High-precision speed reducer gear key groove machining method

Abstract

The invention relates to a high-precision speed reducer gear key groove machining method. The method comprises the following steps that S100, heat treatment deformation is predicted; s200, groove broaching before heating; s300, heat treatment is conducted; s400, detecting after heating; s500, finish milling is conducted after heating; through a heat treatment deformation prediction model, after-heat deviation data are predicted in advance at the front end of the process, heat treatment parameters are optimized, so that heat deformation is controllable, a compensation reserved deformation compensation amount is provided for pre-heat groove broaching, a combined process is formed, the pre-heat groove broaching can be matched with a heat deformation prediction result to carry out deformation compensation, and the heat treatment quality is improved. A workpiece has a targeted compensation structure before entering heat treatment, and then the cutting amount is corrected through a cutting parameter optimization model in combination with pre-heat compensation and post-heat detection, so that the problem of thermal deformation uncontrollability in the key groove machining process is converted into predictable, compensable and correctable process variables; the technical bottlenecks that the symmetry degree is difficult to guarantee, the matching precision fluctuation is large and the machining efficiency is low in traditional gear key groove after-heating hard machining are broken through.

Description

High-precision gear reducer keyway machining method Technical Field

[0001] This invention relates to the field of gear keyway machining technology, and more specifically to a high-precision gear keyway machining method for reducers. Background Technology

[0002] In the process of modern high-end equipment manufacturing moving towards higher precision and higher reliability, high-precision speed reducers, as core components of industrial transmission systems, directly determine the stability and reliability of equipment operation based on their performance. Gear keyways, as key structures for torque transmission, play a decisive role in the meshing transmission accuracy of speed reducers due to their symmetry and fit precision. Insufficient keyway symmetry or poor fit precision can lead to uneven force distribution during gear meshing, generating vibration and noise, reducing transmission efficiency, and even causing premature gear failure, severely hindering the development of my country's high-end equipment manufacturing industry.

[0003] Traditional machining processes mainly employ post-heat wire EDM for keyways, where the keyway is machined after heat treatment using wire EDM, with the material being etched by an electrode wire. Alternatively, pre-heat broaching can be used, where a keyway broach is used before heat treatment to broach and remove material along the axial direction to form the keyway.

[0004] When hard machining keyways in gears after heat treatment, it is difficult to effectively guarantee the symmetry and fit accuracy of the keyways. Wire EDM of keyways after heat treatment may also lead to a decrease in keyway hardness, thereby reducing its wear resistance and fatigue strength, resulting in premature failure of the gear pair. Furthermore, wire EDM is a point-by-point electrical discharge machining process, which has a slow cutting speed and low efficiency in mass production when dealing with high-hardness gear materials.

[0005] If broaching is performed before heat treatment, the workpiece will deform due to the transformation of the material structure and stress release caused by heat treatment. This makes it difficult to guarantee the symmetry and fit accuracy of the keyway, resulting in a keyway size deviation exceeding ±0.05mm, poor accuracy stability, uneven force on gear meshing, vibration (amplitude 1.2mm / s), noise (75dB (A)), reduced transmission efficiency (92%), and even premature failure.

[0006] Therefore, traditional machining processes for gear keyway processing are inefficient and cannot effectively guarantee the symmetry and fit accuracy of the keyway, thus becoming a bottleneck for industry development.

[0007] Therefore, it is necessary to study the machining method of high-precision gear keyway for reducers. Summary of the Invention

[0008] Therefore, the purpose of this invention is to provide a high-precision gear keyway machining method, which can effectively solve the problems of difficulty in ensuring the symmetry and fit accuracy of the keyway and low efficiency in traditional gear keyway machining processes.

