Gear machining forging process control method and system

Abstract

The application discloses a gear machining forging process control method and system, and relates to the technical field of gear machining forging process control; the method comprises the following steps: detecting the local hardness of a workpiece to obtain local hardness distribution data of the workpiece; simulating the cutting process of the workpiece according to the hardness distribution data and initial machining parameters, and collecting cutting forces generated by each cutting point in the cutting process in real time; generating first machining parameters by adjusting the initial machining parameters according to the cutting forces, and cutting the workpiece by using the first machining parameters; collecting a workpiece surface image after cutting, and generating finishing machining parameters of the workpiece after cutting according to the workpiece surface image; and finishing machining the workpiece after cutting by using the finishing machining parameters, so as to ensure that micro ripples on the surface of the workpiece are eliminated in the finishing machining process. The application can realize the detection of the local hardness of the workpiece, adjust machining parameters according to the detection results and the cutting force, effectively eliminate micro ripples on the surface of the workpiece in the cutting and finishing machining processes, and improve the machining precision and consistency of the gear.

Description

Technical Field

[0001] This application relates to the field of gear processing and forging process control technology, and more specifically, to a gear processing and forging process control method and system. Background Technology

[0002] In intelligent gear forging workshops of high-end equipment manufacturing, product quality consistency and production efficiency face hidden challenges. Existing automated production lines rely on cutting parameters (depth of cut, feed rate, spindle speed, etc.) set based on the average characteristics of materials. While these parameters can meet the needs of routine production, they are unable to cope with subtle fluctuations in raw materials.

[0003] Even if the hardness of the blank is within the range allowed by the technical specifications, batch differences, fluctuations in alloy composition, or minor adjustments during heat treatment can still cause its physical properties to deviate from the upper limit of hardness. Traditional control systems lack real-time sensing capabilities, and if standard parameters are continued to be used for machining, the cutting resistance will increase significantly, and the load exceeding the design value for a long period of time poses a hidden threat to the machine tool.

[0004] Excessive cutting resistance can disrupt the stability of the lubricating oil film in the spindle bearing, leading to direct metal-to-metal contact, accelerating uneven wear between the rolling elements and raceways, and consequently causing high-frequency, micro-amplitude, irregular vibrations. These vibrations are difficult to detect and monitor using conventional methods, but they can be transmitted to the cutting edge of the tool, forming microscopic ripples on the workpiece surface that are difficult to detect with conventional measurements.

[0005] These microscopic defects cannot be eliminated in the finishing process and may even be amplified. Finishing is based on the roughing surface, which leads to batch-to-batch fluctuations in key indicators such as gear tooth profile accuracy, tooth direction accuracy, and surface roughness, seriously affecting the interchangeability and reliability requirements of high-end equipment for parts.

[0006] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0007] This application discloses a method and system for controlling the forging process of gear processing, which aims to solve the problem in the prior art where the cutting resistance is continuously high due to slight fluctuations in the hardness of the raw materials within the allowable range, which in turn causes hidden wear and irregular micro-vibrations in the machine tool spindle bearings, and ultimately forms imperceptible micro-ripples on the gear processing surface. These ripples are retained or even amplified in subsequent finishing processes, resulting in a significant decrease in the precision consistency of the final gear products.

[0008] The technical solution of this application is as follows:

[0009] In a first aspect, this application discloses a method for controlling the gear forging process, including:

[0010] Local hardness testing is performed on the workpiece to obtain local hardness distribution data.

[0011] Based on the local hardness distribution data of the workpiece and the initial machining parameters, the cutting process of the workpiece is simulated, and the cutting force generated at each cutting point during the cutting process is collected in real time.

[0012] The initial machining parameters are adjusted according to the cutting force at each cutting point to generate the first machining parameters, and the workpiece is cut using the first machining parameters.

[0013] After the cutting is completed, an image of the workpiece surface is acquired, and the finishing parameters of the workpiece after the cutting is generated based on the micro-ripples on the workpiece surface in the image.

[0014] The workpiece is finished by using finishing parameters to ensure that micro-ripples on the workpiece surface are eliminated during the finishing process.

[0015] This technical solution enables the detection of local hardness of the workpiece and the adjustment of machining parameters based on the detection results and cutting force. This effectively controls the surface quality of the workpiece during cutting and finishing, eliminates micro-ripples, and improves the machining accuracy and consistency of gears. It solves the problem that traditional methods cannot cope with the instability of machining accuracy caused by subtle changes in raw materials and equipment wear.

[0016] Furthermore, after simulating the cutting process of the workpiece based on the local hardness distribution data and initial machining parameters, and collecting the cutting force generated at each cutting point in real time during the cutting process, the process also includes:

[0017] Based on the cutting force at each cutting point and the preset machine tool spindle bearing health model, the local conductivity change of the machine tool spindle bearing is simulated, and the local simulated conductivity data of the machine tool spindle bearing is collected in real time.

[0018] During the cutting process of the workpiece using the first machining parameters, the local conductivity data of the machine tool spindle bearing is collected in real time.

[0019] Based on local simulated conductivity data and local conductivity data, determine whether there is early hidden wear in the machine tool spindle bearing;

[0020] In cases where the machine tool spindle bearing exhibits early latent wear, the steps of acquiring an image of the workpiece surface after cutting and subsequent steps are performed.

[0021] Through this technical solution, this application can determine whether there is early hidden wear in the bearing by simulating and monitoring the local conductivity data of the machine tool spindle bearing in real time, and adjust the subsequent processing steps in time when wear is detected, thereby effectively avoiding the decrease in processing accuracy caused by bearing wear, extending the service life of the equipment, and improving the reliability and stability of production.

[0022] Based on this, the process of cutting the workpiece using the first machining parameters also includes the following steps:

[0023] Real-time acquisition of the first microscopic vibration velocity data of the tool tip;

[0024] After determining whether early latent wear exists in the machine tool spindle bearing based on local simulated conductivity data and local conductivity data, the following steps are also included:

[0025] When the machine tool spindle bearing has early latent wear, the micro-vibration generated by the machine tool spindle bearing and the tool clamping system is simulated, and the second micro-vibration velocity data of the tool tip is collected in real time during the simulation process.

[0026] The step of cutting the workpiece using the first machining parameters also includes:

[0027] Based on the first and second micro-vibration velocity data, the micro-vibration data of the tool tip are quantified;

[0028] Predict the first micro-ripples to form on the workpiece surface after cutting based on micro-vibration data;

[0029] The specific steps for generating finishing parameters after workpiece cutting based on the micro-ripples on the workpiece surface image include:

[0030] Based on the micro-ripples on the workpiece surface in the workpiece surface image and the first micro-ripples formed on the workpiece surface, the finishing parameters after the workpiece cutting is completed are generated.

[0031] Through this technical solution, this application can acquire micro vibration data of the tool tip in real time, quantify micro vibration by combining it with simulation data, predict micro ripples on the workpiece surface, and generate finishing parameters based on the prediction results and actual images, thereby eliminating micro ripples more accurately and further improving machining accuracy and surface quality.

[0032] In one embodiment, the step of performing local hardness detection on the workpiece to obtain local hardness distribution data specifically includes:

[0033] An ultrasonic transducer is used to emit ultrasonic signals to the workpiece and to receive the ultrasonic signals that have penetrated the workpiece.

[0034] Based on the transmitted and received ultrasonic signals, the density and elastic modulus at different locations inside the workpiece are calculated.

[0035] Based on density and elastic modulus, the hardness at different locations inside the workpiece is calculated, and the local hardness distribution data of the workpiece is obtained.

[0036] Through this technical solution, this application can use ultrasonic technology to non-destructively detect the density, elastic modulus and hardness distribution inside the workpiece, thereby obtaining more accurate local hardness data, providing a reliable basis for the adjustment of subsequent processing parameters, and improving the accuracy and efficiency of hardness detection.

[0037] In another embodiment, the step of performing local hardness detection on the workpiece to obtain local hardness distribution data specifically includes:

[0038] A focused ultrasonic transducer is used to emit focused ultrasonic signals to the workpiece and receive ultrasonic signals scattered back from inside the workpiece.

[0039] Based on the emitted focused ultrasonic signal and the scattered ultrasonic signal, the density and distribution data of nanoscale hard particles inside the workpiece are calculated, and the local hardness distribution data of the workpiece is obtained.

[0040] Based on the workpiece's local hardness distribution data and initial machining parameters, the specific steps for simulating the cutting process of the workpiece and collecting the cutting force generated at each cutting point in real time during the cutting process include:

[0041] Based on the density and distribution data of nanoscale hard particles in the local hardness distribution data of the workpiece, the initial machining parameters, and the preset atomic-level geometric model of the tool tip, the cutting process of the workpiece is simulated. The transient cutting force formed at the point where the tool tip atoms collide with the hard particle atoms during the cutting process is collected in real time, thus obtaining the cutting force generated at each cutting point during the cutting process.

[0042] Through this technical solution, this application can detect the density and distribution of nanoscale hard particles by focusing ultrasonic waves, and combine them with atomic-level geometric models to simulate transient cutting forces during the cutting process, thereby gaining a more refined understanding of material properties and cutting mechanisms, and providing a more accurate basis for parameter adjustment for ultra-precision machining.

[0043] Furthermore, the initial machining parameters are adjusted based on the cutting force at each cutting point to generate the first machining parameters. The specific steps for cutting the workpiece using the first machining parameters include:

[0044] Based on the transient cutting force generated at the point where the cutting edge atoms collide with the hard particle atoms, calculate the transient impact force, impact duration, and impact frequency when the cutting edge contacts the hard particles;

[0045] The target area is defined as the area where the duration of the transient impact force generated when the tool tip comes into contact with hard particles exceeds a preset threshold is less than a preset time, and the impact frequency exceeds a preset frequency. Based on the location of the target area, the transient impact force, and the impact frequency, the parameters in the initial processing parameters used for the target area are adjusted to generate the first processing parameters. The first processing parameters include the piezoelectric drive signal of the tool tip used for the target area.

[0046] The workpiece is cut using the first machining parameters to suppress transient impact forces between the tool tip and hard particles.

[0047] Through this technical solution, this application can identify the target area based on the transient impact force, duration and frequency, and adjust the processing parameters accordingly, including using piezoelectric drive signals to suppress transient impact force, thereby effectively reducing the collision between the tool and hard particles, reducing tool wear and improving processing stability.

