Gear heat treatment deformation on-line monitoring and clamping posture self-adaptive adjusting system

Abstract

The invention discloses a gear heat treatment deformation online monitoring and clamping posture self-adaptive adjusting system, and belongs to the technical field of machine manufacturing and heat treatment intelligent control. The system comprises a real-time sensing module, an intelligent prediction module, a dynamic execution module and a central control module. According to the system, temperature, strain and displacement data of the gear in the heat treatment process are collected in real time, the virtual comparison model is driven to conduct synchronous simulation and deformation prediction, the space posture and clamping force distribution of the gear are dynamically adjusted through the multi-degree-of-freedom adjustable clamping platform based on the prediction result, and closed-loop control of prediction feedforward and real-time feedback is formed.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of intelligent control of mechanical manufacturing and heat treatment process, and particularly relates to a gear heat treatment deformation online monitoring and clamping posture self-adaptive adjustment system. BACKGROUND

[0002] Carburizing and quenching is a key process for improving the hardness and wear resistance of gear tooth surface, but the deformation of workpiece caused thereby has been a major problem restricting the precision and consistency of gears. In the heat treatment process, the uneven temperature field, phase transformation stress and cooling difference jointly act to cause complex deformation such as warping and eccentricity of the gear, which deteriorates the tooth profile precision formed by the pre-rolling process and brings about the problem of uneven allowance distribution in the subsequent grinding process.

[0003] At present, the industry mainly relies on optimizing process parameters and improving clamping methods to control deformation, such as using vertical mounting to improve the atmosphere and cooling uniformity. This kind of method belongs to "static prevention", although it has certain effect, but has obvious limitations: first, it cannot respond to the dynamic deformation occurring in the furnace in real time; second, it lacks continuous online monitoring of key states such as temperature, strain and displacement in the heat treatment process, and the deformation analysis relies on offline sampling, which is lagging; third, the clamping system itself is fixed and does not have the ability to adapt to the actual deformation state of the workpiece for self-adaptive posture adjustment, and cannot compensate for the deformation caused by structural asymmetry or cooling micro-variation.

[0004] Therefore, the existing process lacks initiative, adaptability and accuracy in deformation control, often forcing the previous process to reserve a large machining allowance, causing waste of materials and working hours. The industry urgently needs a system that can realize integrated control of real-time perception, online prediction and dynamic execution, and change the heat treatment from a "black box" process to a transparent process that can be controlled in real time. SUMMARY

[0005] The application provides a gear heat treatment deformation online monitoring and clamping posture self-adaptive adjustment system to solve the problems of lack of initiative, adaptability and accuracy in deformation control.

[0006] To solve the above technical problems, the summary is as follows:

[0007] A gear heat treatment deformation online monitoring and clamping posture self-adaptive adjustment system, comprising:

[0008] A real-time perception module: continuously collects temperature, internal stress and shape change data of the gear in the furnace;

[0009] An intelligent prediction module: a virtual comparative model of the gear is constructed in the computer, the real state of the gear in the furnace is simulated synchronously according to the real-time perceived data, and the deformation trend that may occur in the subsequent process is calculated in advance;

[0010] Dynamic execution module: contains a flexible motion clamping platform, receives instructions from the prediction module, and automatically adjusts the placement angle and position of the gear and the clamping force of each support point in real time;

[0011] Central control module: coordinates all the above modules. It combines the advanced prediction and real-time feedback information to form a closed-loop control system, dynamically instructs the execution mechanism to make adjustments, and actively suppresses deformation.

[0012] Preferably, the real-time sensing module comprises:

[0013] A plurality of infrared thermal imagers for acquiring temperature field distribution of the workpiece surface ;

[0014] A plurality of groups of distributed fiber grating sensors pre-buried in key cross sections of the gear or attached to the surface for measuring strain ; at least three laser displacement sensors for non-contact measurement of gear end face and addendum circle displacement ; sensor data is transmitted synchronously to the intelligent prediction module through wireless or wired mode.

[0015] Preferably, the three-dimensional thermal coupling model in the intelligent prediction module, the core control equation of which includes: heat transfer equation:

[0016]

[0017] wherein, is the density, is the specific heat capacity, is the thermal conductivity, is the latent heat term of phase change.