[0009] To achieve the above objectives, the technical solution adopted by this invention is: a high-precision gear keyway machining method for reducers, comprising the following steps: S100: heat treatment deformation prediction; collecting the workpiece material composition, the three-dimensional model after grooving, and heat treatment process parameters; simulating the heat treatment result using the heat treatment deformation prediction model to predict post-heating deviation data; iteratively correcting the heat treatment process parameters to control the post-heating deviation data within the expected range; based on the predicted post-heating deviation data, back-calculating the reserved deformation compensation amount before grooving; S200: pre-heating grooving; rough machining the keyway using a high-speed broaching machine. S300: Heat treatment; Perform heat treatment on the workpiece according to the optimized heat treatment parameters; S400: Post-heat inspection; Inspect the workpiece after heat treatment and obtain post-heat inspection data; S500: Post-heat finish milling; Establish a cutting parameter optimization model, input the post-heat inspection data, correct the dimensional error of the workpiece after heat treatment, and generate cutting parameters according to production needs and accuracy requirements; Perform high-speed finish milling on the keyway according to the cutting parameters to ensure that the dimensional accuracy, symmetry and surface roughness of the keyway meet the actual requirements, and complete the machining.

[0010] Furthermore, the heat treatment prediction model predicts post-heat treatment deviation data through the following steps: S110: Geometric model mesh generation; finite element analysis is performed on the 3D model after grooving, and mesh generation is carried out; S120: Temperature field model establishment; combined with the workpiece material composition and heat treatment process parameters, a temperature field model is established during the heat treatment process to simulate the temperature changes of various parts of the workpiece during the heat treatment process; S130: Thermo-mechanical coupling calculation; combined with the expansion coefficient, phase transformation characteristics and material mechanical properties, thermo-mechanical coupling calculation is used to analyze the thermal stress caused by temperature field changes and its influence on the workpiece geometry; S140: Geometric prediction deviation; based on the relationship between the temperature field and the workpiece geometric deformation, the dimensional deviation and symmetry deviation of the workpiece after heat treatment are predicted.

[0011] Furthermore, the heat treatment includes normalizing, quenching, and tempering performed sequentially; the heat treatment process parameters include normalizing temperature, quenching cooling medium, quenching process parameters, and tempering regime.

[0012] Furthermore, the cutting parameter optimization model includes a dimension correction model and a multi-parameter coupled analysis model; the cutting parameters include cutting amount, cutting speed, feed rate, and depth of cut; the post-heating detection data includes dimensional error and keyway symmetry error; the dimension correction model directly adjusts the cutting amount through addition and subtraction to correct the dimensional error caused by heat treatment; the multi-parameter coupled analysis model generates the corresponding cutting speed, feed rate, and depth of cut based on the corrected cutting amount, and uses them as the initial cutting parameters for post-heating finish milling.

[0013] Furthermore, the multi-parameter coupling analysis model is established through the following method: taking the keyway's dimensional accuracy, symmetry, and surface roughness as objective functions, and cutting speed, feed rate, and depth of cut as adjustment variables, cutting parameter experiments are conducted through orthogonal experimental design to obtain keyway machining test data under different combinations of cutting speed, feed rate, and depth of cut, and the keyway's dimensional accuracy, symmetry, and surface roughness of the corresponding machining results are detected; based on the experimental data, the mapping relationship between cutting speed, feed rate, and depth of cut and keyway dimensional accuracy, symmetry, and surface roughness is constructed, and a multi-parameter coupling analysis model is established.

[0014] Furthermore, the cutting parameter optimization model also includes a cutting parameter intelligent decision-making system, which collects cutting state monitoring data in real time during the finish milling process. The cutting state monitoring data includes at least one or more of cutting force, vibration, or cutting temperature. The cutting parameter intelligent decision-making system dynamically adjusts the initial cutting parameters in real time based on the cutting state monitoring data.

[0015] Furthermore, the intelligent decision-making system for cutting parameters is established based on a deep learning method. The deep learning method uses historical cutting state monitoring data and corresponding cutting parameter corrections as training samples to learn the mapping relationship between cutting state monitoring data and cutting speed, feed rate, and depth of cut. During the hot finishing process, the intelligent decision-making system for cutting parameters calls the trained deep learning model based on the real-time collected cutting state monitoring data and outputs correction values ​​for cutting speed, feed rate, and depth of cut, thereby realizing dynamic adjustment of cutting parameters.

