signal. In order to ensure the accuracy of the calculation results, the frequency range is set to 0-5kHz, and the center of gravity of the spectrum is calculated within this range. The center of gravity data of the spectrum provides an important basis for subsequent offset identification and reflects the main frequency offset of the noise signal.

[0152] Step S236: performing offset identification on the spectrum center of gravity data, thereby obtaining spectrum center of gravity offset data;

[0153] In this embodiment, a reference spectrum center of gravity value is set according to the spectrum center of gravity data. For example, the initial spectrum center of gravity is set to 1.5kHz. By comparing with the reference value, it is detected whether the spectrum center of gravity has shifted significantly. For example, a threshold value (for example, 0.1kHz) is set. When the difference between the spectrum center of gravity and the reference value exceeds the threshold, it is determined that a shift has occurred. The direction and amplitude of the shift will be recorded as shift data. If the amplitude of the spectrum center of gravity shift exceeds the set threshold, it is considered that abnormal noise exists, and further tool abnormality identification is performed. The spectrum center of gravity shift data is obtained for identifying tool state changes.

[0154] Step S237: performing tool abnormality time identification on the cemented carbide tool data according to the spectrum center of gravity offset data, thereby obtaining the cemented carbide tool noise abnormality data.

[0155] In this embodiment, the spectrum center of gravity offset data is matched with the working time data of the cemented carbide tool to identify the time period in which the spectrum center of gravity is offset. By setting a time window (for example, 5 seconds or 10 seconds), the spectrum center of gravity offset in each time window is compared with a preset threshold value (such as a 0.1kHz offset). When the spectrum center of gravity offset exceeds the threshold, the time period is recorded as an abnormal time period. Combined with other tool data (such as cutting force data, temperature data, etc.), it is determined whether it is an abnormal noise phenomenon of the cemented carbide tool. Identify the abnormal noise data of the cemented carbide tool to provide a basis for tool maintenance.

[0156] Optionally, step S3 specifically includes:

[0157] Step S31: extracting servo motor features and stepper motor features according to the metal processing machine tool structure data, thereby obtaining servo motor data and stepper motor data;

[0158] In the present embodiment, the operating parameter data of servo motor and stepper motor are obtained, and these data include current, rotation speed, torque, power, etc. The servo motor data is obtained by the sensor connected to the motor control system, and these sensors record the operating data such as the current and rotation speed of the motor in real time. The stepper motor data is collected by the stepper motor drive controller, mainly recording the movement of each step of the stepper motor and the current and voltage fluctuations. The data acquisition frequency of the sensor is set to 500Hz to ensure the accuracy of data acquisition. After using a suitable data acquisition system (for example, LabVIEW) to collect the data, specific data processing software is used to pre-process it, such as denoising and filtering. According to the collected motor data, the characteristic data of the servo motor and stepper motor are extracted, and the servo motor data and stepper motor data are obtained respectively, so as to prepare data for subsequent analysis.

[0159] Step S32: performing transient torque analysis on the servo motor data to obtain transient torque data of the servo motor;

[0160] In this embodiment, transient torque refers to the instantaneous torque change of the servo motor under dynamic conditions such as starting, acceleration, and deceleration. First, using the data of the servo motor, including current and speed data, the transient torque is calculated by establishing a relationship model between torque, current, and speed. The torque formula is:

[0161] T=K e I;

[0162] Where T is the torque, K e is the torque constant, and I is the current. The current value is obtained from the sensor, and the speed value is measured by a tachometer or encoder. When performing transient analysis on the servo motor data, the acquisition signal window is set to 1 second, and a high-speed data acquisition card is used for real-time monitoring. According to the analysis results, the torque value at each moment is calculated, and the torque-time curve is drawn. This curve reflects the torque fluctuation of the servo motor under different working conditions, and finally the transient torque data of the servo motor is obtained.