[0048] In some preferred embodiments, acquiring workpiece surface images after cutting specifically includes:

[0049] A linearly polarized laser beam of a set wavelength is emitted onto the surface of the workpiece after cutting.

[0050] Collect P-polarized and S-polarized light scattered from the workpiece surface from multiple angles;

[0051] Calculate the intensity ratio and phase difference between the P-polarized light and the S-polarized light scattered back at each angle;

[0052] The density and depth of submicron-level pits and / or scratches formed on the workpiece surface are calculated based on the intensity ratio and phase difference, and an image of the workpiece surface carrying submicron-level pits and / or scratches is generated based on the calculation results.

[0053] Through this technical solution, this application can use linearly polarized laser beams and multi-angle polarized light collection technology to accurately calculate the density and depth of submicron-level pits and scratches on the workpiece surface, thereby generating a high-precision surface image and providing more detailed and accurate surface defect information for the generation of finishing parameters.

[0054] Based on the above, the process of cutting the workpiece using the first machining parameters also includes the following steps:

[0055] Real-time acquisition of the machine tool spindle bearing speed, bearing geometric parameters, and the first cutting force at each first cutting point;

[0056] Based on the bearing speed, bearing geometric parameters, and the first cutting force at each first cutting point, calculate the local contact pressure and shear rate in the bearing contact area;

[0057] When the local contact pressure exceeds the preset pressure value and the shear rate exceeds the preset rate value, the process of depolymerization, repolymerization and formation of local insulating microgels in the lubricating oil is simulated, and the volume fraction and insulation characteristic data of the local insulating microgels are calculated.

[0058] Based on the volume fraction and insulation characteristics data, as well as the effective medium theory, the local conductivity data of the machine tool spindle bearing under disturbance were calculated.

[0059] Based on the disturbed local conductivity data, the local conductivity data collected during the cutting process of the workpiece using the first machining parameters is calibrated to obtain local conductivity calibration data.

[0060] The local conductivity data used in the step of determining whether there is early latent wear in the machine tool spindle bearing, based on local simulated conductivity data and local conductivity data, is the calibrated local conductivity calibration data.

[0061] Through this technical solution, this application can simulate the changes in nanoscale polymers in lubricating oil by real-time acquisition of bearing operating parameters and cutting force, and calculate the local conductivity data of the disturbed area. Then, the acquired conductivity data can be calibrated, thereby more accurately judging the early latent wear of machine tool spindle bearings and improving the reliability of wear diagnosis.

[0062] Preferably, after the step of calibrating the local conductivity data collected during the cutting of the workpiece using the first machining parameters based on the disturbed local conductivity data to obtain local conductivity calibration data, the method further includes:

[0063] The local conductivity calibration data is compared with the preset health threshold.

[0064] When the local conductivity calibration data exceeds the preset health threshold for a duration that reaches or exceeds the set time, the early latent wear risk level of the machine tool spindle bearing is assessed based on the magnitude of the local conductivity calibration data.

[0065] Based on the early latent wear risk level and the remaining time of the workpiece cutting process, a machining parameter adjustment strategy is generated. This strategy guides the system to progressively adjust the first machining parameter in real time during the subsequent cutting process of the workpiece.

[0066] Through this technical solution, this application can assess the bearing wear risk level by comparing the calibrated conductivity data with the health threshold, and generate a progressive real-time adjustment strategy for machining parameters based on the risk level and remaining machining time. This enables intelligent early warning and adaptive control of early hidden wear of machine tool spindle bearings, effectively avoiding sudden equipment failures and ensuring the continuity and safety of production.

[0067] Secondly, this application also discloses a gear forging process control system, comprising:

[0068] The detection module is used to detect the local hardness of the workpiece and obtain local hardness distribution data of the workpiece.

[0069] The simulation module is used to simulate the cutting process of the workpiece based on the local hardness distribution data and initial machining parameters, and to collect the cutting force generated at each cutting point in real time during the cutting process.

[0070] The cutting module is used to adjust the initial machining parameters according to the cutting force at each cutting point, generate the first machining parameters, and use the first machining parameters to cut the workpiece;

[0071] The generation module is used to acquire workpiece surface images after cutting and generate finishing parameters based on the micro-ripples on the workpiece surface in the workpiece surface images.

[0072] The finishing module is used to finish the workpiece after cutting using finishing parameters to ensure that micro-ripples on the workpiece surface are eliminated during the finishing process.

[0073] Through this technical solution, this application can achieve automated control of various processes such as local hardness detection of workpieces, cutting process simulation, machining parameter adjustment, surface image acquisition and finishing through modular design, thereby constructing an efficient and intelligent gear machining and forging process control system, which effectively improves machining accuracy and production efficiency.

[0074] Beneficial effects:

[0075] This application discloses a gear forging process control method that obtains local hardness distribution data of the workpiece by performing local hardness detection, and simulates the cutting process in conjunction with initial machining parameters, while also acquiring cutting forces in real time. The machining parameters are adjusted according to the cutting forces to generate first machining parameters for cutting. After cutting, an image of the workpiece surface is acquired, and finishing parameters are generated based on micro-ripples. Finally, the finishing parameters are used to eliminate micro-ripples on the workpiece surface. This method effectively solves the problems in existing technologies, such as high cutting resistance due to slight fluctuations in raw material hardness, hidden wear of machine tool spindle bearings, irregular micro-vibrations, and the formation of micro-ripples on the workpiece surface. By sensing the material properties of the workpiece and the dynamic changes during the machining process in real time, this application achieves adaptive adjustment of machining parameters, thereby significantly improving the machining accuracy and consistency of gear products, overcoming the limitations of traditional methods in dealing with complex working conditions, and providing more reliable quality assurance for high-end equipment manufacturing. Attached Figure Description

[0076] Figure 1 This application provides a schematic flowchart of a gear forging process control method.

[0077] Figure 2 This application provides a schematic diagram of a gear forging process control system.

[0078] Figure 2 In the diagram: 1 is the detection module; 2 is the simulation module; 3 is the cutting module; 4 is the generation module; and 5 is the finishing module. Detailed Implementation

[0079] The technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0080] See Figure 1 This application proposes a method for controlling the gear forging process, including:

[0081] S10. Perform local hardness detection on the workpiece to obtain local hardness distribution data of the workpiece;

[0082] S20. Based on the local hardness distribution data of the workpiece and the initial machining parameters, simulate the cutting process of the workpiece and collect the cutting force generated at each cutting point in real time during the cutting process.

[0083] S30. Adjust the initial machining parameters according to the cutting force at each cutting point to generate the first machining parameters, and use the first machining parameters to cut the workpiece;

[0084] S40. After the cutting is completed, acquire an image of the workpiece surface and generate finishing parameters based on the micro-ripples on the workpiece surface in the image.

[0085] S50. Use finishing parameters to finish the workpiece after cutting to ensure that micro-ripples on the workpiece surface are eliminated during the finishing process.

[0086] This application detects the local hardness of the workpiece and combines it with the cutting force feedback during the simulated cutting process to dynamically adjust the machining parameters. After the cutting is completed, the micro-ripples are identified by image recognition and finishing parameters are generated, ultimately achieving the finishing of the workpiece surface. This effectively solves the shortcomings of traditional methods in dealing with changes in material properties and equipment wear, and significantly improves machining accuracy and product quality consistency.

[0087] To better understand the gear machining and forging process control method proposed in this application, the following will elaborate on some key terms and operating environments involved.

[0088] "Workpiece" typically refers to the gear blank to be machined. Its material, dimensions, and initial state directly affect the subsequent machining process and the final product quality. "Local hardness detection" refers to the non-uniform measurement of hardness in different areas of the workpiece to obtain a detailed distribution of its internal material properties, which is crucial for subsequent cutting parameter adjustments. "Initial machining parameters" refer to parameters such as cutting depth, feed rate, and spindle speed set based on experience or standards before machining begins. "Cutting force" is the force exerted by the tool on the workpiece during the cutting process; its magnitude and changes reflect the dynamic characteristics of the cutting process and material removal. "First machining parameters" are machining parameters adjusted based on real-time cutting force feedback, used to optimize the cutting process. "Workpiece surface image" refers to workpiece surface morphology data acquired through optical or non-contact measuring equipment, used to identify microscopic defects. "Microscopic ripples" refer to tiny undulations or textures on the workpiece surface that are difficult to detect with the naked eye, and are a key factor affecting surface quality and subsequent finishing effects. "Finishing parameters" are generated based on microscopic ripple information and are used to guide the finishing process to eliminate surface defects. This method is typically implemented in an intelligent workshop environment that integrates sensors, data processing units, CNC machine tools, and finishing equipment.

[0089] The core of the gear forging process control method proposed in this application lies in achieving intelligent control of the gear processing process through a series of refined sensing, simulation, adjustment and finishing steps.

[0090] First, local hardness testing is performed on the workpiece to obtain local hardness distribution data. This step aims to obtain detailed information about the workpiece's material properties, as even workpieces from the same batch may exhibit subtle local differences in internal hardness. For example, a traditional Rockwell or Brinell hardness tester can be used to measure hardness at multiple preset points on the workpiece surface, and then an interpolation algorithm can be used to construct a local hardness distribution map of the workpiece. Alternatively, a portable ultrasonic hardness tester can be used to infer local hardness by measuring the propagation speed or attenuation of ultrasonic waves in the material. This method allows for rapid measurement without damaging the workpiece surface.

[0091] Secondly, based on the workpiece's local hardness distribution data and initial machining parameters, the cutting process is simulated, and the cutting forces generated at each cutting point during the cutting process are collected in real time. In this stage, finite element analysis (FEA) software can be used, with the detected local hardness data as material property input, combined with initial machining parameters (such as tool geometry, feed rate, depth of cut, etc.) to simulate the interaction between the tool and the workpiece. During the simulation, the software calculates and outputs the cutting force experienced by each cutting point at different times. For example, a virtual sensor array can be set up to simulate densely distributed cutting points along the cutting path and record the instantaneous cutting force data at each point.