[0016] The beneficial effects of the above technical solution are as follows: This invention uses a heat treatment deformation prediction model to predict post-heat deviation data in advance at the front end of the process, optimize heat treatment parameters, make heat deformation controllable, and provide a compensation reserve for deformation compensation for pre-heat grooving, forming a combined process. This allows pre-heat grooving to be used in conjunction with the heat deformation prediction results for deformation compensation, so that the workpiece has a targeted compensation structure before entering heat treatment, creating stable boundary conditions for subsequent high-precision correction.

[0017] Based on this, heat treatment is performed according to the predicted and optimized process parameters, ensuring that the post-heat deformation of the workpiece stabilizes and converges within the expected range. Subsequently, real geometric deviation data is obtained through post-heat inspection and directly applied to the post-heat finish milling process. The cutting parameter optimization model, on the one hand, combines pre-heat compensation to correct the cutting amount, and on the other hand, combines heat treatment deformation compensation technology with dynamic correction during the finish milling process. This forms a continuous closed loop of "pre-heat prediction - compensation optimization - pre-heat grooving - stable heat treatment - post-heat inspection - post-heat finish milling correction," transforming the uncontrollable thermal deformation problem in the keyway machining process into a predictable, compensable, and correctable process variable. This fundamentally breaks through the technical bottlenecks of traditional gear keyway post-heat hard machining, such as difficulty in ensuring symmetry, large fluctuations in fit accuracy, and low machining efficiency, significantly improving the machining accuracy of keyways and ensuring that the keyway accuracy meets design requirements. Detailed Implementation

[0018] The present invention will be further described in detail below with reference to specific embodiments: This embodiment aims to provide a high-precision gear keyway machining method, which is mainly used for high-precision machining of gear keyways, addressing the problem that traditional gear keyway machining processes are difficult to effectively guarantee the symmetry and fit accuracy of the keyways and are inefficient.

[0019] A high-precision gear keyway machining method includes the following steps: S100: Heat treatment deformation prediction; collect the workpiece material composition, the three-dimensional model after grooving, and the heat treatment process parameters.

[0020] The workpiece material must contain at least carbon, chromium, and molybdenum. In this embodiment, microalloying technology is also introduced to improve the material's thermal stability and wear resistance while ensuring the strength (increased by 20%-25%) and toughness of the gear matrix. This is achieved by refining the grain size (increasing the grain size grade by 2 levels). Strict control of the steel inclusion grade (≤1.5) and grain size is required. Combined with forging ratio optimization (forging ratio ≥3:1) and isothermal normalizing process, the internal microstructure of the material is improved, reducing the heat treatment deformation rate by over 25%, thus providing a blank foundation for high-precision keyway machining.

[0021] The 3D model after grooving is drawn using 3D modeling software such as SolidWorks to accurately describe the dimensions, shape, and geometric features of the gear and keyway, as well as the basic machining allowance reserved after grooving. This 3D model is the 3D model before and after grooving, not the pre-grooving or final cost model, and a basic machining allowance is reserved after grooving. The heat treatment in this embodiment includes normalizing, quenching, and tempering performed sequentially; the normalizing process uses isothermal normalizing to refine the grains, reducing heat treatment deformation by more than 30% and increasing strength by 20%-25%, meeting the requirements of high-end equipment.

[0022] Heat treatment process parameters include normalizing temperature, quenching cooling medium, quenching process parameters, and tempering regime. The quenching cooling medium includes the type and concentration of the quenching fluid; the quenching process parameters include the cooling rate (e.g., 510℃ / s); and the tempering regime includes the tempering temperature and temperature tolerance, such as a tempering temperature of ±5℃. Precise control of the tempering temperature (±5℃) reduces residual stress. Tempering temperatures that are too high or too low will affect the hardness and toughness of the material and may even induce new stresses.

[0023] A heat treatment deformation prediction model is used to simulate the heat treatment results, predict post-heat treatment deviation data, and provide data support for subsequent process compensation (pre-heat grooving and post-heat finish milling). The predicted post-heat treatment deviation data mainly includes dimensional deviation and symmetry deviation.