[0018] Thermoelastic constitutive equation:

[0019]

[0020] wherein is the stress increment, is the elastoplastic matrix related to temperature and plastic strain , is the total strain increment, is the thermal expansion coefficient, is the unit matrix.

[0021] The deformation prediction algorithm solves the above coupling equations, takes real-time monitoring data as boundary conditions, and iteratively calculates the deformation field in the future time .

[0022] Further, the intelligent prediction module adopts a data model hybrid driving strategy:

[0023] A library of deformation pattern features is constructed, and real-time monitored deformation data is compared with typical deformation patterns in the library Fast matching is performed, and similarity is calculated :

[0024]

[0025] When , a fast prediction based on feature patterns is started; otherwise, a complete three-dimensional coupled simulation is run.

[0026] The module continuously self-learns and optimizes the prediction model parameters according to historical process data.

[0027] Preferably, the dynamic execution module is a multi-degree-of-freedom adjustable clamping platform, and the adjustment strategy is based on the following optimization model:

[0028] Define the posture adjustment vector , including the tilt angle, translation amount, and multi-point clamping force;

[0029] Establish the objective function , aiming to minimize the deviation of the predicted deformation from the target geometry , while minimizing the change in clamping force:

[0030]

[0031] wherein is a weight coefficient, is the clamping force adjustment amount.

[0032] Solving the optimization problem, the optimal adjustment instruction is obtained, and the actuator is driven to act.

[0033] Preferably, the dynamic execution module is integrated into the vertical mounting module of the "carburizing and quenching gear vertical heat treatment and reference reconstruction integrated process":

[0034] The adjustable clamping platform replaces the fixed support assembly in the original process, and can dynamically adjust the verticality, circumferential deflection, and bottom support reaction force distribution of the gear during heat treatment.

[0035] The system applies differentiated posture control strategies at different stages such as carburizing, uniform heating, and quenching according to the predicted deformation trend.

[0036] Preferably, the central control module uses a fuzzy self-adaptive PID control algorithm:

[0037] The predicted deformation amount and real-time deformation amount and its rate of change as inputs

[0038] online tuning of PID controller's proportional, integral, and derivative parameters by fuzzy rules ;

[0039] controller's output drive the actuator, which in its discretized form is

[0040]

[0041] The algorithm can effectively deal with the control challenges brought by the strong nonlinearity and large time delay characteristics of the heat treatment process.

[0042] A gear heat treatment process optimization method adopts the online monitoring and clamping posture self-adaptive adjustment system, and includes the following steps:

[0043] S1: Real-time synchronous monitoring and modeling: In the gear heat treatment process, temperature, strain, and displacement data are synchronously collected, and a digital twin model is driven to perform real-time simulation.

[0044] S2: Deformation trend prediction and evaluation: Based on a hybrid driving strategy, the deformation amount in a future period of time is predicted, and it is evaluated whether it exceeds the allowable threshold ;

[0045] S3: Adaptive posture decision and adjustment: If the predicted deformation is out of tolerance, an optimization algorithm is started to calculate the optimal adjustment posture , and a control actuator is controlled to implement dynamic adjustment;

[0046] S4: Closed-loop verification and process learning: The actual deformation after adjustment is compared with the prediction result, the model parameters are corrected, and the process data of this time are stored in a knowledge base for autonomous optimization of subsequent processes.

[0047] Preferably, in step S3, a gradient compensation strategy is adopted during the quenching cooling stage:

[0048] According to the infrared thermal imager data, the cooling speed of each region of the gear is identified in real time ;

[0049] For regions with excessively fast cooling speed, the clamping force or heat dissipation conditions of the region are fine-tuned through an actuator, so that the cooling tends to be uniform, and the compensation amount is proportional to the cooling speed gradient:

[0050]

[0051] wherein is a material-related compensation coefficient.

[0052] A computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of a method of optimizing a gear heat treatment process.