[0163] Step S33: performing winding insulation abnormality analysis on the stepper motor data, thereby obtaining the stepper motor winding insulation abnormality data;

[0164] In this embodiment, the winding insulation performance of the stepper motor is a key parameter for the normal operation of the motor. If the winding insulation is damaged, the motor will short-circuit or overheat. In this step, first, the working state of the winding is monitored by the voltage and current signals of the motor. The current and voltage waveform data of the stepper motor are collected using a high-precision voltage sensor and a current sensor. Then, based on the current-voltage relationship, the insulation resistance of the stepper motor winding is measured using an insulation resistance tester, and the standard value of the insulation resistance is set to be abnormal below 5MΩ. The collected signal is subjected to spectrum analysis to analyze whether high-frequency noise or pulse signals appear in the motor during operation. These abnormal signals are related to the insulation damage of the winding. The signal is subjected to FFT (Fast Fourier Transform) processing to extract the spectrum features. When a frequency component beyond the normal operating range appears in the spectrum, it is determined that the winding insulation is abnormal. According to the above analysis, the abnormal insulation data of the stepper motor winding is obtained, and the abnormal time point and abnormal type are recorded to provide a basis for subsequent maintenance.

[0165] Step S34: performing motor overload integration according to the servo motor transient torque data and the stepper motor winding insulation abnormality data, thereby obtaining the motor overload data.

[0166] In this embodiment, based on the transient torque data of the servo motor, it is analyzed whether the motor is overloaded under dynamic load conditions. A threshold is set. For example, when the torque exceeds 120% of the rated torque, the motor is considered to be overloaded. According to the abnormal insulation data of the stepper motor winding, if the winding insulation is abnormal and the current exceeds the rated range (for example, exceeding 1.2 times the rated current), it can be determined that the motor has an overload problem. By integrating these two data sources, when the transient torque data of the servo motor exceeds the set threshold and the insulation abnormality of the stepper motor winding occurs, combined with the working state of the motor, the motor is determined to be overloaded. A time window analysis is performed on the motor overload situation to determine the specific overload duration and frequency. Finally, the motor overload data is generated, including the time period, duration and overload intensity of the overload, as a reference for subsequent maintenance and adjustment.

[0167] Optionally, step S32 is specifically:

[0168] Step S321: collecting torque signals of the servo motor data to obtain torque signal data;

[0169] In this embodiment, a torque sensor (such as a torque sensor or a strain gauge) is installed at the shaft end of the servo motor or the load side of the motor to directly measure the output torque of the motor. The sensor connects the signal to a data acquisition system (for example, a NIDAQ system) via a cable, and the sampling frequency of the system is set to 2kHz to ensure accurate capture of the signal. The acquisition process of the torque signal includes converting the current and voltage values ​​obtained by the sensor into torque data. The current signal is converted into torque data by setting a torque constant (for example, a torque constant of 0.5Nm / A). The collected torque signal data is saved in the form of time series data and prepared for the calculation of subsequent steps.

[0170] Step S322: Calculate the instantaneous change rate according to the torque signal data, so as to obtain the instantaneous change rate of the torque signal;

[0171] In this embodiment, the instantaneous rate of change is calculated based on the torque signal data. The instantaneous rate of change reflects the speed of change of the torque signal and is obtained by performing differential processing on the torque signal in a short time window. Assuming that the torque signal data is T(t1), T(t2), ..., T(tn), where each T(ti) represents the torque value at time ti, the instantaneous rate of change calculation formula is:

[0172]

[0173] Among them, Δt is the time interval between two consecutive sampling points, which is usually set to 1 millisecond. Fast Fourier transform (FFT) is used to filter the high-frequency components of the torque signal and eliminate low-frequency noise, thereby obtaining smooth instantaneous rate of change data. Through this method, the instantaneous rate of change of the torque signal at each moment can be obtained, providing data support for the subsequent high-rate of change torque signal extraction.