[0092] Next, the initial machining parameters are adjusted based on the cutting forces at each cutting point to generate the first machining parameters, which are then used to cut the workpiece. If an abnormal increase in cutting force is detected at certain cutting points during the simulated cutting process, this may indicate that the material hardness in that area is too high or the cutting conditions are unsatisfactory. In this case, the control system will optimize and adjust the initial machining parameters based on this cutting force data. For example, if the cutting force in a certain area is too high, the system may reduce the feed rate or depth of cut in that area, or adjust the spindle speed to avoid excessive tool wear or workpiece surface damage. The adjusted parameters are the first machining parameters, which the machine tool will then use for actual cutting of the workpiece.

[0093] Subsequently, after cutting, an image of the workpiece surface is acquired, and finishing parameters are generated based on the micro-ripples on the workpiece surface in the image. After cutting, the workpiece surface quality needs to be evaluated. A high-resolution industrial camera with appropriate light source can be used to photograph the workpiece surface and acquire its image. Image processing algorithms can identify and quantify the micro-ripples present in the image, such as analyzing the periodicity and amplitude of the surface texture using Fourier transform. Based on the characteristics of these micro-ripples (such as peak height, wavelength, and distribution density), the system generates a set of finishing parameters. For example, if the micro-ripples are deep and dense, the finishing parameters may set a larger grinding allowance or a finer grinding path.

[0094] Finally, the workpiece is finished using finishing parameters to ensure the elimination of microscopic ripples on the workpiece surface. The finishing stage is crucial for improving workpiece surface quality. The machine tool performs grinding, polishing, or honing operations on the workpiece based on the previously generated finishing parameters. For example, if the finishing parameters indicate the need to eliminate deep ripples in a specific area, the finishing equipment precisely controls the tool or abrasive to perform multiple fine removal operations in that area until the surface achieves the required smoothness and precision.

[0095] The gear forging process control method proposed in this application forms a closed-loop intelligent control system by closely integrating the local hardness detection of the workpiece, the simulation of the cutting process and the real-time acquisition of cutting force, the dynamic adjustment of processing parameters, the analysis of the surface image after cutting, and the generation of finishing parameters.

[0096] Specifically, before machining begins, a localized hardness test is performed on the workpiece to obtain detailed data on the distribution of its internal material properties. This data provides accurate input for subsequent cutting simulations, making the simulation results closer to reality. For example, if a region of the workpiece is found to have excessively high hardness, the simulation system will predict that this region may generate greater cutting forces during machining.

[0097] Next, based on these local hardness distribution data and initial machining parameters, the cutting process of the workpiece is simulated. During the simulation, the cutting forces generated at each cutting point are collected in real time. This simulation can not only predict the distribution of cutting forces but also evaluate the impact of different machining parameters on cutting forces in a virtual environment. When the simulation results show that the cutting forces at certain cutting points exceed the preset range, the system will adjust the initial machining parameters based on these cutting force data to generate the first machining parameters. For example, if the simulation shows that the cutting force in a certain area is too large, the system may automatically reduce the feed rate or depth of cut in that area to avoid accelerated tool wear or workpiece surface damage during actual machining.

[0098] Subsequently, the machine tool performs actual cutting on the workpiece using the generated first machining parameters. Since the first machining parameters are dynamically adjusted based on the local hardness characteristics of the workpiece and the simulated cutting force, they can better adapt to the actual situation of the workpiece and effectively suppress any adverse phenomena that may occur during the cutting process, such as excessive cutting force or unstable cutting process.

[0099] After cutting, images of the workpiece surface are acquired. By analyzing these images, microscopic ripples on the workpiece surface are identified and quantified. These microscopic ripples are traces of the interaction between the tool and the workpiece during the cutting process, and their morphology and distribution directly reflect the cutting quality. Based on the characteristics of these microscopic ripples, the system generates a set of finishing parameters. For example, if image analysis shows deep periodic ripples on the workpiece surface, the finishing parameters may be set to use a specific grinding path and grinding depth to effectively remove these defects.

[0100] Finally, the generated finishing parameters are used to perform finishing on the workpiece after cutting. The goal of finishing is to completely eliminate micro-ripples on the workpiece surface, thereby achieving higher surface finish and precision requirements. In this way, the method of this application ensures that even if micro-defects are caused by differences in material properties or micro-vibrations of the equipment during the roughing stage, they can be effectively corrected in the subsequent finishing process, ultimately producing high-quality gear products.

[0101] The core innovation of this application lies in the introduction of "local hardness detection" and "cutting process simulation and real-time cutting force acquisition," enabling precise perception of workpiece material properties and dynamic prediction of the cutting process. By performing local hardness detection on the workpiece, this application can obtain detailed distribution data of the internal material properties of the workpiece. This allows the adjustment of machining parameters to no longer rely solely on average values, but to be optimized based on the actual local characteristics of the workpiece. For example, when a region of the workpiece is found to have excessively high hardness, the system can pre-adjust the cutting parameters for that region, avoiding potential problems caused by excessive cutting force.

[0102] Furthermore, this application achieves accurate prediction and real-time feedback of the cutting process by "simulating the cutting process of the workpiece based on the local hardness distribution data and initial machining parameters, and collecting the cutting force generated at each cutting point during the cutting process in real time." This simulation capability allows the system to predict the cutting force distribution before actual cutting occurs and adjust the initial machining parameters based on the prediction results to generate the first machining parameters. This contrasts sharply with traditional methods that rely solely on preset parameters or post-processing adjustments, greatly improving the predictability and control accuracy of machining.

[0103] Furthermore, this application, after the cutting process is completed, achieves refined evaluation of machining quality and intelligent guidance for the finishing process by "acquiring workpiece surface images and generating finishing parameters based on the micro-ripples on the workpiece surface in the images." Through high-resolution image analysis, this application can identify and quantify these micro-ripples, and generate customized finishing parameters accordingly, ensuring the complete elimination of these defects during the finishing process, thereby significantly improving the surface quality and precision consistency of the final product.

[0104] In some embodiments of this application described above, after simulating the cutting process of the workpiece based on the local hardness distribution data and initial machining parameters, and real-time acquiring the cutting force generated at each cutting point during the cutting process, the method further includes:

[0105] Based on the cutting force at each cutting point and the preset machine tool spindle bearing health model, the local conductivity change of the machine tool spindle bearing is simulated, and the local simulated conductivity data of the machine tool spindle bearing is collected in real time.

[0106] During the cutting process of the workpiece using the first machining parameters, the local conductivity data of the machine tool spindle bearing is collected in real time.

[0107] Based on local simulated conductivity data and local conductivity data, determine whether there is early hidden wear in the machine tool spindle bearing;

[0108] In cases where the machine tool spindle bearing exhibits early latent wear, the steps of acquiring an image of the workpiece surface after cutting and subsequent steps are performed.

[0109] Specifically, the pre-defined machine tool spindle bearing health model can be understood as a mathematical or simulation model established based on historical operating data, bearing design parameters, material properties, and lubrication conditions. Its purpose is to predict the normal local conductivity variation trend of the machine tool spindle bearing under specific cutting forces. This model can reflect the influence of factors such as bearing lubricating oil film thickness and contact area microstructure on local conductivity under ideal or healthy conditions. Local conductivity data refers to the conductivity values ​​of specific areas of the machine tool spindle bearing acquired in real time by sensors. These values ​​can sensitively reflect changes in local contact resistance caused by the integrity of the bearing lubricating oil film, the degree of lubricating oil contamination, or microscopic damage to the bearing surface (such as microcracks or pitting). In practical applications, the changing trends of local simulated conductivity data are compared with those of local conductivity data. If the local conductivity data shows persistent, weak, high-frequency current fluctuations above the background noise, and the changing trend of the local conductivity data is consistent with that of the local simulated conductivity data, then early latent wear of the machine tool spindle bearing is considered to exist. This indicates that the bearing may have begun to show abnormalities, such as lubrication failure, micro-wear, or fatigue damage. When early latent wear of the machine tool spindle bearing is determined, subsequent workpiece surface image acquisition and finishing steps will be conditionally performed. This means that if the health of the machine tool spindle bearing is poor, subsequent processing steps can be paused or adjusted to avoid producing defective workpieces, thereby effectively avoiding resource waste.

[0110] This application's solution addresses the issue of poor machine tool condition affecting machining quality by introducing a real-time monitoring and diagnostic mechanism for the health of machine tool spindle bearings. Specifically, before the cutting process begins, based on cutting force simulation results and a pre-set machine tool spindle bearing health model, the expected changes in local conductivity of the bearing under current cutting conditions can be predicted, thus obtaining simulated local conductivity data. During actual cutting of the workpiece using the first machining parameters, the local conductivity data of the machine tool spindle bearing is collected in real time. By comparing the real-time collected local conductivity data with the simulated local conductivity data, early latent wear of the machine tool spindle bearing can be effectively detected. This comparison mechanism can capture minute anomalies that occur in the bearing during actual operation, such as lubricating oil film rupture and changes in micro-contact points. These anomalies may not immediately manifest as macroscopic vibration or noise in the early stages, but they can be reflected through changes in conductivity. Once early latent wear of the machine tool spindle bearing is detected, the system will selectively execute subsequent workpiece surface image acquisition and finishing steps. This conditional execution strategy avoids unnecessary subsequent operations when the machine tool is not in a suitable condition for high-quality machining, thereby preventing the production of defective products and providing early warnings for machine tool maintenance.

[0111] Through the above technical solution, this application can effectively diagnose early latent wear of machine tool spindle bearings, thereby significantly improving the reliability and processing quality of gear forging. This solution avoids continuing subsequent processing when the machine tool is in poor condition, effectively reducing scrap rate and production costs. Furthermore, real-time monitoring of the machine tool spindle bearing health enables predictive maintenance of the machine tool, extending its service life and reducing unplanned downtime. This strategy of incorporating machine tool health status into the processing control makes the entire processing system more intelligent and robust, ensuring the stable production of high-quality gear products even in complex and variable processing environments.

[0112] In some embodiments of this application described above, the process of cutting the workpiece using the first machining parameters further includes the following steps:

[0113] Real-time acquisition of the first microscopic vibration velocity data of the tool tip;

[0114] After determining whether early latent wear exists in the machine tool spindle bearing based on local simulated conductivity data and local conductivity data, the following steps are also included:

[0115] When the machine tool spindle bearing has early latent wear, the micro-vibration generated by the machine tool spindle bearing and the tool clamping system is simulated, and the second micro-vibration velocity data of the tool tip is collected in real time during the simulation process.

