An aeronautical part machining heat affected zone identification method

Abstract

This invention discloses a method for identifying the heat-affected zone (HAZ) during machining of aerospace parts. Addressing the problems of existing HAZ detection methods, such as high destructiveness, insufficient accuracy, and incompatibility with multi-pass machining conditions, this invention constructs a critical threshold database for HAZ based on phase transition dynamics through pre-calibration; it simultaneously acquires multi-source signals during the cutting process and completes spatiotemporal registration, extracting effective features after preprocessing; it employs dynamic heat source intensity calculation combined with multi-pass heat accumulation correction, and uses a moving heat source solution to invert the transient temperature field inside the workpiece. Combined with a phase transition dynamics model and acoustic emission phase transition characteristic correction, the three-dimensional spatial distribution and damage level of the HAZ are obtained. This invention enables non-destructive online detection with high accuracy, is suitable for machining complex curved surface parts in aerospace applications, and can be widely used for quality control in the machining of difficult-to-machine aerospace parts.

Description

Technical Field

[0001] This invention relates to the field of aerospace precision manufacturing technology, and in particular to a method for identifying the heat-affected zone during the machining of aerospace parts. Background Technology

[0002] Core aerospace components such as engine blades, integral bladed disks, and fuselage structural parts often utilize difficult-to-machine materials like titanium alloys and nickel-based high-temperature alloys. These materials have low thermal conductivity and high high-temperature strength, resulting in a significant amount of cutting heat being generated in the cutting zone during machining. This leads to phase transformation, grain growth, and a decrease in microhardness in the workpiece surface material, forming a heat-affected zone (HAZ). The HAZ significantly reduces the fatigue strength, corrosion resistance, and service life of the parts, making it one of the core indicators for quality control in aerospace component machining. Furthermore, aerospace components often have complex curved surface structures and commonly employ multi-pass, intensive milling processes, resulting in a significant cumulative effect of residual heat from adjacent toolpaths, further increasing the difficulty of controlling and identifying the HAZ.

[0003] Currently, heat-affected zone (HAZ) detection in the industry is mainly divided into two technical routes: offline detection and online detection. Offline detection, represented by metallographic microscopy and microhardness testing, requires cutting and sampling of parts before microscopic observation. This is a destructive test and cannot be applied to full-scale inspection of finished parts. Furthermore, it has a long testing cycle and high cost, allowing only post-processing sampling and not real-time guidance for process adjustments. Online detection often uses infrared thermometry to collect the surface temperature of the workpiece and determine the boundary of the HAZ using a fixed phase transition temperature as a threshold. However, this method only obtains instantaneous temperature information of the workpiece surface and cannot infer the internal temperature gradient distribution and thermal exposure time, ignoring the cumulative effect of material microstructure transformation over time. It also fails to consider the cumulative thermal error caused by the residual heat from multiple cutting passes, resulting in generally low identification accuracy. Some existing technologies attempt to introduce acoustic emission signals to assist in HAZ determination, but this is only used as a qualitative basis and is not quantitatively coupled with temperature field calculations and phase transition dynamics models, making it impossible to achieve accurate quantitative identification of the three-dimensional spatial distribution and damage level of the HAZ.