[0174] Step S323: performing change rate statistics according to the instantaneous change rate of the torque signal, so as to obtain high instantaneous change rate torque signal data;

[0175] In this embodiment, a change rate threshold is set, for example, 0.05Nm / s. This threshold is determined by historical data analysis. When the instantaneous change rate exceeds the threshold, it is considered that the torque signal at that moment changes more drastically. The instantaneous change rate data is compared with the set threshold. If the change rate at a certain moment exceeds the threshold, the torque signal at that moment is considered to be a high instantaneous change rate signal. The signal data with high instantaneous change rates are screened and recorded, including the time point when they occur and the corresponding torque value. These torque signals with high instantaneous change rates indicate that the motor has experienced rapid load changes or unstable operating conditions, and further analysis will be performed in subsequent steps.

[0176] Step S324: identifying a mutation point of the torque signal data according to the high instantaneous change rate torque signal data, thereby obtaining torque signal mutation point data;

[0177] In this embodiment, mutation points are identified based on torque signal data with high instantaneous change rates. A mutation point refers to a point where the torque signal changes significantly in a short period of time, which is usually related to overload, impact load or other abnormal conditions of the motor. First, a sliding window algorithm is used to divide the high-change rate signal into multiple time windows, and the change amplitude of the torque signal in each window is calculated. If the torque change in a certain time window exceeds a preset threshold (for example, the torque change amplitude exceeds 5Nm), it is considered that a mutation has occurred at that time point. The identification of mutation points is based on the change rate and amplitude, combined with the extreme value detection algorithm of the torque signal, and by analyzing the change trend of the torque, it is determined whether it is a mutation. The identified mutation point data (including occurrence time and change amplitude) is recorded as the torque signal mutation point data to provide a basis for subsequent motor state evaluation.

[0178] Step S325: Visualize the transient fluctuation of the servo motor according to the torque signal mutation point data, so as to obtain the transient torque data of the servo motor.

[0179] In this embodiment, the transient fluctuation diagram of the servo motor is constructed in combination with the mutation point data of the torque signal. Use a visualization tool (such as the Matplotlib library in MATLAB or Python) to draw the torque change curve of the servo motor in different time periods according to the mutation point data. The specific operation includes: starting from the mutation point, drawing the change trend of the torque signal, marking the mutation point and emphasizing its change amplitude. In order to more intuitively display the transient fluctuation of the motor, time axis, ruler and other information can also be added to the figure to observe the fluctuation of the motor torque. Through this method, the law of transient fluctuations of the servo motor during operation can be clearly demonstrated, and the mutation and transient fluctuation of the torque can be displayed through a visual chart, providing reliable data support for subsequent fault diagnosis and maintenance decisions.

[0180] Optionally, step S33 is specifically:

[0181] Step S331: collecting insulation resistance data of the stepper motor, thereby obtaining insulation resistance data of the stepper motor;

[0182] In this embodiment, it is necessary to accurately collect the insulation resistance of the stepper motor. Use a high-precision digital megohmmeter (such as Hioki3153 or Fluke 1507) to perform an insulation resistance test to ensure that the test voltage is 500V, which meets the working requirements of the stepper motor. The test probe is contacted with the stator and rotor parts of the stepper motor respectively to obtain the insulation resistance value of the motor winding. The resistance data during the test should be recorded every 10 seconds to obtain a stable resistance reading. During the test, it is also necessary to ensure that the ambient temperature is around 25°C, and record the temperature and humidity data for subsequent analysis. During the whole process, if the resistance value is lower than a predetermined threshold (for example, 50MΩ), it is marked as abnormal data and further analyzed. The collected insulation resistance data is stored in a database for subsequent temperature statistics and modeling analysis.

[0183] Step S332: performing temperature statistics according to the stepper motor insulation resistance data, thereby obtaining high-temperature stepper motor insulation resistance data;

[0184] In this embodiment, the collected insulation resistance data is selected and paired with the corresponding recorded temperature data. According to the historical data analysis, the temperature threshold is set to 50°C, and the insulation resistance data under temperature conditions higher than this is screened out. Next, the temperature compensation algorithm is applied to adjust the insulation resistance data, and the resistance value of each temperature point is corrected using a standard temperature coefficient (such as -0.1% / °C). For the resistance values ​​at different temperatures, the resistance distribution in each temperature interval (such as every 5°C interval) is statistically analyzed to obtain the resistance change trend under high temperature. The output high-temperature stepper motor insulation resistance data will be used for subsequent modeling to further analyze the impact of high temperature environment on insulation resistance.