[0116] The step of cutting the workpiece using the first machining parameters also includes:

[0117] Based on the first and second micro-vibration velocity data, the micro-vibration data of the tool tip are quantified;

[0118] Predict the first micro-ripples to form on the workpiece surface after cutting based on micro-vibration data;

[0119] The specific steps for generating finishing parameters after workpiece cutting based on the workpiece surface micro-ripples in the workpiece surface image include:

[0120] Based on the micro-ripples on the workpiece surface image and the first micro-ripples formed on the workpiece surface, the finishing parameters after the workpiece cutting is completed are generated.

[0121] Specifically, real-time acquisition of the first micro-vibration velocity data of the tool tip refers to acquiring the micro-vibration velocity information of the tool tip during the cutting process using a high-precision vibration sensor (such as a laser Doppler vibrometer or piezoelectric accelerometer) mounted on the tool or tool clamping system when the workpiece is being cut by the first machining parameter. This data reflects the dynamic response of the tool tip under actual cutting conditions.

[0122] In this simulation, when early latent wear is observed in the machine tool spindle bearing, the micro-vibrations generated by the machine tool spindle bearing and tool clamping system are simulated, and the second micro-vibration velocity data of the tool tip is acquired in real time during the simulation. This means that when a potential bearing problem is detected, the system runs a dedicated simulation model that considers factors such as bearing wear state, spindle speed, and cutting force to predict and generate possible micro-vibration modes and velocity data at the tool tip. The second micro-vibration velocity data is theoretical vibration data obtained based on the simulation results and is used for comparison and correction with the actually acquired first micro-vibration velocity data.

[0123] In practical applications, the micro-vibration data of the tool tip is quantified based on the first and second micro-vibration velocity data. This can be understood as fusing, analyzing, and processing actual measurement data and simulated prediction data to obtain a comprehensive and more accurate quantitative index characterizing the micro-vibration state of the tool tip. For example, by comparing the amplitude, frequency, phase, and other characteristics of the two sets of data, abnormal vibration components existing in the actual cutting process can be identified, or the simulated data can be corrected to better reflect the real situation.

[0124] Furthermore, the first micro-ripples formed on the workpiece surface after cutting are predicted based on the quantified micro-vibration data. This prediction process can be based on pre-established physical models or machine learning models that correlate the micro-vibration characteristics of the tool tip with the formation mechanism of micro-ripples on the workpiece surface. By inputting the quantified micro-vibration data, the model can output characteristics of the micro-ripples that may appear on the workpiece surface, such as the amplitude, wavelength, and distribution of the ripples.

[0125] Therefore, the specific steps for generating finishing parameters after workpiece cutting include generating finishing parameters based on the micro-ripples on the workpiece surface as seen in the workpiece surface image and the first micro-ripples formed on the workpiece surface. This means that when generating finishing parameters, the system no longer relies solely on the micro-ripples reflected in the actual workpiece surface image acquired after cutting, but rather comprehensively considers this actual observation data along with the first micro-ripples predicted based on micro-vibration data. This combination provides more comprehensive information; for example, the predicted micro-ripples can supplement potential blind spots or uncertainties in the image data, or be used to verify the accuracy of the image data, thus making the generated finishing parameters more targeted and accurate.

[0126] This application's solution addresses the problem of insufficient accuracy in predicting micro-ripples on the workpiece surface and generating finishing parameters when dealing with micro-vibrations caused by factors such as early latent wear of machine tool spindle bearings, by introducing real-time acquisition and simulation prediction of micro-vibration data at the tool tip. Specifically, when early latent wear exists in the machine tool spindle bearing, it may cause minute, difficult-to-observe vibrations in the spindle system, which are further transmitted to the tool tip, thus forming micro-ripples on the workpiece surface. Traditional solutions may only detect these ripples through surface images after cutting, but by then it is too late and the precise formation mechanism cannot be traced.

[0127] By acquiring the first micro-vibration velocity data of the tool tip in real time, the system can directly obtain the actual dynamic behavior of the tool tip during cutting. Simultaneously, when early latent wear of the bearing is detected, by simulating the micro-vibration generated by the machine tool spindle bearing and tool clamping system, and acquiring the second micro-vibration velocity data of the tool tip in real time, a theoretical micro-vibration prediction related to bearing wear can be obtained. Quantitatively fusing these two types of data allows for a more accurate characterization of the true micro-vibration state of the tool tip, including the micro-vibration component caused by bearing wear.

[0128] Based on this quantified micro-vibration data, the system can predict the first micro-ripples that may form on the workpiece surface after cutting. This prediction is forward-looking, providing information about potential surface defects before the actual surface image is acquired. Finally, when generating finishing parameters, the micro-ripples reflected in the actual acquired workpiece surface image are comprehensively considered together with the predicted first micro-ripples. This dual verification and information complementarity mechanism makes the generation of finishing parameters more comprehensive and accurate, and can more effectively guide the subsequent finishing process to eliminate micro-ripples caused by micro-vibrations.

[0129] Through the above technical solution, this application can identify and predict earlier and more accurately the impact of tool tip micro-vibration caused by factors such as early latent wear of machine tool spindle bearings on workpiece surface quality. This forward-looking predictive capability means that the generation of finishing parameters no longer relies solely on post-process observation, but incorporates dynamic microscopic information during the cutting process. Therefore, the accuracy and specificity of finishing parameters can be significantly improved, ensuring that the finishing process can more effectively eliminate microscopic ripples on the workpiece surface, thereby improving the final machining accuracy and surface quality of gears, reducing scrap rate, and extending the service life of tools and machine tools.

[0130] Specifically, the steps for performing local hardness testing on a workpiece to obtain local hardness distribution data include:

[0131] An ultrasonic transducer is used to emit ultrasonic signals to the workpiece and to receive the ultrasonic signals that have penetrated the workpiece.

[0132] Based on the transmitted and received ultrasonic signals, the density and elastic modulus at different locations inside the workpiece are calculated.

[0133] Based on density and elastic modulus, the hardness at different locations inside the workpiece is calculated, and the local hardness distribution data of the workpiece is obtained.

[0134] The ultrasonic transducer can be a piezoelectric ceramic transducer, capable of converting electrical energy into ultrasonic mechanical vibrations and vice versa. In practical applications, the transducer is placed on the surface of a workpiece and emits ultrasonic signals into the workpiece via contact or non-contact methods (e.g., through a coupling agent). As the emitted ultrasonic signal propagates within the workpiece, it is affected by factors such as the density, elastic modulus, and grain structure of the workpiece material, resulting in attenuation, reflection, and refraction. Once the ultrasonic signal penetrates the workpiece, it is received by the same or another ultrasonic transducer.

[0135] Furthermore, based on the emitted ultrasonic signal and the received ultrasonic signal after penetrating the workpiece, the density and elastic modulus at different locations within the workpiece can be calculated. For example, by measuring the propagation speed and attenuation of ultrasonic waves within the workpiece, acoustic theoretical models (such as elastic wave theory) can be used to inversely calculate parameters such as sound velocity and acoustic impedance at various points within the workpiece, thereby deriving the density and elastic modulus at that location. The propagation speed is closely related to the material's density and elastic modulus, while the attenuation reflects the microstructure and defects within the material.

[0136] Based on this, the hardness at different locations within the workpiece can be further calculated using the calculated density and elastic modulus. For example, empirical formulas or physical models can be used to establish a correlation between density and elastic modulus and the material's hardness (such as Vickers hardness, Rockwell hardness, etc.). These models are typically based on the principles of materials mechanics and solid-state physics, taking into account the influence of the material's microstructure on macroscopic hardness. By performing the above calculations at multiple locations within the workpiece, the local hardness distribution data of the workpiece can ultimately be obtained.

[0137] In materials science and engineering, a universal, single empirical formula or physical model for relating a material's density and elastic modulus to its hardness (such as Vickers hardness or Rockwell hardness) is generally unavailable for all types of materials. This is because material hardness is a complex mechanical property, influenced not only by the strength of atomic bonding (related to elastic modulus) and atomic arrangement (related to density), but also by a wide range of factors including the material's microstructure (such as grain size, grain boundaries, phase composition, and dislocation density), processing history (such as heat treatment and cold working), the types and amounts of alloying elements, and testing methods and conditions. Therefore, a simple formula cannot comprehensively capture these complex influences.

[0138] However, for specific types of materials, especially within a relatively narrow family of materials or alloy systems, empirical correlations with a certain range of applicability can indeed be established through a large amount of experimental data. These correlations are usually not based on simple algebraic formulas derived from universal physical laws, but rather on fitting data from experimental data using methods such as statistical regression analysis.

[0139] For example, for certain types of steel (such as alloy steel commonly used in gears), a positive correlation can be observed between the elastic modulus and hardness within a certain heat treatment state and composition range. While density is a fundamental property of materials, its direct and simple correlation with hardness is not as obvious as that of the elastic modulus. When establishing such empirical formulas, the following forms are usually considered:

[0140] 1. Empirical formula based on elastic modulus:

[0141] For some materials, hardness (H) may exhibit an approximately linear or nonlinear relationship with elastic modulus (E). For example, a relationship of the form `H = A * E + B` or `H = A * E` can be fitted. n The relationship is +B`, where A, B, and n are constants determined based on experimental data. These constants may vary for different steel grades or heat treatment states.

[0142] 2. Multivariable empirical formulas for density and elastic modulus:

[0143] In more complex models, if density has a significant impact on hardness in a specific material system, it can be incorporated into empirical formulas. For example, a polynomial relationship of the form `H = C1 + C2 * E + C3 * ρ + C4 * E * ρ + ...` can be established, where H represents hardness (e.g., HV Vickers hardness), E represents elastic modulus, ρ represents density, and C1, C2, C3, C4, etc., are coefficients determined through multiple regression analysis based on extensive experimental data. Establishing such a formula requires a wider range of experimental data covering different combinations of density and elastic modulus.