[0024] The heat treatment prediction model predicts post-heat deviation data through the following steps: S110: Geometric model mesh generation; Finite element analysis is performed on the three-dimensional model after grooving. The workpiece is decomposed into multiple small units through mesh generation, which provides a basis for the simulation and analysis of the subsequent heat treatment process.

[0025] S120: Establish a temperature field model; using the workpiece's three-dimensional model and the material's thermal conductivity, establish a heat conduction model; combining the workpiece's material composition and heat treatment process parameters, calculate the temperature distribution of the workpiece during heat treatment, and establish a temperature field model during heat treatment; using the finite element analysis (FEA) method, combined with mesh generation, decompose the workpiece into small units for fine calculation of the temperature field, and obtain the temperature changes of each part.

[0026] S130: Thermo-mechanical coupling calculation; Since changes in the temperature field can cause thermal expansion or phase transition, the mechanical properties of the material (such as the elastic modulus) affect the magnitude and manner of deformation.

[0027] By combining the workpiece material's expansion coefficient, phase transformation characteristics, and material mechanical properties, and through thermo-mechanical coupling calculations, the thermal stress and deformation caused by temperature changes are calculated using a mechanical model. The thermal stress caused by temperature field changes and its influence on the workpiece geometry are analyzed, and the deformation trend of the workpiece during heat treatment is obtained.

[0028] S140: Geometric prediction deviation; Based on the relationship between the temperature field and the geometric deformation of the workpiece, combined with parameters such as the coefficient of thermal expansion and the phase transformation model, the deformation of the workpiece after heat treatment is simulated. By comparing the workpiece data before heat treatment, the dimensional deviation and symmetry deviation of the workpiece after heat treatment are predicted, providing data support for subsequent process compensation. It can be combined with finite element simulation software to simulate the three-dimensional model after heat treatment, so as to intuitively reflect the influence of thermal deformation.

[0029] After the simulation is completed, the heat treatment process parameters are iteratively corrected to optimize the heat treatment process parameters, and the simulation prediction is carried out again to control the post-heating deviation data within the expected range and ensure that the thermal deformation is stable and controllable.

[0030] Meanwhile, based on the predicted post-heat deviation data, the allowable deformation compensation amount for pre-heat grooving is calculated to correct the machining allowance and the three-dimensional model after grooving.

[0031] Because thermal stress after heat treatment can cause local deformation on or inside the workpiece, such as warping or shrinkage, especially the volume change that occurs when the material transforms from austenite to martensite during quenching, the keyway size and symmetry will shift. By predicting heat treatment deformation before processing, the deformation can be known in advance, thereby optimizing the heat treatment process parameters and the grooving machining allowance. A reasonable compensation amount is reserved in the pre-heat processing stage (in this embodiment, the pre-heat compensation amount is 0.1-0.2mm), forming a heat treatment deformation compensation technology to ensure that the keyway accuracy meets the design requirements.

[0032] S200: Pre-heat broaching; the keyway of the workpiece blank is rough machined using a high-speed broaching machine, with a machining allowance reserved in conjunction with deformation compensation. In this example, a high-speed broaching machine (broaching speed 80-120m / min) is used to remove 80% to 90% of the machining allowance. The broach material is high-speed steel (W6Mo5Cr4V2, hardness HRC63-66), the tooth rise is 0.02-0.05mm, and the groove width accuracy is controlled within ±0.05mm, leaving a uniform allowance for finish milling after heat treatment.

[0033] This embodiment combines conventional machining processes and conducts comparative experiments on various process schemes, including "pre-heat grooving + post-heat grinding," "pre-heat milling + post-heat precision boring," and "pre-heat grooving + post-heat precision milling." Based on machining accuracy testing, tool wear analysis, and production efficiency evaluation, the "pre-heat grooving + post-heat precision milling" combination was ultimately determined to be the optimal solution. Pre-heat grooving, performed before heating, can quickly remove most of the excess material, reducing the post-heat machining load and ensuring machining efficiency. Post-heat precision milling employs high-speed milling technology, capable of handling the precision machining of high-hardness workpieces after heating. Combined with a high-precision CNC system, it achieves high-precision keyway forming.