[0053] The present application realizes active intervention and dynamic compensation of heat treatment deformation by constructing a closed-loop control system combining real-time monitoring, digital twin prediction and multi-degree-of-freedom adaptive clamping execution, and has the following significant advantages: the system can perceive the deformation trend in real time during the heat treatment process, and dynamically adjust the clamping posture and clamping force based on the prediction results, thereby inhibiting the deformation before or during its occurrence, breaking through the limitations of traditional static processes; significantly reducing the deformation dispersion of the gear after heat treatment, making the pre-grinding allowance more uniform and controllable, and providing a high-quality reference with better geometric consistency for subsequent finishing; due to effective control of deformation, the pre-grinding allowance in the hobbing stage can be moderately reduced, thereby shortening the gear grinding time, reducing the consumption of grinding wheels and energy consumption, and realizing the comprehensive optimization of processing efficiency, quality stability and manufacturing cost; the system continuously corrects the prediction model through closed-loop data and accumulates process knowledge, which can realize continuous self-optimization and process iteration, and adapt to the production requirements of different gear models and batches. BRIEF DESCRIPTION OF DRAWINGS

[0054] Figure 1 : overall architecture schematic diagram;

[0055] Figure 2 : working process and data-model hybrid driving strategy schematic diagram of the intelligent prediction module;

[0056] Figure 3 : structural schematic diagram of the dynamic execution module integrated with the vertical heat treatment tooling;

[0057] Figure 4 : principle and effect comparison schematic diagram of active inhibition of "tooth sag" deformation;

[0058] Figure 5 : adjustment principle schematic diagram of reference dynamic unification and following;

[0059] Figure 6 : data flow schematic diagram of group furnace collaborative monitoring and individualized adaptive grinding;

[0060] Figure 7 : precision alignment schematic diagram;

[0061] Figure 8 : product schematic Figure 1 ;

[0062] Figure 9 : product schematic Figure 2 ;

[0063] Figure 10 : route schematic Figure 1 ;

[0064] Figure 11 Route map Figure 2 . DETAILED DESCRIPTION

[0065] The present application is further described in the following examples, comparative examples, and performance test experiments, which do not limit the scope of the application claimed.

[0066] In the following description of embodiments, specific details are set forth, such as particular system configurations, techniques, etc., in order to provide a thorough understanding of embodiments of the application. However, persons having ordinary skill in the art will appreciate that embodiments of the application can be practiced without the specific details, or with other methods, components, materials, etc. In other instances, well-known structures, devices, circuits, and materials have not been described in detail in order to avoid obscuring the application.

[0067] It should be understood that the term "comprises" when used in this specification and the appended claims, specifies the presence of stated features, integers, steps, operations, elements, and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0068] It should also be understood that the term "and / or" when used in this specification and the appended claims, means any one or more of the associated listed items, and that includes one or more of the associated listed items.

[0069] As used in this specification and the appended claims, the term "if' can be construed to mean "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrase "if it is determined" or "if [a described condition or event] is detected" can be construed to mean "upon determining" or "in response to determining" or "upon detecting [the described condition or event]" or "in response to detecting [the described condition or event]," depending on the context.

[0070] In addition, the terms "first," "second," "third," etc. are used herein only to describe different instances of elements, and are not intended to imply or create an ordinality between elements.

[0071] Reference within the specification of this application to "one embodiment" or "some embodiments" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the application. The appearances of the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," and the like in the specification of the application do not necessarily refer to the same embodiment, unless otherwise noted. The terms "including," "comprising," "having," and variations thereof mean "including but not limited to," unless expressly specified otherwise.

[0072] Embodiment 1: Gear heat treatment deformation dynamic regulation principle example based on core algorithm model

[0073] Reference is made to the accompanying drawings Figure 1 , the implementation steps are as follows:

[0074] S1: Vertical mounting and initial process parameter calculation.

[0075] Reference is made to the accompanying drawings Figure 3 , the implementation steps are as follows:

[0076] The formula described in the claim is used: .

[0077] Based on historical data training, the system selects the optimization coefficient for this type of shaft gear: , the furnace size correction amount C = 5mm.

[0078] The calculation is: .

[0079] The system automatically adjusts the adjacent workpiece spacing to about 24mm accordingly, ensuring uniform atmosphere circulation and laying the foundation for active control.

[0080] S2: Online monitoring and data model hybrid driven prediction.

[0081] Reference is made to the accompanying drawings Figure 2 , at the end of the carburizing stage, the system monitors the following real-time data:

[0082] Infrared thermal imager: The average temperature of the upper end (near the spline end) of the gear is 918℃, and the lower end is 923℃, with a negative temperature difference (ΔT = 5K) of 5℃.

[0083] Fiber Bragg grating sensor: The average micro-strain detected in the middle of the gear width is +220με.