[0004] In summary, existing heat-affected zone identification technologies have shortcomings such as destructive detection, insufficient identification accuracy, incompatibility with multi-pass processing conditions of aerospace parts, and lack of deep quantitative coupling of multi-source signals. These shortcomings make it difficult to meet the non-destructive, high-precision, and online quality control requirements of high-value aerospace manufacturing for high-end parts. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a method for identifying the heat-affected zone (HAZ) in the machining of aerospace parts. Using phase transition dynamics as the core criterion, and combining techniques such as multi-source signal spatiotemporal registration, dynamic heat source intensity calculation, multi-pass heat accumulation correction, and acoustic emission phase transition characteristic correction, this method achieves non-destructive, high-precision, in-situ online three-dimensional identification and damage classification of the HAZ, adapting to the actual production needs of multi-pass machining of complex curved surface parts in aerospace. To achieve the above objective, the technical solution adopted by this invention includes the following steps: S1 Pre-calibration: For the corresponding materials of the aerospace parts to be processed, a database of critical thresholds for thermal effects based on phase transformation dynamics is constructed, and the quantitative mapping relationship of temperature-thermal action time-microstructure transformation under different cutting conditions is calibrated. At the same time, the thermal accumulation correction coefficient of multi-pass cutting and the acoustic emission characteristic threshold corresponding to the solid phase transformation of the material are calibrated. S2 Online Acquisition and Spatiotemporal Registration: During the cutting process, infrared temperature field signals, acoustic emission signals, and three-dimensional cutting force signals of the workpiece cutting area are acquired simultaneously, as well as real-time tool coordinate data output by the machine tool CNC system; using the machine tool CNC coordinates as a unified reference, the time axis synchronization and spatial position mapping of multi-source signals are completed, so that each spatial coordinate point on the workpiece cutting path corresponds to a set of multi-physical quantity signals; S3 Signal Preprocessing and Feature Extraction: Interference removal and temperature correction are performed on the infrared temperature field, noise reduction and decomposition of the acoustic emission signal are performed and phase transition related features are extracted, and the cutting force signal is processed to obtain cutting specific energy and instantaneous heat source intensity parameters. S4 Force-Heat-Organization Coupling Inversion Identification: The instantaneous heat source intensity is calculated based on the triaxial cutting force. Based on the instantaneous heat source intensity and the initial temperature of the workpiece matrix, the heat accumulation correction coefficient is called to correct the residual heat superposition effect of multi-pass cutting. The transient temperature field inside the workpiece is inverted by the moving heat source solution. The temperature field data is substituted into the phase transformation dynamics model to calculate the microstructure transformation variables at each location. When the acoustic emission phase transformation characteristic quantity exceeds the calibration threshold, the boundary range of the heat-affected zone is expanded accordingly to correct the systematic error of pure temperature inversion, and the three-dimensional spatial distribution and damage level of the heat-affected zone are obtained. S5 Result Reconstruction and Output: Automatically determine the out-of-standard areas based on preset process thresholds and output an identification report.

[0006] As a further aspect of the present invention, in step S1, the mapping relationship between temperature, heat treatment time, and microstructure transformation is fitted; and three levels of heat-affected zone critical thresholds are set in combination with the performance requirements of aerospace parts: the initial threshold corresponds to a microstructure transformation of ≥5% or a decrease in microhardness of ≥3%, the moderate threshold corresponds to a microstructure transformation of 15%~30% or a decrease in hardness of 5%~10%, and the severe threshold corresponds to a microstructure transformation of ≥30% or a decrease in hardness of >10%.

[0007] As a further aspect of the present invention, in step S1, the thermal accumulation correction coefficient is calibrated through cutting tests under different pass spacings and different tool travel intervals. The correction coefficient has a corresponding mapping relationship with the pass spacing, tool travel interval, and initial substrate temperature. In step S4, if the distance between the current cutting position and the previous cutting path is less than 3 times the width of the heat-affected zone and the interval is less than the material cooling characteristic time, then the thermal accumulation temperature increment is superimposed to correct the initial substrate temperature.

[0008] As a further aspect of the present invention, in step S2, the 1ms scanning cycle of the machine tool PLC is used as a unified timestamp reference to align the time axes of signals with different sampling rates, with a time synchronization error ≤1ms; the tool coordinates are mapped to the workpiece coordinate system through coordinate system transformation, thereby completing the pixel-space coordinate transformation of the infrared temperature field and the propagation attenuation correction of the acoustic emission signal, and realizing a one-to-one spatial correspondence between the signal and the cutting position.

[0009] As a further aspect of the present invention, in step S3, the acoustic emission signal is decomposed using an adaptive weighted variational mode decomposition algorithm for noise reduction. The number of mode decompositions is automatically adjusted using the real-time cutting force fluctuation amplitude as a weighting factor. The low-frequency noise components of machine tool vibration and tool impact are removed by mutual information entropy screening, while the mid-to-high frequency components corresponding to plastic deformation and thermal phase transition are retained. The center frequency, phase transition peak amplitude, and energy ratio of the characteristic frequency band are extracted as phase transition characteristic quantities.

[0010] As a further aspect of the present invention, in step S3, the abnormal regions of chip obstruction and cutting fluid atomization are identified in the infrared temperature field through inter-frame difference. The temperature of the obstructed region is completed by radial basis function interpolation, constrained by Fourier's law of heat conduction. The surface temperature measurement value is corrected by combining the pre-calibrated cutting fluid film emissivity coefficient to obtain the true workpiece surface temperature field.

[0011] As a further aspect of the present invention, in step S4, the calculation of the instantaneous heat source intensity based on the triaxial cutting force is specifically achieved through the following formula: ; Where Q(t) is the instantaneous heat source intensity at time t, Fx(t), Fy(t), and Fz(t) are the triaxial cutting forces at time t, vx, vy, and vz are the cutting velocity components in the corresponding directions, and η is the heat-work conversion coefficient. The heat distribution coefficient is the ratio of temperature T to strain rate. The function; The heat distribution coefficient The calculation formula is: ; in, The baseline heat distribution coefficient was obtained through cutting tests. For ambient temperature, The melting point of the material; The temperature influence coefficient is determined through the material's thermophysical parameters; For reference strain rate, It is a strain rate sensitivity index, determined through cutting test calibration or material constitutive model.