[0185] Step S333: constructing an insulation resistance-temperature relationship model according to the stepper motor insulation resistance data and the high-temperature stepper motor insulation resistance data, thereby obtaining an insulation resistance-temperature relationship model;

[0186] In this embodiment, the collected insulation resistance data of the stepper motor and the correction data under high temperature conditions are used to establish a mathematical relationship between resistance and temperature using a regression analysis method (such as linear regression or polynomial regression). The specific steps include: fitting the resistance value at each temperature point, calculating its correlation with temperature, and obtaining a temperature coefficient (such as 0.05% / °C). For high temperature data greater than 50°C, a nonlinear regression algorithm (such as exponential regression or power function regression) is further used to optimize the model to ensure that it can accurately reflect the changing trend of insulation resistance at different temperatures. The final model includes a functional relationship between temperature and insulation resistance, and the accuracy of the model is measured by the degree of fit (such as R 2 value) to ensure its effectiveness under actual working conditions.

[0187] Step S334: performing insulation material aging simulation on the insulation resistance-temperature relationship model to obtain insulation material aging data;

[0188] In this embodiment, the simulation process includes setting specific aging conditions (for example, high temperature operation for 1000 hours, the temperature range is 60°C to 80°C), and then predicting the insulation resistance change of the motor under this condition based on the temperature relationship model. By setting the time interval (for example, a simulation is performed every 200 hours), the aging process of the insulating material is simulated to obtain the resistance change data after aging. During the simulation process, it is necessary to consider the accelerated effect of temperature on resistance and make corrections based on the workload of the stepper motor. In the simulation results, if the resistance value is lower than the preset safety threshold (such as 30MΩ), it is recorded as an indication of material aging. By comparing the aging data of different time periods, detailed data on the aging of the insulating material is obtained and output in the form of a chart to facilitate subsequent analysis.

[0189] Step S335: obtaining the stepper motor operating voltage data;

[0190] In this embodiment, a voltage monitoring device (such as a Fluke 435-II power quality analyzer or PowerLogicPM8000 series) is connected to the stepper motor drive system to collect the working voltage of the motor. The voltage data is recorded once per second to ensure that the voltage fluctuation is within a stable range. The voltage data will be stored in the data recording system and can be exported to CSV format for analysis according to specific needs. The data collection of the working voltage includes the waveforms of the DC voltage and the AC voltage, recording the high and low voltages and the voltage fluctuation range to ensure that sufficient data support can be provided in the subsequent voltage failure judgment process.

[0191] Step S336: performing breakdown voltage statistics according to the insulation material aging data, thereby obtaining insulation material aging breakdown voltage data;

[0192] In this embodiment, according to the change of insulation resistance with the aging process, a breakdown voltage prediction model is used to simulate the breakdown voltage of the insulating material at different aging stages. The prediction of the breakdown voltage adopts known electrical theory and material properties (for example, the breakdown strength of the insulating material usually decreases with the decrease of the resistance value). The breakdown voltage data under different aging degrees are calculated by combining the resistance change trend of the motor during the aging process through the breakdown voltage standard provided by experimental data or literature. The breakdown voltage value of each aging stage is counted, and the statistical results are output as a basis for subsequent judgment of whether the motor has insulation failure.

[0193] Step S337: Perform voltage insulation failure determination based on the stepper motor operating voltage data and the insulation material aging breakdown voltage data, thereby obtaining the stepper motor winding insulation abnormality data.

[0194] In this embodiment, the actual working voltage of the stepper motor is compared with the breakdown voltage data obtained by simulation. If the working voltage of the motor exceeds the breakdown voltage threshold of the aging material, it is judged as insulation failure. A breakdown voltage safety threshold is set, such as 80% breakdown voltage. If the motor working voltage exceeds the threshold, it is recorded as insulation abnormality. In this process, further analysis is performed on the voltage fluctuations under different working conditions to ensure the accuracy of the judgment. Finally, the judgment result (i.e., whether there is winding insulation abnormality) is output in the form of data and marked as abnormal data for subsequent maintenance and preventive measures.