[0084] The system will deform monitoring data vector The input prediction module, with the "axis gear bowl deformation mode M1" in the feature library, performs similarity calculation. The dot product similarity S1=0.87 is calculated, which is higher than the threshold 0.8. Therefore, the system starts the fast prediction path based on the feature mode.

[0085] The prediction module combines temperature field data According to the thermoelastic constitutive relation, it is quickly deduced that during the upcoming quenching process, due to the existence of axial temperature gradient, the asynchronous cooling will cause the gear to produce about 0.08mm "bowl mouth" warping deformation (upper end concave), and the maximum deformation point is located on the upper end face of the gear ring.

[0086] S3: adaptive posture decision and fuzzy PID control execution.

[0087] The control module compares the predicted deformation (0.08mm) with the preset allowable threshold (0.03mm), and determines that active intervention is needed. The system establishes a posture adjustment optimization model:

[0088] Decision variables: additional clamping platform pressure force increment ΔF1 applied to the upper end face, and height fine adjustment amount Δh2, Δh3, Δh4 of the bottom three hydraulic support columns.

[0089] Objective function That is, while minimizing the predicted deformation, the change amplitude of the clamping force is constrained.

[0090] After solving, the optimal adjustment instruction is obtained: increase the additional pressure force of the upper end face by 50N, and simultaneously lift the bottom support columns (corresponding to the predicted deformation concave area) by 0.02mm.

[0091] The fuzzy adaptive PID controller receives this instruction. It takes the real-time monitored end face runout (initially 0.01mm) and its rate of change as input, dynamically adjusts the PID gain through the fuzzy rule matrix, and outputs a smooth control signal to drive the servo motor to accurately complete the above force and position adjustment within 15 seconds.

[0092] S4: gradient compensation during quenching process and effect verification.

[0093] Entering the oil quenching stage, the real-time monitoring and display of the infrared thermal imager shows that the cooling speed of one side of the gear is about 15℃ / s faster than the other side. The system immediately starts the gradient compensation strategy, according to the formula For this material (17CrNiMo6), the system calls the compensation coefficient η=0.003mm·s / ℃ in the database.

[0094] The compensation amount C=0.00315=0.045mm is calculated.

[0095] The system controls the adjustable guide vane to partially shield the side that cools faster, adjusts the quenching oil flow field, and makes the cooling tend to be uniform.

[0096] After the process, through off-line three-coordinate detection, the radial runout of the gear after quenching is 0.025 mm, and the cumulative error of the pitch is 0.055 mm. Compared with the data of the same type of gear in the original process, the radial runout after quenching is 0.045-0.065 mm, and the cumulative pitch is 0.037-0.075 mm, which shows obvious advantages in controlling deformation dispersion. The whole process data is encrypted and stored for further iteration optimization of the prediction model of this type of gear.

[0097] Example 2 is applied to the heat treatment homogeneity control of a large modulus planetary gear of a shield machine

[0098] S1: High-density sensor network arrangement and initial state modeling.

[0099] Reference is made to the accompanying drawings Figure 8 and 10 In view of the high value and large modulus of the gear, before heat treatment, 8 high-temperature fiber Bragg grating sensors are pre-embedded in the key stress concentration areas such as the tooth root fillet and the spoke plate. After vertical mounting, the initial geometric point cloud in the clamped state is obtained through 3D scanning, aligned with the CAD model, and a high-fidelity digital twin is established. The initial detection normal W is 174.674 mm (process requirement).

[0100] S2: Stress accumulation monitoring and early warning during long-period carburizing process.

[0101] During the long carburizing diffusion stage (the total time of the original process is more than 70 hours), the sensor network works continuously. In the late stage of the second strong carburizing stage (930℃, 25 hours), the system monitors that the heavy spoke side of the asymmetric structure of the gear has a compression creep strain that is 35% higher than the other side. The digital twin model simulates and predicts that after the final quenching, a non-negligible eccentric deformation will be generated, which may cause the pre-grinding radial runout to be out of tolerance (the pre-grinding radial runout of the No. 336 workpiece in the original specification is 0.16 mm).

[0102] S3: Self-adaptive clamping force dynamic distribution based on stress state.

[0103] Before high-temperature tempering (650℃, 6 hours), the system executes the active intervention strategy. The adjustment target is not to change the macro position of the gear, but to dynamically adjust the radial clamping force distribution on the gear rim by controlling the current of 12 independent electromagnetic clamping units integrated in the vertical mounting tool.