[0144] 3. Correlation based on physical models:

[0145] From a physical model perspective, hardness is more directly related to the microscopic mechanical parameters of a material, such as yield strength and shear modulus. For example, for metallic materials, hardness is usually proportional to yield strength (H ≈ 3 * σy, where σy is the yield strength). Yield strength is related to microscopic mechanisms such as dislocation motion and lattice resistance, which are indirectly affected by atomic bonding strength (elastic modulus).

[0146] In practical applications of gear forging process control, when this application mentions that "an empirical formula or physical model can be used to establish a correlation between density and elastic modulus and the hardness of the material," those skilled in the art will understand that: for a specific gear material being processed (e.g., a certain alloy steel), an empirical correlation model is established for that specific material system through extensive preliminary experimental testing and data analysis. This model may be a regression equation, the coefficients of which are calibrated based on the actual measurement data of the material. For example, for a specific gear steel, after knowing its density and elastic modulus, this calibrated empirical formula can be used to predict its local hardness value relatively accurately, thereby providing a basis for subsequent cutting process simulation and machining parameter adjustment.

[0147] This application utilizes the propagation characteristics of ultrasonic waves within materials to achieve non-destructive testing of localized hardness within a workpiece. The propagation speed and attenuation characteristics of ultrasonic signals within the workpiece directly reflect the material's physical properties, including density and elastic modulus. These physical properties are intrinsically related to the material's hardness. Specifically, after the ultrasonic signal emitted by the ultrasonic transducer penetrates the workpiece, its propagation time, waveform changes, and energy attenuation are received and analyzed. Through the analysis of these signals, the density and elastic modulus at different locations within the workpiece can be accurately calculated. Since the hardness of a material reflects its resistance to plastic deformation, and density and elastic modulus are key parameters determining the material's mechanical properties, a quantitative relationship between density and elastic modulus and hardness can be established. This quantitative relationship can be established through extensive experimental testing and data analysis, creating an empirical correlation model for this specific material system. Therefore, this method can infer the microscopic hardness distribution within the workpiece from macroscopic changes in ultrasonic signals, thereby acquiring localized hardness distribution data of the workpiece.

[0148] In the above-mentioned gear machining and forging process control method, the step of performing local hardness detection on the workpiece to obtain local hardness distribution data specifically includes:

[0149] A focused ultrasonic transducer is used to emit focused ultrasonic signals to the workpiece and receive ultrasonic signals scattered back from inside the workpiece.

[0150] Based on the emitted focused ultrasonic signal and the scattered ultrasonic signal, the density and distribution data of nanoscale hard particles inside the workpiece are calculated, and the local hardness distribution data of the workpiece is obtained.

[0151] Based on the workpiece's local hardness distribution data and initial machining parameters, the specific steps for simulating the cutting process of the workpiece and collecting the cutting force generated at each cutting point in real time during the cutting process include:

[0152] Based on the density and distribution data of nanoscale hard particles in the local hardness distribution data of the workpiece, initial machining parameters, and a preset atomic-level geometric model of the tool tip, the cutting process of the workpiece is simulated. The transient cutting force generated at the point of collision between the tool tip atoms and hard particle atoms during the cutting process is collected in real time, thus obtaining the cutting force generated at each cutting point during the cutting process. The atomic-level geometric model of the tool tip is a precise description of the tool tip at the atomic level. Traditional tool models usually focus on macroscopic geometry, but when cutting at the nanoscale or processing materials containing nanoscale hard particles, the microscopic features of the tool tip, such as atomic arrangement, crystal structure, interatomic spacing, and surface defects, become crucial. The atomic-level geometric model of the tool tip in this application is established through molecular dynamics (MD) simulation or other atomic-scale simulation techniques. It includes information such as the atomic type, atomic coordinates, and interatomic interaction potential function of the tool tip material, thereby accurately characterizing the microscopic morphology and mechanical response properties of the tool tip.

[0153] Nanoscale hard particles within a workpiece refer to atoms of hard phase particles with sizes on the nanometer scale distributed within the workpiece material matrix. These particles typically possess higher hardness and strength than the matrix material, such as carbides and nitrides. In the previous local hardness detection step, the density and distribution data of these nanoscale hard particles were calculated using a focused ultrasonic transducer. This data provides information on the spatial location and quantity of the particles within the workpiece.

[0154] The core of simulating the cutting process by combining a pre-defined atomic-level geometric model of the tool tip with the atomic model of nanoscale hard particles within the workpiece lies in constructing a multi-scale or atomic-scale simulation environment. In this environment, the atomic-level model of the tool tip is integrated with the atomic model of the workpiece material containing nanoscale hard particles. The simulation process typically employs molecular dynamics methods, which calculate the forces acting on each atom (based on the interatomic interaction potential function) and update the position and velocity of the atoms according to Newton's second law, thereby simulating the dynamic evolution of the material under cutting forces.

[0155] During simulated cutting, as the tool tip (defined by its atomic-level geometry) gradually approaches and comes into contact with the workpiece material (especially its nanoscale hard particles), the simulation system calculates the interaction forces between the tool tip atoms and the hard particle atoms in real time. This simulation of direct collision means that the simulation is no longer just a calculation at the level of macroscopic continuous medium mechanics, but is precise down to the atomic-atom interaction. When the atoms of the tool tip come into contact with the atoms of the hard particles, transient, highly localized impact forces are generated due to the repulsive forces between atoms. The simulation system monitors the distance and interaction potential energy between these atoms in real time, and once a critical value is reached, it considers a collision to have occurred and calculates the resulting transient cutting force.

[0156] This ability to simulate direct collisions is crucial for understanding and optimizing gear machining and forging processes. Traditional macroscopic cutting models often fail to capture these microscopic transient mechanical events, which are key factors leading to localized tool wear, micro-ripples on the workpiece surface, and even the initiation of microcracks within the material. Atomic-level simulations can precisely quantify the magnitude, duration, and frequency of these transient impact forces, providing a more refined and accurate basis for subsequent machining parameter adjustments. For example, simulation results can predict which areas of hard particles will violently collide with the tool tip, allowing for adjustments to cutting parameters in those areas, such as reducing feed rate, changing depth of cut, or introducing micro-vibrations to assist cutting, thereby mitigating impact, protecting the tool, and improving workpiece surface quality.

[0157] Specifically, a focused ultrasonic transducer is a device capable of emitting and receiving focused ultrasonic signals. By precisely controlling the propagation direction and focus of the ultrasonic waves, it concentrates ultrasonic energy into a specific micro-region within the workpiece. When the focused ultrasonic signal penetrates the workpiece and encounters internal nanoscale hard particles, scattering occurs. By analyzing the characteristics of the emitted focused ultrasonic signal and the ultrasonic signal scattered back from the workpiece, such as scattering intensity and frequency shift, the density and spatial distribution data of these nanoscale hard particles can be calculated. This data can more accurately reflect the hardness inhomogeneity of the workpiece material at the microscopic level, thus obtaining more refined local hardness distribution data. The preset atomic-level geometric model of the cutting tool tip can be understood as a precise description of the tool tip's geometry, crystal structure, and atomic arrangement at the atomic scale. This model, combined with the density and distribution data of nanoscale hard particles in the local hardness distribution data of the workpiece, can be used to simulate the interaction between the atoms at the tool tip and the atoms of the hard particles inside the workpiece at the atomic level. In practical applications, through molecular dynamics simulations or other atomic-level simulation techniques, the transient cutting force generated at the point of collision between the atoms at the tool tip and the atoms of the hard particles during the cutting process can be acquired in real time. This transient cutting force is a highly localized force with an extremely short duration. It can more realistically reflect the mechanical response at the microscopic level during the cutting process, providing a more accurate basis for subsequent machining parameter adjustments.

[0158] This application's solution, by introducing focused ultrasonic testing technology, can delve into the microstructure of workpiece materials, accurately identifying and quantifying the density and distribution of nanoscale hard particles. These nanoscale hard particles are key factors affecting the local hardness and cutting performance of materials, and traditional macroscopic hardness testing methods often struggle to capture their fine characteristics. By acquiring hardness distribution data at this microscopic level, more realistic and detailed material input can be provided for subsequent cutting process simulation. Furthermore, this application incorporates a pre-defined atomic-level geometric model of the tool tip, enabling the simulation of direct collisions between tool tip atoms and nanoscale hard particle atoms within the workpiece during the cutting process. This atomic-level simulation can capture the transient cutting forces generated at the moment of collision, which are key factors leading to the formation of micro-ripples on the workpiece surface and localized tool wear. This atomic-level collision simulation provides more accurate and real-time cutting force data, especially for areas of high local stress caused by microscopic inhomogeneities. It is precisely because such detailed transient cutting force data can be obtained that subsequent machining parameter adjustments can more precisely and specifically suppress these adverse effects at the microscopic level.

[0159] Through the above technical solution, this application can significantly improve the accuracy of acquiring local hardness distribution data of workpieces, especially in identifying and quantifying the density and distribution of nanoscale hard particles inside the workpiece, thereby providing more accurate material property input for cutting process simulation. Furthermore, by combining the atomic-level geometric model of the tool tip, this application can achieve accurate simulation and real-time acquisition of the transient cutting force generated at the point of collision between tool tip atoms and hard particle atoms during the cutting process. This microscopic mechanical analysis makes the prediction of cutting force more refined and realistic, especially when dealing with materials containing microscopic hard phases, effectively avoiding errors caused by insufficient macroscopic simulation. Therefore, this application can provide a more reliable and refined basis for subsequent machining parameter adjustments, thereby effectively suppressing the formation of microscopic ripples on the workpiece surface and extending tool life while ensuring machining efficiency.

[0160] Furthermore, the initial machining parameters are adjusted according to the cutting force at each cutting point to generate the first machining parameters. The specific steps of cutting the workpiece using the first machining parameters include:

[0161] Based on the transient cutting force generated at the point where the cutting edge atoms collide with the hard particle atoms, calculate the transient impact force, impact duration, and impact frequency when the cutting edge contacts the hard particles;

[0162] The target area is defined as the area where the duration of the transient impact force generated when the tool tip comes into contact with hard particles exceeds a preset threshold is less than a preset time, and the impact frequency exceeds a preset frequency. Based on the location of the target area, the transient impact force, and the impact frequency, the parameters in the initial processing parameters used for the target area are adjusted to generate the first processing parameters. The first processing parameters include the piezoelectric drive signal of the tool tip used for the target area.