[0195] Optionally, the present specification further provides a metal processing machine tool operating state prediction system, which is used to execute the metal processing machine tool operating state prediction method as described above, and the metal processing machine tool operating state prediction system includes:

[0196] A metal processing machine tool model building module is used to obtain metal processing machine tool design data, and extract geometric structure design features based on the metal processing machine tool design data, so as to obtain geometric structure design data; and build a metal processing machine tool model based on the geometric structure design data, so as to obtain a metal processing machine tool model;

[0197] The tool abnormal operation state analysis module is used to extract the features of carbon tool steel tools and carbide tools according to the structure data of metal processing machine tools, so as to obtain the data of carbon tool steel tools and carbide tools; analyze the abnormal operation state of tools according to the data of carbon tool steel tools and carbide tools, so as to obtain the data of abnormal operation state of tools;

[0198] The motor overload analysis module is used to extract servo motor features and stepper motor features according to the metal processing machine tool structure data, so as to obtain servo motor data and stepper motor data; and to perform motor overload analysis according to the servo motor data and stepper motor data, so as to obtain motor overload data;

[0199] The cutting simulation module is used to fuse the abnormal state of the machining machine tool according to the motor overload data and the abnormal operation state data of the tool, so as to obtain the abnormal state data of the machining machine tool; and to perform cutting simulation according to the metal machining machine tool model, so as to obtain the cutting simulation data of the metal machining machine tool;

[0200] The abnormal state prediction module of metal processing machine tools is used to predict the abnormal state of metal processing machine tool cutting simulation data according to the abnormal state data of the processing machine tools, so as to obtain the abnormal state prediction data of the metal processing machine tools and upload it to the metal processing machine tool management platform to perform the abnormal state prediction task.

[0201] The metal processing machine tool operation status prediction system of the present invention can implement any metal processing machine tool operation status prediction method of the present invention, and is used to combine the operation and signal transmission medium between various modules to complete the metal processing machine tool operation status prediction method. The internal modules of the system cooperate with each other, thereby improving the machine tool fault warning and maintenance efficiency.

[0202] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0203] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for predicting the operating status of a metal processing machine tool, characterized in that: The following steps are involved: Step S1: acquiring metal processing machine tool design data, and performing geometric structure design feature extraction according to the metal processing machine tool design data, thereby obtaining geometric structure design data; Constructing a metal processing machine tool model according to the geometric structure design data, thereby obtaining a metal processing machine tool model; Step S2: extracting features of carbon tool steel tools and carbide tools according to the metal processing machine tool structure data, thereby obtaining carbon tool steel tool data and carbide tool data; analyzing tool abnormal operation status according to the carbon tool steel tool data and carbide tool data, thereby obtaining tool abnormal operation status data; Step S3: extracting servo motor features and stepper motor features according to the metal processing machine tool structure data, thereby obtaining servo motor data and stepper motor data; performing motor overload analysis according to the servo motor data and stepper motor data, thereby obtaining motor overload data; Step S4: performing abnormal state fusion of the machining machine tool according to the motor overload data and the abnormal operation state data of the tool, thereby obtaining abnormal state data of the machining machine tool; performing cutting simulation according to the metal machining machine tool model, thereby obtaining cutting simulation data of the metal machining machine tool; Step S5: Perform abnormal state prediction on the metal processing machine tool cutting simulation data according to the abnormal state data of the processing machine tool, so as to obtain the abnormal state prediction data of the metal processing machine tool, and upload it to the metal processing machine tool management platform to execute the abnormal state prediction task.