[0034] S300: Heat treatment; The workpiece is heat treated according to the optimized heat treatment parameters; The heat treatment process is strictly carried out according to the heat treatment deformation prediction, which will not be elaborated here.

[0035] S400: Post-heat treatment inspection; In this embodiment, the German Klingberg gear measuring center (measurement accuracy ±0.002mm) and its matching inspection software are used to perform 100% online inspection of keyway symmetry, slot width, surface roughness and other indicators. The inspection data is uploaded to the MES system (Manufacturing Execution System) in real time to perform precision inspection on the heat-treated workpiece and obtain post-heat treatment inspection data. The post-heat treatment inspection data includes at least dimensional error and keyway symmetry error. The post-heat treatment inspection data is compared with the predicted post-heat treatment deviation data to establish a simulated deviation dataset, and the heat treatment deformation prediction model is optimized to form a closed loop.

[0036] S500: Finish milling after heat treatment; establish a cutting parameter optimization model, input post-heat test data, correct the dimensional error of the workpiece after heat treatment, and generate cutting parameters according to production needs and accuracy requirements.

[0037] The cutting parameter optimization model includes a dimension correction model and a multi-parameter coupled analysis model; the cutting parameters addressed in this embodiment include cutting quantity, cutting speed, feed rate, and depth of cut.

[0038] Based on the dimensional deviation prediction data provided by the heat treatment prediction model and the actual inspection data, the dimensional correction model directly calculates and adjusts the cutting amount using addition and subtraction to correct the dimensional errors caused by heat treatment, ensuring that the cut dimensional parameters are within the specified tolerance range. If a dimension in the actual inspection data is greater than the final predicted post-heat treatment deviation data (e.g., the keyway width exceeds the upper tolerance upwards), the cutting amount is increased (e.g., increasing the depth of cut or milling allowance) to compensate for the excessive dimensional error, ensuring that the final dimension meets the workpiece requirements. Conversely, if a dimension is smaller than the final predicted post-heat treatment deviation data (e.g., the keyway width exceeds the lower tolerance downwards), the cutting amount is reduced to avoid excessive material removal.

[0039] Based on the corrected cutting amount, the multi-parameter coupled analysis model generates the corresponding cutting speed, feed rate and depth of cut, which are used as the initial cutting parameters for hot finishing milling. That is, under the premise that the dimensional error has been corrected by the cutting amount, a set of the most suitable cutting speed v, feed rate f and depth of cut ap are found so that the finish milled keyway meets the requirements of dimensional accuracy, symmetry and roughness.

[0040] In this embodiment, the multi-parameter coupled analysis model is established using the following method: the dimensional accuracy (tolerance ±0.05mm), symmetry (tolerance ±0.01mm), and surface roughness (Ra≤1.6μm) of the keyway are used as objective functions, and cutting speed, feed rate, and depth of cut are used as adjustment variables. First, orthogonal experimental design is used to conduct cutting parameter experiments on heat-treated samples of the same material. Fine milling experiments are performed on different combinations of cutting speed, feed rate, and depth of cut, and the dimensional accuracy error, symmetry error, and surface roughness of the corresponding keyway are detected. Experimental data on keyway machining under different combinations of cutting speed, feed rate, and depth of cut are obtained, and the dimensional accuracy, symmetry, and surface roughness of the corresponding machining results are detected, forming an experimental sample dataset of cutting parameters and machining quality indicators.

[0041] Based on the experimental sample dataset, a multivariate regression method was used to establish the relationship between cutting speed v, feed rate f, depth of cut ap, and dimensional accuracy error E. d Symmetry error E s The mathematical mapping function between surface roughness Ra and surface roughness Ra is obtained. The explicit mathematical mapping relationship is then derived: ;in The results are calculated from the experimental data using a quadratic response surface polynomial form, as follows: Based on this, with minimum surface roughness as the optimization objective and dimensional accuracy error and symmetry error not exceeding preset tolerances as constraints, a multi-objective constrained optimization model is constructed. Within the allowable range of the process, the optimal combination of cutting speed, feed rate, and depth of cut is obtained as the initial cutting parameters for post-heat finishing milling. Therefore, the optimization model is constructed as follows: This model enables post-heat finishing milling to simultaneously achieve geometric error correction, symmetry stability control, and surface quality compliance in a single process, avoiding the problem of relying on repeated trial cuts based on experience in traditional processes.