[0104] On the side of the heavy spoke where higher stress was monitored, the clamping force was reduced from the standard value of 200 N to 150 N, allowing for more complete stress relaxation at high temperatures in this area.

[0105] On the other side of the symmetry, the clamping force was increased to 250 N, providing a counterbalancing constraint.

[0106] This adaptive clamping strategy, combining "loose-tight", guides the micro-creep of the metal at high temperatures, leading to a more uniform internal stress distribution, and thus reducing the driving force for subsequent quenching distortion at the source.

[0107] S4: Precise shape control during the salt quenching process.

[0108] In the final salt quenching stage (830°C, salt bath quenching), in addition to monitoring the temperature field, the system focuses on monitoring the phase transformation process through the pre-embedded sensors. When the system identifies the start of martensitic transformation on the tooth surface through the strain inflection point, it immediately activates the "pulse-like micro-stirring" program. The salt bath stirrer is controlled to intermittently increase stirring at a specific frequency, breaking the vapor film and ensuring uniform cooling of the tooth part; at the same time, the gear body part is maintained under relatively gentle cooling conditions to reduce thermal stress. This zoned cooling strategy achieves the coordinated control of high hardness on the tooth part and low stress on the body.

[0109] Implementation results: After applying this system, the 10 planetary gear shafts in this batch after quenching, the pre-grinding radial runout was detected to be stable within the range of 0.06-0.10 mm (there was an abnormal value of 0.16 mm in the original process data), and the cumulative error of the pitch was ≤0.12 mm. The "exposed amount" in the grinding process remained stable, effectively avoiding the need for repair or scrap due to excessive deformation of individual parts, providing a guarantee for the batch production stability of this high-value product.

[0110] Example 3: Solving the transmission error transmission error of locomotive pinion shaft

[0111] Referring to the attached Figure 5 , the problem of long time for circle alignment and excessive runout in the carbon layer process of numerical control turning, leading to poor precision of the tooth part after quenching, is solved.

[0112] S1: Integrated intelligent alignment and reference data binding.

[0113] Referring to the attached Figure 9 and 11 , before the carbon layer process, the workpiece is hoisted onto the lathe integrated with the reference alignment module of the present invention. The operator does not need to manually use the circle bar to take measurements.

[0114] The system controls the three radial measurement units of the measurement platform to extend synchronously, and the measuring head automatically abuts against the near-tooth-space circle.

[0115] High-precision displacement sensors collect data, and the controller calculates the eccentricity e and the tilt angle θ in real time. According to the gear parameters (Z = 17, n = 4), the system automatically calculates the eccentricity e and the tilt angle θ according to the formula Four tooth spaces are measured in the circumferential direction, and the whole circle is evaluated twice.

[0116] Within 30 seconds, the system automatically adjusts the gear to an eccentricity e ≤ 0.015 mm and a tilt angle θ ≤ 0.001 rad, and locks it. At the same time, the system binds the "software reference" in this state with the coordinates of the center hole at both ends to be turned and repaired.

[0117] S2: Reference following and real-time compensation in the finishing turning process.

[0118] Reference the attached Figure 2 When turning φ128 and other external circles, the traditional method relies on the worn center support to support the machined surface. This system is different:

[0119] At the support roller where the center support contacts the workpiece, a micro-force sensor and a micro-displacement actuator are built in.

[0120] During the turning process, the reference alignment module continuously monitors the real-time position of the tooth pitch circle reference (virtual axis) at a frequency of 10 Hz.

[0121] Once the virtual axis is found to have deviated by more than 2 μm due to cutting force or thermal deformation, the controller immediately calculates the compensation amount and drives the actuator of the center support to adjust the position of the support roller, so that the rotation axis of the workpiece always follows the "tooth pitch circle reference axis".

[0122] This achieves the dynamic unification of the machining reference (center hole center) and the design reference (tooth pitch circle), rather than static conversion.

[0123] S3: Data closed loop and process optimization.

[0124] After finishing turning, the system automatically re-measures the tooth runout and records it. The batch data shows that after finishing turning, when positioning with the center holes at both ends, the tooth pitch circle runout at points A and B is stable within 0.02 mm (the original process requirement is ≤ 0.05 mm, and it often exceeds the tolerance).