[0012] The moving heat source solution is a fast solution method using the moving heat source Green's function. It uses the surface temperature field measured by infrared as the boundary constraint and the instantaneous heat source intensity as the internal input to solve the temperature-time history at different depths inside the workpiece. The temperature-time history at each location is substituted into the phase transformation dynamics model to calculate the microstructure transformation variables at each point in space. The isosurface where the microstructure transformation variables are equal to the initial threshold is determined as the physical boundary of the heat-affected zone.

[0013] As a further aspect of the present invention, in step S4, the heat-affected zone is divided into three damage levels: mild, moderate, and severe, based on the tissue transformation variable.

[0014] As a further aspect of the present invention, it also includes a model self-iterative optimization step: periodically selecting the spare area of ​​the part for offline metallographic inspection and hardness testing, comparing the actual measured values ​​with the online identification values, correcting the thermal conductivity coefficient and phase transformation kinetic parameters by the least squares method, and introducing a tool wear compensation coefficient to dynamically correct the heat source intensity based on the tool wear amount, and iteratively updating the threshold database.

[0015] Compared with the prior art, the present invention has the following significant technical advancements and beneficial effects: 1. Significantly improved recognition accuracy: Breaking through the traditional judgment logic of a single surface temperature threshold, the heat-affected zone is defined from the essence of material phase transformation. The coupling effect of temperature and thermal action time is quantified by combining phase transformation dynamics model. At the same time, acoustic emission phase transformation characteristics are introduced for cross-correction. The relative error of heat-affected zone depth recognition can be stably controlled within 7%, which is far better than the error level of more than 30% of the traditional infrared thermometry method. It can meet the stringent quality control requirements of aerospace precision machining.

[0016] 2. Adapted to multi-pass machining conditions in aerospace: In response to the residual heat superposition characteristics of dense toolpaths in complex curved surface parts of aerospace, a multi-pass heat accumulation correction coefficient matrix is ​​constructed to dynamically correct the initial temperature of the substrate. In the scenario of small pitch and close-tooth milling, the recognition error can be reduced by more than 15%, which fits the actual machining process characteristics of aerospace parts and solves the engineering pain points that are generally ignored by existing technologies.

[0017] 3. Non-destructive online real-time inspection: No need to cut and sample parts, it can realize in-situ synchronous inspection of all machined surfaces, avoiding the scrapping damage of high-value aerospace parts caused by traditional metallographic inspection, and significantly reducing the cost of quality verification; the identification process is carried out simultaneously with the cutting process, and the single-segment toolpath identification delay is down to the second level, without the need for additional machine stop inspection procedures, and without affecting the production cycle.

[0018] 4. Strong engineering adaptability and robustness: Sensor deployment does not require modification of the machine tool's main structure and can be directly connected to existing CNC systems; it also has a model self-iteration and tool wear compensation mechanism, which can adapt to production variables such as material batch fluctuations and tool life cycle, and the recognition accuracy continues to improve with iteration, making it suitable for mass production and promotion. Attached Figure Description

[0019] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart illustrating the steps of a method for identifying the heat-affected zone during machining of aerospace parts according to the present invention. Detailed Implementation

[0020] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0021] Example 1 This embodiment is applied to the five-axis CNC milling process of titanium alloy blades for aero-engines. It is used to identify the spatial distribution and damage level of the heat-affected zone (HAZ) generated during machining online, replacing traditional offline metallographic destructive testing, and achieving in-situ quality control of high-value aerospace parts. The HAZ identification method for aerospace parts machining described in this embodiment uses a pre-calibrated database as a benchmark. Through a complete process of online acquisition and spatiotemporal registration of multi-source signals, signal preprocessing and feature extraction, multi-physics coupling inversion, result output, and model iteration, it achieves accurate identification of the HAZ. Please refer to [link to relevant documentation]. Figure 1 As shown, this is achieved through the following steps: S1 Pre-calibration: Constructing a database of critical thresholds for thermal effects based on phase transition dynamics The core objective of this step is to pre-establish a quantitative benchmark system for determining the heat-affected zone (HAZ) of the corresponding material of the aerospace parts to be processed. This abandons the traditional fixed temperature threshold-based determination logic and defines the HAZ from the essential level of changes in microstructure and properties, providing a calibration basis for subsequent online identification. This step specifically includes three calibration parts: (1) Calibration of the mapping relationship between temperature, thermal treatment time and tissue transformation. First, standard samples from the same batch and under the same heat treatment condition as the aerospace parts to be processed are selected. The samples are then clamped on CNC machining equipment of the same model as the actual production equipment, ensuring that the clamping rigidity, cooling conditions, tool type, and other boundary conditions are consistent with the actual machining conditions. Miniature temperature sensing elements are pre-embedded at different depths inside the samples, and acoustic signal acquisition devices, infrared temperature measurement devices, and cutting force measurement devices are installed in conjunction with them. All acquisition devices are synchronized through a synchronous triggering system to achieve a unified time reference, ensuring the synchronicity of the acquisition of various physical quantities.