[0163] The workpiece is cut using the first machining parameters to suppress transient impact forces between the tool tip and hard particles.

[0164] Specifically, transient impact force refers to the peak force generated by the mechanical action over an extremely short period of time when the tool tip and nanoscale hard particles inside the workpiece come into instantaneous contact and separate. Impact duration refers to the extremely short time interval from the onset to the end of this transient impact force. Impact frequency refers to the number of transient impacts between the tool tip and hard particles within a certain time. Calculating these parameters helps to quantify and characterize the dynamic interaction between the tool and hard particles.

[0165] The target area can be understood as a localized region on the workpiece where transient impact forces are particularly significant and may negatively impact machining quality. Specifically, when the tool tip contacts hard particles, if the resulting transient impact force exceeds a preset threshold, its duration is less than a preset time, and the impact frequency exceeds a preset frequency, then that region is identified as the target area. The preset threshold, preset time, and preset frequency are critical values ​​set based on experience, material properties, or experimental data, used to accurately identify areas requiring special treatment.

[0166] In practical applications, the first machining parameter is a set of optimized parameters used when cutting the target area, with the aim of effectively suppressing transient impact forces. For example, the first machining parameter may include a piezoelectric drive signal for the tool tip. The piezoelectric drive signal controls the piezoelectric actuator to cause the tool tip to vibrate or displace at a microscale, thereby changing the contact mode between the tool and hard particles, such as by microscopic avoidance or changing the cutting angle, to reduce or disperse transient impact forces. The parameters of the piezoelectric drive signal, such as frequency, amplitude, and phase, are dynamically adjusted according to the location of the target area, the transient impact force, and the impact frequency to achieve the best suppression effect.

[0167] This application's solution, through in-depth analysis of the transient cutting force generated at the point of collision between the tool tip atoms and hard particle atoms, can accurately calculate the transient impact force, impact duration, and impact frequency. The transient impact force refers to the peak value of the cutting force generated during a single collision between the tool tip and the hard particle. When time-series data F(t) of the transient cutting force is obtained through a high-frequency force sensor or atomic-level simulation, an impact event typically manifests as a rapidly rising and then rapidly falling mechanical pulse. To calculate the transient impact force, these independent impact events must first be identified. This can be achieved by setting a force threshold: when the transient cutting force F(t) rapidly rises from the baseline level and exceeds the preset force threshold, an impact event is considered to have begun; when the force reaches its maximum and then rapidly falls below the threshold, the impact event is considered to have ended. Within each identified impact event, the transient impact force is defined as the maximum peak force of that mechanical pulse. For example, if an impact event is detected between time t1 and t2, then the transient impact force F_impact = max(F(t)) for t in [t1, t2].

[0168] Impact duration refers to the effective length of time the transient impact force acts during a single collision between the tool tip and a hard particle. After obtaining time-series data of transient cutting forces and identifying independent impact events, impact duration can be defined as the time interval from when the impact force first exceeds a preset threshold (e.g., 10% or 20% of the peak impact value) to when it falls back below that threshold. More precisely, it can also be defined as the time from when the impact force begins to rise significantly (e.g., exceed the baseline noise level) to when it completely decays (e.g., returns to the baseline level). Due to the transient nature of atomic-level collisions, this time is typically very short, possibly on the order of nanoseconds to microseconds. For example, for an impact event starting at time t_start and ending at time t_end, the impact duration Δt = t_end - t_start.

[0169] Impact frequency refers to the average rate of occurrence of transient impact events between the cutting edge of a tool and hard particles over a certain time period. Calculating the impact frequency requires identifying all independent transient impact events within a relatively long observation window. Once all impact events have been identified and counted, the impact frequency can be calculated as follows:

[0170] The impact frequency f = (total number of impact events identified within the observation time window) / (length of the observation time window). For example, if 1000 independent transient impact events are identified within a 1-second observation period, then the impact frequency is 1000 Hz. This frequency reflects the density of hard particles encountered by the tool during cutting, or the combined effect of the distribution density of hard particles in the workpiece material and the tool cutting speed.

[0171] These detailed mechanical parameters enable the system to identify target areas with excessively large, short-duration, and high-frequency transient impacts—these areas are key to tool wear and surface defects. By adjusting initial machining parameters and generating first machining parameters incorporating piezoelectric drive signals to the tool tip for these target areas, the system can actively perform microscopic control of the tool tip. For example, through the piezoelectric drive signal, the tool tip can precisely displace or vibrate on microsecond or even nanosecond timescales, effectively suppressing transient impacts before the tool is about to violently collide with hard particles by microscopic avoidance, changing the local cutting angle, or dispersing impact energy. This avoids direct and severe impacts from hard particles on the tool tip during traditional cutting processes, thus solving the tool wear and machining quality problems caused by transient impacts.

[0172] Through the above technical solution, this application can achieve accurate identification and effective suppression of transient impact forces during gear machining and forging. Specifically, by quantifying the transient impact force, impact duration, and impact frequency, potential damage areas during cutting can be accurately located. Furthermore, by introducing a piezoelectric drive signal as part of the first machining parameter, active control of the tool tip at the microscopic level is achieved, thereby significantly reducing the transient impact force generated between the tool tip and hard particles inside the workpiece. This not only effectively extends the tool's service life and reduces the frequency of tool replacement, but also significantly improves the surface quality and machining accuracy of the workpiece, avoiding microscopic cracks or surface defects caused by transient impact forces, thereby improving overall machining efficiency and product reliability.

[0173] Specifically, acquiring workpiece surface images after cutting may include the following steps:

[0174] A linearly polarized laser beam of a set wavelength is emitted onto the surface of the workpiece after cutting.

[0175] Collect P-polarized and S-polarized light scattered from the workpiece surface from multiple angles;

[0176] Calculate the intensity ratio and phase difference between the P-polarized light and the S-polarized light scattered back at each angle;

[0177] The density and depth of submicron-level pits and / or scratches formed on the workpiece surface are calculated based on the intensity ratio and phase difference, and an image of the workpiece surface carrying submicron-level pits and / or scratches is generated based on the calculation results.

[0178] In this context, emitting a linearly polarized laser beam with a set wavelength refers to using a laser to generate a beam with a specific wavelength and a single polarization direction, and then illuminating the surface of a workpiece that has already been cut. This set wavelength is typically selected based on the microstructure dimensions of the surface to be inspected and the optical properties of the material to ensure effective interaction between the laser and the surface features. The use of linearly polarized laser beams helps to precisely control the polarization state of the incident light, providing a stable input for subsequent polarization analysis.

[0179] Furthermore, collecting P-polarized and S-polarized light scattered from the workpiece surface from multiple angles refers to receiving light reflected or scattered from the workpiece surface at different scattering angles by setting up multiple detectors or rotating detectors. P-polarized light refers to light whose electric field vector is parallel to the incident plane (the plane formed by the incident ray and the surface normal), while S-polarized light refers to light whose electric field vector is perpendicular to the incident plane. By separating and collecting these two polarization states of light, richer information about the surface structure can be obtained.

[0180] Therefore, calculating the intensity ratio and phase difference between P-polarized and S-polarized light scattered back at each angle involves measuring the light intensity of P-polarized and S-polarized light collected at different angles and calculating their intensity ratio (e.g., Rp / Rs) and phase difference (Δδ). These parameters are key indicators characterizing the interaction between light and surface microstructures, and their variations are closely related to the geometry and material properties of micro-features such as surface roughness, pits, and scratches.

[0181] Specifically, calculating the density and depth of submicron-level pits and / or scratches on a workpiece surface based on intensity ratio and phase difference refers to using a pre-established optical model or empirical relationship to invert the measured intensity ratio and phase difference data into physical parameters of the submicron-level pits and / or scratches on the workpiece surface. For example, scattering theory, effective medium theory, or numerical simulation methods can be used to correlate the optical response with parameters such as the density, average size, and depth of surface defects. Through this calculation, the distribution and severity of surface defects can be quantified.

[0182] Finally, generating a workpiece surface image with submicron-level pits and / or scratches based on the calculation results means visually presenting the density and depth information of the submicron-level pits and / or scratches obtained from the above calculations as a workpiece surface image. This image can clearly show the location, size, and distribution of surface defects, providing intuitive and accurate guidance for subsequent finishing processes.

[0183] This application's solution utilizes the principle of interaction between polarized light and the microstructure of a workpiece surface to achieve precise detection of submicron-level defects. This technical solution enables non-contact, high-precision detection of submicron-level defects on workpiece surfaces. By quantifying the density and depth of these submicron-level defects, extremely detailed and accurate surface condition information can be provided for the finishing process, ensuring that finishing can effectively eliminate these micro-defects and significantly improve the surface quality, wear resistance, and service life of gears.

[0184] This application further proposes the above-mentioned gear machining and forging process control method, which further includes the following steps during the cutting process of the workpiece using the first machining parameters:

[0185] Real-time acquisition of the machine tool spindle bearing speed, bearing geometric parameters, and the first cutting force at each first cutting point;

[0186] Based on the bearing speed, bearing geometric parameters, and the first cutting force at each first cutting point, calculate the local contact pressure and shear rate of the bearing contact area;

[0187] When the local contact pressure exceeds the preset pressure value and the shear rate exceeds the preset rate value, the process of depolymerization, repolymerization and formation of local insulating microgels in the lubricating oil is simulated, and the volume fraction and insulation characteristic data of the local insulating microgels are calculated.

[0188] Based on the volume fraction and insulation characteristics data, as well as the effective medium theory, the local conductivity data of the machine tool spindle bearing under disturbance were calculated.

[0189] Based on the disturbed local conductivity data, the local conductivity data collected during the cutting process of the workpiece using the first machining parameters is calibrated to obtain local conductivity calibration data.

[0190] The local conductivity data used in the step of determining whether there is early latent wear in the machine tool spindle bearing, based on local simulated conductivity data and local conductivity data, is the calibrated local conductivity calibration data.

[0191] Specifically, during the cutting process using the first machining parameters, it is necessary to collect the bearing speed, bearing geometric parameters, and the first cutting force at each first cutting point of the machine tool spindle in real time. The bearing speed can be monitored in real time using a speed sensor mounted on the spindle; the bearing geometric parameters refer to the inherent parameters of the bearing, such as its dimensions, clearance, and roller or ball diameters, which are usually known during bearing design or installation; the first cutting force at each first cutting point can be measured in real time using a force sensor mounted on the tool or workpiece fixture. These parameters are the fundamental data for evaluating the lubrication status and stress conditions of the bearing contact area.