[0125] The system correlates and analyzes the "finishing runout value" with the "pre-carbon layer runout value" and the "post-heat treatment runout value" through big data analysis. Analysis shows that when the pre-carbon layer heat treatment deformation is within a certain range, through the finishing process of this system, the final pre-grinding runout can be predicted and controlled within a very narrow range. Based on this, the system suggests to the process database: for such gears, when the heat treatment deformation is below a certain threshold, it is possible to reduce the grinding allowance of the hobbing normal line by 0.1 mm.

[0126] Embodiment 4: Group furnace collaborative optimization and adaptive grinding for batch petrochemical idler wheels

[0127] Referring to the accompanying drawings Figure 6 , applications in multi-workpiece simultaneous processing and big data learning are demonstrated.

[0128] S1: Group furnace loading and overall monitoring strategy.

[0129] Referring to the accompanying drawings Figure 6 , 6 pieces of left and right box idlers, a total of 12 pieces, are carburized and quenched in the same furnace. The system is expanded to a "one master and multiple slave" architecture:

[0130] One master module monitors the overall atmosphere and temperature field of the entire furnace.

[0131] Each vertical loading rack for the gear integrates a simplified sensing unit (2 temperature measurement points and 1 key cross-section strain gauge per piece) and a micro-actuator (only supporting fine adjustment function), forming an "intelligent clamping slave station".

[0132] All slave station data are wirelessly aggregated to the master module.

[0133] S2: Early abnormality identification and adjustment based on statistical process.

[0134] During the strong penetration stage at 920℃ (25 hours), the master module analyzes the distribution of the 12 workpiece tooth surface temperatures in real time. The system finds that the temperature of "right box 06" is always about 8℃ lower than the average furnace temperature, with large fluctuations. The digital twin model deduces that this may lead to insufficient penetration depth and deformation pattern different from the same batch.

[0135] The system does not directly adjust this workpiece (as it may interfere with the carburizing atmosphere), but marks it as a "key attention piece" and assigns a higher monitoring frequency to its slave station. At the same time, the master module fine-tunes the circulating fan speed in the area where the workpiece is located, trying to improve the local atmosphere uniformity.

[0136] S3: Big data analysis and adaptive grinding programming after quenching.

[0137] After quenching and shot peening of this batch of workpieces, the system performs rapid automatic laser scanning on the 12 gears before unified numerical control fine turning, obtaining the tooth top circle and end face three-dimensional topography data of each gear.

[0138] The system analysis finds that the left box gears generally exhibit slight "taper" deformation (0.05-0.08mm larger at one end than the other), while the right box gears mainly exhibit "saddle" deformation (about 0.03-0.06mm concave in the middle). This is related to the mirror symmetry of left and right box gear structures but different loading positions in the furnace, and the data are recorded in the "product deformation pattern" knowledge base.

[0139] In the subsequent gear grinding process, the system no longer uses a unified grinding program. After the operator clamps the workpiece, the system automatically retrieves the three-dimensional scanning data of the workpiece and compares it with the standard model to generate a personalized grinding allowance distribution map.

[0140] The CNC system of the gear grinding machine receives this distribution map. For example, for the "left box No. 03" workpiece, the system automatically assigns a higher rough grinding feed rate to the end with larger deformation in the grinding program and performs additional shape correction cycles in the fine grinding stage to achieve efficient grinding by "changing to change".

[0141] S4: Full life cycle process map generation.

[0142] After the 12 gears in this batch are completely processed, the system generates a "digital process passport" for each gear, including the temperature-strain curve throughout the heat treatment process, key adjustment instructions, geometry data after finish machining, pre-grinding allowance three-dimensional map, and final accuracy test report. These structured data provide extremely valuable precise input for the process development of subsequent products of the same type, transforming traditional "trial and error" optimization into "prediction and verification" precision iteration, significantly improving the maturity and reliability of batch manufacturing.