[0042] Following the cutting parameters, the keyway was then precision milled at high speed to ensure that its dimensional accuracy, symmetry, and surface roughness met the actual requirements, thus completing the machining process. A Taiwan-made Yu-Chia CNC machining center (positioning accuracy ±0.005mm, repeatability ±0.003mm) was used, and a CBN end mill (10mm diameter, 4 teeth) was employed for high-speed milling (v=200-300m / min, f=800-1200mm / min). Combined with heat treatment deformation compensation technology, the keyway symmetry was achieved to a precision of ±0.01mm and a surface roughness Ra of 1.6μm.

[0043] Furthermore, in this embodiment, the cutting parameter optimization model also includes a cutting parameter intelligent decision-making system. The cutting parameter intelligent decision-making system collects cutting state monitoring data in real time during the finish milling process. The cutting state monitoring data includes at least cutting force, vibration, and cutting temperature.

[0044] The intelligent decision-making system for cutting parameters dynamically adjusts the initial cutting parameters in real time based on cutting condition monitoring data. This system is built upon a deep learning approach. The deep learning method uses historical cutting condition monitoring data and corresponding cutting parameter corrections as training samples, assuming predetermined constant dimensional accuracy, symmetry, and surface roughness, to learn the mapping relationship between the cutting condition monitoring data and cutting speed, feed rate, and depth of cut. The deep learning model uses the cutting condition monitoring data as the feature vector input and the cutting parameter corrections as the output, undergoing supervised learning training. This allows the model to learn the nonlinear mapping relationship between the cutting condition and the cutting parameter corrections, resulting in the trained deep learning model.

[0045] During the post-heat finishing milling process, the CNC machining center first calls the initial cutting parameters output by the multi-parameter coupling model. The intelligent decision-making system for cutting parameters collects cutting force (range 0-5000N, accuracy 1%), vibration (range ±50g, frequency 0-10kHz), and cutting temperature (infrared thermometry, accuracy ±2℃) in real time through sensors installed on the CNC machining center. It constructs cutting state monitoring data and forms a real-time feature vector for input. Then, the real-time feature vector is fed into the trained deep learning model to calculate the cutting parameter correction amount. The correction amount is then superimposed on the initial cutting parameters (or the previous cutting parameters) to obtain the latest cutting parameters and send them to the CNC machining center, realizing online dynamic adjustment of the cutting process (optimization cycle ≤100ms). The intelligent decision-making system for cutting parameters can dynamically adjust the cutting speed, feed rate, and depth of cut based on real-time changes in the cutting process. This ensures that the finish milling process maintains a stable cutting state even under conditions of material hardness changes after heating, release of local residual stress, and fluctuations in cutting load. This avoids vibration, local overcutting, or surface burns, further stabilizing the dimensional accuracy, symmetry, and surface roughness of the keyway. Through continuous feedback optimization, it can achieve dynamic adaptation of cutting parameters, significantly improving production efficiency and reducing machining energy consumption.

Claims

1. A method for machining keyways in high-precision reducer gears, characterized in that: Includes the following steps: S100: Heat treatment deformation prediction; The process involves collecting the workpiece's material composition, a 3D model after grooving, and heat treatment process parameters. A heat treatment deformation prediction model is used to simulate the heat treatment results and predict post-heat deviation data. The heat treatment process parameters are iteratively corrected to control the post-heat deviation data within the expected range. Based on the predicted post-heat deviation data, the allowable deformation compensation amount for pre-heat grooving is calculated. S200: Pre-heat grooving; the keyway is rough-machined using a high-speed broaching machine, with a machining allowance reserved based on the deformation compensation amount. S300: Heat treatment; the workpiece is heat-treated according to the optimized heat treatment parameters. S400: Post-heat inspection. The workpiece is inspected after heat treatment to obtain post-heat inspection data; S5 00: Finish milling after heating; Establish a cutting parameter optimization model, input post-heating inspection data, correct the dimensional errors of the workpiece after heat treatment, and generate cutting parameters according to production needs and accuracy requirements; According to the cutting parameters, the keyway is precision milled at high speed to ensure that the dimensional accuracy, symmetry and surface roughness of the keyway meet the actual requirements, and the machining is completed.