[0192] Furthermore, based on the real-time acquired bearing speed, bearing geometric parameters, and the first cutting force at each first cutting point, the local contact pressure and shear rate of the bearing contact area can be calculated. Local contact pressure refers to the force per unit area borne by the bearing rollers or balls in contact with the inner and outer raceways, and can be calculated using Hertzian contact theory combined with actual loads. Shear rate refers to the velocity gradient of the lubricating oil under shear action within the bearing contact area, and is related to factors such as bearing speed, geometric parameters, and lubricating oil film thickness. These calculations help quantify the severity of the bearing's operating environment.

[0193] When the calculated local contact pressure exceeds a preset pressure value and the shear rate exceeds a preset rate value, it indicates that the bearing operating conditions are relatively harsh, and the lubricating oil may undergo significant changes. In this case, it is necessary to simulate the processes of depolymerization, recombination, and formation of locally insulating microgels in the lubricating oil. Nanopolymer depolymerization refers to the breakage of large molecular chains in the lubricating oil into smaller molecules under high temperature and high shear; recombination refers to the recombination of these smaller molecules under specific conditions to form new polymers; locally insulating microgels refer to gel-like substances with locally insulating properties formed by the lubricating oil under extreme conditions. These processes alter the dielectric properties and conductivity of the lubricating oil. By establishing corresponding physicochemical models, the volume fraction and insulating property data of the locally insulating microgels, such as dielectric constant and resistivity, can be calculated.

[0194] In practical applications, based on the calculated volume fraction and insulation properties of the local insulating microgels, combined with the effective medium theory, the disturbed local conductivity data of the machine tool spindle bearing can be calculated. The effective medium theory is a theory used to calculate the macroscopic physical properties of composite materials; here, it is used to calculate the overall conductivity of lubricating oil containing microgels, reflecting the contribution of changes in the lubricating oil's state to its conductivity.

[0195] Therefore, based on the disturbed local conductivity data obtained from the above calculations, the local conductivity data collected in real time during the cutting process of the workpiece using the first machining parameters can be calibrated to obtain local conductivity calibration data. The calibration process aims to subtract or correct the conductivity deviation caused by changes in the lubricating oil state from the original collected data, thereby obtaining conductivity data that more accurately reflects the bearing wear state.

[0196] Finally, in the step of determining whether the machine tool spindle bearing has early latent wear based on local simulated conductivity data and local conductivity data, the local conductivity data will be calibrated local conductivity data.

[0197] This application's solution addresses the problem of interference from changes in the lubricating oil itself in directly acquired local conductivity data, which is a common issue in traditional methods, by introducing a modeling and calibration mechanism for changes in lubricating oil conditions. Specifically, by real-time monitoring of bearing speed, geometric parameters, and cutting force, the system can accurately assess the local contact pressure and shear rate in the bearing contact area. These parameters are key factors inducing physicochemical changes in the lubricating oil. When these parameters reach critical values, the system simulates the depolymerization and repolymerization of nanoscale polymers in the lubricating oil, as well as the formation of local insulating microgels, and quantifies the impact of these changes on the lubricating oil conductivity, thereby obtaining disturbed local conductivity data. Subsequently, this disturbed data is used to calibrate the actually acquired local conductivity data, effectively eliminating conductivity deviations caused by changes in the lubricating oil itself. This allows the final calibrated local conductivity data used for wear assessment to more accurately reflect the true wear state of the machine tool spindle bearing.

[0198] Through the above technical solution, this application can significantly improve the accuracy and reliability of early latent wear detection in machine tool spindle bearings. By eliminating the interference of lubricating oil condition changes on conductivity measurement, the system can more accurately distinguish between conductivity changes caused by bearing wear and those caused by lubricating oil aging, thereby avoiding false alarms or missed alarms and ensuring accurate assessment of bearing health. This helps to promptly detect and address potential bearing problems, extend equipment lifespan, reduce maintenance costs, and ensure stable machining quality.

[0199] This application further proposes a scheme for processing the aforementioned local conductivity calibration data to achieve more refined management and preventive control of early latent wear of machine tool spindle bearings.

[0200] Specifically, after the step of calibrating the local conductivity data collected during the cutting process of the workpiece using the first machining parameters, based on the disturbed local conductivity data, the method further includes:

[0201] The local conductivity calibration data is compared with the preset health threshold.

[0202] When the local conductivity calibration data exceeds the preset health threshold for a duration that reaches or exceeds the set time, the early latent wear risk level of the machine tool spindle bearing is assessed based on the magnitude of the local conductivity calibration data.

[0203] Based on the early latent wear risk level and the remaining time of the workpiece cutting process, a machining parameter adjustment strategy is generated. This machining parameter adjustment strategy is used to guide the system to progressively adjust the first machining parameter in real time during the subsequent cutting process of the workpiece.

[0204] The "preset health threshold" can be understood as the upper limit of the local conductivity of the machine tool spindle bearing under normal health conditions. Its setting can be based on extensive historical operating data, machine tool design specifications, or expert experience. This threshold defines the normal and abnormal fluctuation ranges of the bearing's local conductivity. Furthermore, the "set time" refers to the length of time required for the local conductivity calibration data to continuously exceed the preset health threshold. Setting this duration is to avoid misjudgments caused by instantaneous fluctuations or measurement errors, ensuring that the detected anomalies are persistent and genuine early signs of latent wear. For example, the set time can be several seconds, tens of seconds, or longer, depending on the dynamic characteristics of the machine tool and the sensitivity requirements for wear response.

[0205] The "Early Latent Wear Risk Level" is a quantitative assessment of the potential wear severity of machine tool spindle bearings. This assessment considers not only whether the local conductivity calibration data exceeds a threshold, but also the duration and magnitude of this exceedance. For example, multiple risk levels can be set, such as "Low Risk," "Medium Risk," and "High Risk," each corresponding to different combinations of conductivity exceedance magnitude and duration. The classification of the "Early Latent Wear Risk Level" for machine tool spindle bearings is primarily based on the comparison between the local conductivity calibration data and a preset health threshold, as well as the duration of this comparison. The specific classification method is as follows:

[0206] First, the prerequisite for assessing the risk level of early latent wear is that the local conductivity calibration data must consistently exceed the preset health threshold for a duration equal to or exceeding the set time. If the local conductivity calibration data only briefly exceeds the preset health threshold but for a duration less than the set time, the system will classify it as normal fluctuation and will not conduct a risk level assessment.

[0207] Once the above prerequisites are met—that is, the local conductivity calibration data continuously exceeds the preset health threshold for a set time—the system will assess the early latent wear risk level of the machine tool spindle bearing based on the magnitude of the local conductivity calibration data (i.e., the extent to which it exceeds the preset health threshold). This application provides examples of the following three risk level classifications:

[0208] 1. Low Risk Level: When the local conductivity calibration data consistently exceeds a preset health threshold (e.g., set to X) for a set time (e.g., 5 seconds), and its value fluctuates between the preset health threshold X and 1.2 times X, the system will assess it as a "low risk" level. This indicates that the bearing has slight early signs of wear, but has not yet reached a serious level.

[0209] 2. Medium Risk Level: When the local conductivity calibration data consistently exceeds the preset health threshold X for a set time (e.g., 5 seconds), and its value fluctuates between 1.2 times X and 1.5 times X, the system will assess it as a "medium risk" level. This indicates that the early latent wear of the bearing is relatively high, requiring closer attention and appropriate intervention.

[0210] 3. High Risk Level: When the local conductivity calibration data continuously exceeds the preset health threshold X for a set time (e.g., 5 seconds), and its value exceeds 1.5 times X, the system will assess it as "high risk". This indicates that the early latent wear of the bearing has reached a relatively serious level, and immediate measures may be needed to avoid further damage or failure.

[0211] It should be noted that the preset health threshold X, its multiples (such as 1.2X, 1.5X), and the set time (such as 5 seconds) mentioned above are all exemplary values. In practical applications, these parameters can be adjusted and optimized according to specific machine tool models, bearing types, lubricant characteristics, processing requirements, and historical operating data to ensure the accuracy and practicality of the risk assessment. Through this graded assessment, the system can adopt differentiated response strategies for different levels of early latent wear risks, thereby achieving more refined predictive maintenance and processing control.

[0212] In practical applications, the "machining parameter adjustment strategy" refers to a series of instructions automatically or semi-automatically generated by the system to adjust the primary machining parameters based on different risk levels and the remaining time in the workpiece cutting process. This strategy aims to reduce bearing load and suppress wear development by progressively adjusting the primary machining parameters, such as cutting speed, feed rate, depth of cut, or toolpath, while maximizing machining quality and efficiency. For example, for low risk, the strategy might suggest fine-tuning the feed rate; for medium risk, it might suggest reducing the cutting speed and shortening the depth of cut; and for high risk, it might suggest pausing machining and performing an inspection.

[0213] This application's solution addresses the problem of insufficient response in existing solutions after detecting early latent wear by introducing a refined analysis and risk assessment mechanism for local conductivity calibration data. Specifically, comparing local conductivity calibration data with a preset health threshold clearly distinguishes between normal operating conditions and potential abnormal conditions. When local conductivity calibration data consistently exceeds the preset health threshold for a set period, it indicates that the early latent wear of the bearing is not an instantaneous phenomenon but a potential problem with a certain degree of persistence. Based on this, the risk level of early latent wear is assessed according to the magnitude of the local conductivity calibration data, enabling the system to quantify the severity of wear, thereby avoiding simple yes-or-no judgments and achieving a deeper understanding of the bearing's health status. Furthermore, by combining the early latent wear risk level with the remaining time in the workpiece cutting process to generate a machining parameter adjustment strategy, the system can dynamically and gradually adjust the first machining parameter according to the current risk level and the urgency of the machining task. This adjustment is not a one-size-fits-all stop or continue, but rather, through refined control, effectively suppresses further wear development while ensuring machining continuity, extends bearing life, and provides sufficient time for subsequent maintenance.