[0143] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A system for online monitoring and adaptive adjustment of gear heat treatment deformation, characterized in that, include: Real-time sensing module: continuously collects data on temperature, internal stress, and shape changes of the gears inside the furnace; Intelligent prediction module: A virtual comparison model of the gear is built in the computer. Based on real-time sensed data, the actual state of the gear in the furnace is simulated synchronously, and the deformation trend that may occur in subsequent processes is calculated in advance. Dynamic execution module: includes a flexible clamping platform that receives instructions from the prediction module and adjusts the gear placement angle, position, and clamping force at each support point in real time and automatically. Central control module: Coordinates all the above modules. It combines advance prediction with real-time feedback information to form a closed-loop control system, dynamically directing the actuators to make adjustments and proactively suppressing deformation.

2. The gear heat treatment deformation online monitoring and clamping posture adaptive adjustment system according to claim 1, characterized in that, The core simulation capabilities of the intelligent prediction module are as follows: The virtual comparison model of this module strictly follows the physical properties of gear materials and simulates the heat transfer process in the gear through calculation. At the same time, it combines the laws of thermal expansion and deformation of materials under stress to calculate the internal stress and deformation caused by uneven temperature and material structure transformation. The system uses real-time collected data as input conditions for model calculation and predicts the specific deformation of each point on the gear in the future through iterative calculation.

3. The gear heat treatment deformation online monitoring and clamping posture adaptive adjustment system according to claim 2, characterized in that, The intelligent prediction module employs the following strategy: The system has pre-learned and stored a library of various typical gear deformation cases. During operation, it quickly compares the deformation data monitored in real time with the patterns in the case library. If a highly similar case is found, the system directly calls the pattern of that case for rapid prediction, which greatly shortens the calculation time. If no similar case is found, the system starts a complete and high-precision 3D simulation calculation.

4. The gear heat treatment deformation online monitoring and clamping posture adaptive adjustment system according to claim 1, characterized in that, The adjustment decisions of the dynamic execution module are based on intelligent optimization algorithms: When adjustments are needed, the system will comprehensively consider multiple adjustment methods to minimize the predicted deformation of the adjusted gear and make it close to the ideal shape; and minimize the variation of the clamping force to make the adjustment process smooth and reliable. The system automatically obtains the optimal adjustment scheme by solving this multi-objective optimization problem.

5. The gear heat treatment deformation online monitoring and clamping posture adaptive adjustment system according to claim 1, characterized in that, The central control module employs an intelligent adjustment algorithm to handle complex processes: The heat treatment process is characterized by nonlinearity and hysteresis. The control algorithm of this system dynamically adjusts its control "strength" and "rhythm" based on the prediction error, real-time error and their changing trends. It comprehensively considers the current deviation, the accumulation of past deviations and the trend of deviation changes, and outputs control commands to effectively cope with the complex working conditions of heat treatment.

6. The gear heat treatment deformation online monitoring and clamping posture adaptive adjustment system according to claim 1, characterized in that, The dynamic execution module can be directly used as the core driving component of the vertical tooling, dynamically and precisely adjusting the verticality of the gear, the circumferential sway, and the distribution of the bottom support force throughout the carburizing and quenching process.

7. A method for optimizing gear heat treatment process based on the above system, characterized in that, Includes the following steps: Step 1: Real-time monitoring and synchronous modeling: During the heat treatment process, various data are collected synchronously, and the virtual comparison model is driven to perform real-time simulation, so that the digital simulation keeps pace with the current state; Step 2: Early Warning and Deformation Assessment: Using the aforementioned prediction strategy, determine whether the workpiece will deform beyond tolerance in the subsequent process, thus achieving early warning of quality problems; Step 3: Intelligent decision-making and dynamic correction: Once it is predicted that the deformation will exceed the tolerance, the system immediately starts the optimization algorithm, calculates the best adjustment plan, and directs the execution mechanism to actively intervene and correct during the process; Step Four: Closed-Loop Learning and Process Evolution: Compare the actual adjustment results with the predicted values ​​to correct and optimize the prediction model parameters. All process data is stored in a knowledge base, enabling the system and process solutions to continuously learn and upgrade.

8. The method according to claim 7, characterized in that, In the crucial quenching and cooling stage, the method further includes: Based on the monitoring of uneven cooling rates in different areas of the gear surface, the actuator dynamically fine-tunes the local clamping state or guides heat dissipation to provide targeted "compensation" for areas with uneven cooling. The degree of compensation is proportional to the degree of difference in cooling rate in that area, thereby actively promoting uniform overall cooling and reducing deformation.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements all the steps of the gear heat treatment process optimization method as described in claim 7 or 8.

Images