2. The high-precision gear keyway machining method according to claim 1, characterized in that: The heat treatment prediction model predicts post-heat treatment deviation data through the following steps: S110: Geometric model mesh generation; Finite element analysis is performed on the 3D model after grooving, and mesh generation is carried out; S120: Temperature field model establishment; Combining the workpiece material composition and heat treatment process parameters, a temperature field model is established during the heat treatment process to simulate the temperature changes of various parts of the workpiece during heat treatment; S130: Thermo-mechanical coupling calculation; Combining the expansion coefficient, phase transformation characteristics, and material mechanical properties, thermo-mechanical coupling calculation is used to analyze the thermal stress caused by temperature field changes and its influence on the workpiece geometry; S140: Geometric prediction deviation; Based on the relationship between the temperature field and the workpiece geometric deformation, the dimensional deviation and symmetry deviation of the workpiece after heat treatment are predicted.

3. The high-precision gear keyway machining method according to claim 1, characterized in that: Heat treatment includes normalizing, quenching, and tempering performed sequentially; heat treatment process parameters include normalizing temperature, quenching cooling medium, quenching process parameters, and tempering regime.

4. The high-precision gear keyway machining method according to claim 1, characterized in that: The cutting parameter optimization model includes a dimension correction model and a multi-parameter coupled analysis model; the cutting parameters include cutting amount, cutting speed, feed rate, and depth of cut; the post-heating detection data includes dimensional error and keyway symmetry error; the dimension correction model directly adjusts the cutting amount through addition and subtraction to correct the dimensional error caused by heat treatment; the multi-parameter coupled analysis model generates the corresponding cutting speed, feed rate, and depth of cut based on the corrected cutting amount, and uses them as the initial cutting parameters for post-heating finish milling.

5. The high-precision gear keyway machining method according to any one of claims 1-4, characterized in that: The multi-parameter coupled analysis model is established through the following method: taking the keyway's dimensional accuracy, symmetry, and surface roughness as objective functions, and cutting speed, feed rate, and depth of cut as adjustment variables, cutting parameter experiments are conducted through orthogonal experimental design to obtain keyway machining test data under different combinations of cutting speed, feed rate, and depth of cut, and the corresponding machining results are detected in terms of keyway dimensional accuracy, keyway symmetry, and surface roughness; based on the experimental data, the mapping relationship between cutting speed, feed rate, and depth of cut and keyway dimensional accuracy, keyway symmetry, and surface roughness is constructed, and the multi-parameter coupled analysis model is established.

6. The high-precision gear keyway machining method according to claim 5, characterized in that: The cutting parameter optimization model also includes a cutting parameter intelligent decision-making system, which collects cutting state monitoring data in real time during the finish milling process. The cutting state monitoring data includes at least one or more of cutting force, vibration, or cutting temperature. The cutting parameter intelligent decision-making system dynamically adjusts the initial cutting parameters in real time based on the cutting state monitoring data.

7. The high-precision gear keyway machining method according to claim 6, characterized in that: The intelligent decision-making system for cutting parameters is established based on deep learning methods. The deep learning method uses historical cutting state monitoring data and corresponding cutting parameter corrections as training samples to learn the mapping relationship between cutting state monitoring data and cutting speed, feed rate and depth of cut. During the hot finishing process, the intelligent decision-making system for cutting parameters calls the trained deep learning model based on the real-time collected cutting status monitoring data, and outputs correction values ​​for cutting speed, feed rate and depth of cut, thereby realizing dynamic adjustment of cutting parameters.