[0214] See Figure 2 The specific embodiments of this application also disclose a gear processing and forging process control system, including: a detection module 1, a simulation module 2, a cutting module 3, a generation module 4, and a finishing module 5.

[0215] Detection module 1 is used to detect the local hardness of the workpiece and obtain local hardness distribution data of the workpiece;

[0216] Simulation module 2 is used to simulate the cutting process of the workpiece based on the local hardness distribution data and initial machining parameters, and to collect the cutting force generated at each cutting point in real time during the cutting process.

[0217] The cutting module 3 is used to adjust the initial machining parameters according to the cutting force at each cutting point, generate the first machining parameters, and use the first machining parameters to cut the workpiece;

[0218] The generation module 4 is used to acquire workpiece surface images after cutting and generate finishing parameters for the workpiece after cutting based on the micro-ripples on the workpiece surface in the workpiece surface images.

[0219] The finishing module 5 is used to finish the workpiece after cutting using finishing parameters to ensure that micro-ripples on the workpiece surface are eliminated during the finishing process.

[0220] The gear processing and forging process control system provided in this application has a working principle and execution steps that are basically the same as the gear processing and forging process control method described above, and will not be repeated here.

[0221] The core innovation of this application lies in its modular design, which enables comprehensive perception, intelligent decision-making, and adaptive control of the machining process. The detection module 1 accurately acquires local hardness distribution data of the workpiece, breaking the reliance of traditional systems on average material properties. The simulation module 2 predicts the cutting process before actual cutting, allowing the system to evaluate and optimize machining parameters in a virtual environment, avoiding the lag in post-processing adjustments inherent in traditional systems. The collaborative work of the cutting module 3, generation module 4, and finishing module 5 ensures end-to-end quality control from roughing to finishing. In particular, the identification of micro-ripples by the generation module and the precise elimination by the finishing module effectively solve the problem of traditional systems' difficulty in detecting and correcting roughing defects. Therefore, the system of this application can significantly improve the machining accuracy, surface quality, and production efficiency of gear products, providing a more advanced and reliable solution for the high-end equipment manufacturing field.

[0222] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for controlling the gear forging process, characterized in that, include: Local hardness testing is performed on the workpiece to obtain local hardness distribution data. Based on the local hardness distribution data of the workpiece and the initial machining parameters, the cutting process of the workpiece is simulated, and the cutting force generated at each cutting point during the cutting process is collected in real time. The initial machining parameters are adjusted according to the cutting force at each cutting point to generate the first machining parameters, and the workpiece is cut using the first machining parameters. After the cutting is completed, an image of the workpiece surface is acquired, and the finishing parameters of the workpiece after the cutting is generated based on the micro-ripples on the workpiece surface in the image. The workpiece after cutting is finished by using finishing parameters to ensure that micro-ripples on the workpiece surface are eliminated during the finishing process. The step of simulating the cutting process of the workpiece based on the local hardness distribution data and initial machining parameters, and collecting the cutting force generated at each cutting point in real time during the cutting process, further includes: Based on the cutting force at each cutting point and the preset machine tool spindle bearing health model, the local conductivity change of the machine tool spindle bearing is simulated, and the local simulated conductivity data of the machine tool spindle bearing is collected in real time. During the cutting process of the workpiece using the first machining parameters, the local conductivity data of the machine tool spindle bearing is collected in real time. Based on local simulated conductivity data and local conductivity data, determine whether there is early hidden wear in the machine tool spindle bearing; In the case of early latent wear in the machine tool spindle bearing, perform the step of acquiring an image of the workpiece surface after cutting and the subsequent steps. The process of cutting the workpiece using the first machining parameters also includes the following steps: Real-time acquisition of the first microscopic vibration velocity data of the tool tip; The step of determining whether the machine tool spindle bearing has early latent wear based on local simulated conductivity data and local conductivity data also includes: When the machine tool spindle bearing has early latent wear, the micro-vibration generated by the machine tool spindle bearing and the tool clamping system is simulated, and the second micro-vibration velocity data of the tool tip is collected in real time during the simulation process. The step of cutting the workpiece using the first machining parameters is followed by: Based on the first and second micro-vibration velocity data, the micro-vibration data of the tool tip are quantified; Predict the first micro-ripples to form on the workpiece surface after cutting based on micro-vibration data; The specific steps for generating finishing parameters after workpiece cutting based on the micro-ripples on the workpiece surface image include: Based on the micro-ripples on the workpiece surface in the workpiece surface image and the first micro-ripples formed on the workpiece surface, the finishing parameters after the workpiece cutting is completed are generated.

2. The gear machining and forging process control method according to claim 1, characterized in that, The specific steps for performing local hardness testing on a workpiece to obtain local hardness distribution data include: An ultrasonic transducer is used to emit ultrasonic signals to the workpiece and to receive the ultrasonic signals that have penetrated the workpiece. Based on the transmitted and received ultrasonic signals, the density and elastic modulus at different locations inside the workpiece are calculated. Based on density and elastic modulus, the hardness at different locations inside the workpiece is calculated, and the local hardness distribution data of the workpiece is obtained.

3. The gear machining and forging process control method according to claim 1, characterized in that, The specific steps for performing local hardness testing on a workpiece to obtain local hardness distribution data include: A focused ultrasonic transducer is used to emit focused ultrasonic signals to the workpiece and receive ultrasonic signals scattered back from inside the workpiece. Based on the emitted focused ultrasonic signal and the scattered ultrasonic signal, the density and distribution data of nanoscale hard particles inside the workpiece are calculated, and the local hardness distribution data of the workpiece is obtained. Based on the workpiece's local hardness distribution data and initial machining parameters, the specific steps for simulating the cutting process of the workpiece and collecting the cutting force generated at each cutting point in real time during the cutting process include: Based on the density and distribution data of nanoscale hard particles in the local hardness distribution data of the workpiece, the initial machining parameters, and the preset atomic-level geometric model of the tool tip, the cutting process of the workpiece is simulated. The transient cutting force formed at the point where the tool tip atoms collide with the hard particle atoms during the cutting process is collected in real time, thus obtaining the cutting force generated at each cutting point during the cutting process.

4. The gear machining and forging process control method according to claim 3, characterized in that, The steps of adjusting the initial machining parameters based on the cutting force at each cutting point to generate the first machining parameters, and then using the first machining parameters to cut the workpiece specifically include: Based on the transient cutting force generated at the point where the cutting edge atoms collide with the hard particle atoms, calculate the transient impact force, impact duration, and impact frequency when the cutting edge contacts the hard particles; The target area is defined as the area where the duration of the transient impact force generated when the tool tip comes into contact with hard particles exceeds a preset threshold is less than a preset time, and the impact frequency exceeds a preset frequency. Based on the location of the target area, the transient impact force, and the impact frequency, the parameters in the initial processing parameters used for the target area are adjusted to generate the first processing parameters. The first processing parameters include the piezoelectric drive signal of the tool tip used for the target area. The workpiece is cut using the first machining parameters to suppress transient impact forces between the tool tip and hard particles.

5. The gear machining and forging process control method according to any one of claims 1 to 4, characterized in that, Acquiring images of the workpiece surface after cutting specifically includes: A linearly polarized laser beam of a set wavelength is emitted onto the surface of the workpiece after cutting. Collect P-polarized and S-polarized light scattered from the workpiece surface from multiple angles; Calculate the intensity ratio and phase difference between the P-polarized light and the S-polarized light scattered back at each angle; The density and depth of submicron-level pits and / or scratches formed on the workpiece surface are calculated based on the intensity ratio and phase difference. Based on the calculation results, generate an image of the workpiece surface carrying submicron-level pits and / or scratches.

6. The gear machining and forging process control method according to claim 1, characterized in that, The process of cutting the workpiece using the first machining parameters also includes the following steps: Real-time acquisition of the machine tool spindle bearing speed, bearing geometric parameters, and the first cutting force at each first cutting point; Based on the bearing speed, bearing geometric parameters, and the first cutting force at each first cutting point, calculate the local contact pressure and shear rate of the bearing contact area; When the local contact pressure exceeds the preset pressure value and the shear rate exceeds the preset rate value, the process of depolymerization, repolymerization and formation of local insulating microgels in the lubricating oil is simulated, and the volume fraction and insulation characteristic data of the local insulating microgels are calculated. Based on the volume fraction and insulation characteristics data, as well as the effective medium theory, the local conductivity data of the machine tool spindle bearing under disturbance were calculated. Based on the disturbed local conductivity data, the local conductivity data collected during the cutting process of the workpiece using the first machining parameters is calibrated to obtain local conductivity calibration data. The local conductivity data used in the step of determining whether there is early latent wear in the machine tool spindle bearing, based on local simulated conductivity data and local conductivity data, is the calibrated local conductivity calibration data.

7. The gear machining and forging process control method according to claim 6, characterized in that, After the step of calibrating the local conductivity data collected during the cutting process of the workpiece using the first machining parameters, based on the disturbed local conductivity data, the method further includes: The local conductivity calibration data is compared with the preset health threshold. When the local conductivity calibration data exceeds the preset health threshold for a duration that reaches or exceeds the set time, the early latent wear risk level of the machine tool spindle bearing is assessed based on the magnitude of the local conductivity calibration data. Based on the early latent wear risk level and the remaining time of the workpiece cutting process, a machining parameter adjustment strategy is generated. This machining parameter adjustment strategy is used to guide the system to progressively adjust the first machining parameter in real time during the subsequent cutting process of the workpiece.

8. A gear forging process control system for executing the steps of the method according to any one of claims 1 to 7, characterized in that, include: The detection module is used to detect the local hardness of the workpiece and obtain local hardness distribution data of the workpiece. The simulation module is used to simulate the cutting process of the workpiece based on the local hardness distribution data and initial machining parameters, and to collect the cutting force generated at each cutting point in real time during the cutting process. The cutting module is used to adjust the initial machining parameters according to the cutting force at each cutting point, generate the first machining parameters, and use the first machining parameters to cut the workpiece; The generation module is used to acquire workpiece surface images after cutting and generate finishing parameters based on the micro-ripples on the workpiece surface in the workpiece surface images. The finishing module is used to finish the workpiece after cutting using finishing parameters to ensure that micro-ripples on the workpiece surface are eliminated during the finishing process.

Images