2. The method for predicting the operating status of a metal processing machine tool according to claim 1, characterized in that: Step S1 is specifically as follows: Step S11: acquiring metal processing machine tool design data, and performing geometric structure design feature extraction according to the metal processing machine tool design data, thereby obtaining geometric structure design data; Step S12: extracting the geometric structure features of the bed and the spindle structure features according to the geometric structure design data, thereby obtaining the geometric structure data of the bed and the spindle structure data; Step S13: assembling the machine tool model with the bed geometry data according to the spindle structure data, thereby obtaining the machine tool model; Step S14: performing virtual simulation on the machine tool model to obtain virtual simulation data of the machine tool model; Step S15: Optimizing the collision of metal processing parts on the machine tool model according to the virtual simulation data of the machine tool model, thereby obtaining a metal processing machine tool model.

3. The method for predicting the operating status of a metal processing machine tool according to claim 2, characterized in that: Step S15 is specifically as follows: Step S151: performing collision area identification according to the virtual simulation data of the machine tool model, thereby obtaining collision area data of the machine tool; Step S152: extracting metal fixture collision features from the machine tool collision area data, thereby obtaining metal fixture collision data; Step S153: adjusting the safety clearance according to the metal fixture collision data, thereby obtaining the metal machine tool safety clearance data; Step S154: Optimizing the machine tool model according to the metal machine tool safety clearance data, thereby obtaining a metal processing machine tool model.

4. The method for predicting the operating status of a metal processing machine tool according to claim 1, characterized in that: Step S2 is specifically as follows: Step S21: extracting features of carbon tool steel tools and hard alloy tool tools according to the metal processing machine tool structure data, thereby obtaining carbon tool steel tool data and hard alloy tool data; Step S22: performing cutting force anomaly analysis on the carbon tool steel tool data, thereby obtaining cutting force anomaly data of the carbon tool steel tool; Step S23: performing noise anomaly analysis on the cemented carbide tool data, thereby obtaining cemented carbide tool noise anomaly data; Step S24: Integrate the abnormal operation status of the tool according to the abnormal cutting force data of the carbon tool steel tool and the abnormal noise data of the cemented carbide tool, so as to obtain the abnormal operation status data of the tool.

5. The method for predicting the operating status of a metal processing machine tool according to claim 4, characterized in that: Step S22 is specifically as follows: Step S221: performing cutting simulation on the carbon tool steel tool data, thereby obtaining the carbon tool steel tool cutting data; Step S222: performing Fourier transform on the carbon tool steel tool cutting data to obtain the carbon tool steel tool cutting frequency domain data; Step S223: performing noise frequency statistics according to the cutting frequency domain data of the carbon tool steel tool, so as to obtain high-frequency noise data and low-frequency noise data; Step S224: performing fluctuation statistics according to the low-frequency noise data, thereby obtaining low-frequency noise high fluctuation data; Step S225: performing an aggravation evaluation on the low-frequency noise high fluctuation data to obtain the low-frequency noise high fluctuation aggravation data; Step S226: Perform cutting force anomaly integration of carbon tool steel cutting tools according to the low-frequency noise high fluctuation aggravation data and the high-frequency noise data, so as to obtain cutting force anomaly data of carbon tool steel cutting tools.

6. The method for predicting the operating status of a metal processing machine tool according to claim 4, characterized in that: Step S23 is specifically as follows: Step S231: extracting noise sensing features from cemented carbide tool data to obtain noise sensing data; Step S232: performing filtering and noise reduction processing according to the noise sensing data, thereby obtaining noise sensing filtering and noise reduction data; Step S233: performing power spectrum drawing on the noise sensor filtering noise reduction data, thereby obtaining the noise sensor filtering noise reduction power spectrum; Step S234: performing density calculation according to the noise sensing filter noise reduction power spectrum, thereby obtaining the noise sensing filter noise reduction power spectrum density data; Step S235: Calculating the spectral centroid of the noise sensor filter noise reduction power spectrum according to the noise sensor filter noise reduction power spectrum density data, thereby obtaining spectral centroid data; Step S236: performing offset identification on the spectrum center of gravity data, thereby obtaining spectrum center of gravity offset data; Step S237: performing tool abnormality time identification on the cemented carbide tool data according to the spectrum center of gravity offset data, thereby obtaining the cemented carbide tool noise abnormality data.

7. The method for predicting the operating status of a metal processing machine tool according to claim 1, characterized in that: Step S3 is specifically as follows: Step S31: extracting servo motor features and stepper motor features according to the metal processing machine tool structure data, thereby obtaining servo motor data and stepper motor data; Step S32: performing transient torque analysis on the servo motor data to obtain transient torque data of the servo motor; Step S33: performing winding insulation abnormality analysis on the stepper motor data, thereby obtaining the stepper motor winding insulation abnormality data; Step S34: performing motor overload integration according to the servo motor transient torque data and the stepper motor winding insulation abnormality data, thereby obtaining the motor overload data.

8. The method for predicting the operating status of a metal processing machine tool according to claim 7, characterized in that: Step S32 is specifically as follows: Step S321: collecting torque signals of the servo motor data to obtain torque signal data; Step S322: Calculate the instantaneous change rate according to the torque signal data, so as to obtain the instantaneous change rate of the torque signal; Step S323: performing change rate statistics according to the instantaneous change rate of the torque signal, so as to obtain high instantaneous change rate torque signal data; Step S324: identifying a mutation point of the torque signal data according to the high instantaneous change rate torque signal data, thereby obtaining torque signal mutation point data; Step S325: Visualize the transient fluctuation of the servo motor according to the torque signal mutation point data, so as to obtain the transient torque data of the servo motor.

9. The method for predicting the operating status of a metal processing machine tool according to claim 7, characterized in that: Step S33 is specifically as follows: Step S331: collecting insulation resistance data of the stepper motor, thereby obtaining insulation resistance data of the stepper motor; Step S332: performing temperature statistics according to the stepper motor insulation resistance data, thereby obtaining high-temperature stepper motor insulation resistance data; Step S333: constructing an insulation resistance-temperature relationship model according to the stepper motor insulation resistance data and the high-temperature stepper motor insulation resistance data, thereby obtaining an insulation resistance-temperature relationship model; Step S334: performing insulation material aging simulation on the insulation resistance-temperature relationship model to obtain insulation material aging data; Step S335: obtaining the stepper motor operating voltage data; Step S336: performing breakdown voltage statistics according to the insulation material aging data, thereby obtaining insulation material aging breakdown voltage data; Step S337: Perform voltage insulation failure determination based on the stepper motor operating voltage data and the insulation material aging breakdown voltage data, thereby obtaining the stepper motor winding insulation abnormality data.

10. A metal processing machine tool operation status prediction system, characterized in that: Used to execute the metal processing machine tool operation state prediction method as claimed in claim 1, the metal processing machine tool operation state prediction system comprises: A metal processing machine tool model building module is used to obtain metal processing machine tool design data, and extract geometric structure design features based on the metal processing machine tool design data, so as to obtain geometric structure design data; and build a metal processing machine tool model based on the geometric structure design data, so as to obtain a metal processing machine tool model; The tool abnormal operation state analysis module is used to extract the features of carbon tool steel tools and carbide tools according to the structure data of metal processing machine tools, so as to obtain the data of carbon tool steel tools and carbide tools; analyze the abnormal operation state of tools according to the data of carbon tool steel tools and carbide tools, so as to obtain the data of abnormal operation state of tools; The motor overload analysis module is used to extract servo motor features and stepper motor features according to the metal processing machine tool structure data, so as to obtain servo motor data and stepper motor data; and to perform motor overload analysis according to the servo motor data and stepper motor data, so as to obtain motor overload data; The cutting simulation module is used to fuse the abnormal state of the machining machine tool according to the motor overload data and the abnormal operation state data of the tool, so as to obtain the abnormal state data of the machining machine tool; and to perform cutting simulation according to the metal machining machine tool model, so as to obtain the cutting simulation data of the metal machining machine tool; The abnormal state prediction module of metal processing machine tools is used to predict the abnormal state of metal processing machine tool cutting simulation data according to the abnormal state data of the processing machine tools, so as to obtain the abnormal state prediction data of the metal processing machine tools and upload it to the metal processing machine tool management platform to perform the abnormal state prediction task.

Images