[0022] A cutting parameter matrix covering the entire actual production process was established, including multiple combinations of working conditions with different cutting speeds, feed rates, depths of cut, and cooling methods. Group cutting experiments were conducted using the controlled variable method. After each group of experiments, samples were taken along the cutting section. After metallographic sample preparation processes such as mounting, grinding, polishing, and etching, data on changes in microstructure, grain growth, phase composition ratio, and microhardness under different thermal conditions were quantitatively obtained through microscopic observation and hardness testing.

[0023] A phase transition kinetic model was used to fit the experimental data, establishing a quantitative mapping relationship between temperature, thermal treatment time, and microstructure transformation variables, thus clarifying the evolution law of thermally induced microstructure transformation in materials. Considering the fatigue performance, dimensional accuracy, and service life requirements of aerospace components, three levels of thermally affected critical thresholds were set as the judgment criteria. Initial threshold: When the tissue transformation reaches a preset ratio or the microhardness shows a detectable decrease, this threshold is the physical boundary of the heat-affected zone. Below this threshold, the tissue properties of the area are not significantly different from those of the matrix. Moderate threshold: The corresponding microstructure transformation and hardness reduction are within the design allowable range, and the mechanical properties of this area can still meet the usage requirements of the parts; Severe threshold: When the change in tissue volume and hardness exceeds the design allowable value, the fatigue strength and service performance of this area will be significantly deteriorated, which is an area that must be controlled during the processing.

[0024] (2) Calibration of the cumulative heat correction factor for multi-pass cutting For the complex curved surfaces of aerospace parts undergoing multi-pass milling, a residual heat superposition effect occurs between adjacent cutting paths. When the residual heat generated by the previous cutting pass has not completely dissipated, the subsequent cutting pass will be performed at a higher initial temperature of the substrate, leading to an expansion of the heat-affected zone. To correct the identification error caused by this effect, this embodiment conducts multi-pass cutting calibration tests in advance.

[0025] Different combinations of pass spacing and tool path intervals were set in the experiment to simulate the arrangement and timing of adjacent toolpaths in actual machining. The expansion of the heat-affected zone caused by residual heat from the previous pass was measured under different working conditions. The mapping relationship between the heat accumulation correction coefficient and the pass spacing, tool path interval, and initial substrate temperature was fitted to form a heat accumulation correction coefficient matrix. In the subsequent online identification process, when the distance between the current cutting position and the previous cutting path is less than 3 times the width of the heat-affected zone and the interval time is less than the characteristic time for the material to cool to a stable temperature, this coefficient matrix is ​​used for residual heat superposition correction.

[0026] (3) Acoustic emission phase transition characteristic threshold calibration When materials undergo solid-state phase transitions, they release characteristic acoustic signals, which can serve as a direct detection basis for microstructural transformation. This calibration section uses spectral analysis of the full-waveform acoustic emission signals acquired in the experiments to separate the characteristic frequency bands corresponding to the solid-state phase transition. The acoustic signal characteristics are then correlated with the microstructural transformation variables obtained from metallographic analysis to determine the acoustic emission energy threshold corresponding to the onset of the phase transition, as well as the characteristic energy range corresponding to different degrees of microstructural transformation. Simultaneously, the range of values ​​for the phase transition acoustic emission contribution weighting coefficients is calibrated through data fitting, providing benchmark parameters for subsequent multi-field coupling correction.

[0027] S2 Online Acquisition and Spatiotemporal Registration The purpose of this step is to simultaneously acquire multiple physical signals during the actual cutting process and solve the problems of inconsistent sampling rates of different sensors and spatial misalignment of signals caused by tool movement, so as to achieve accurate matching of multiple signals in time and space.

[0028] (1) Deployment of multi-source signal acquisition In actual machining, the data acquisition system is deployed without modifying the main structure of the machine tool. The infrared temperature measurement device is fixed in the observation position of the machine tool protection, with the lens aimed at the contact area between the tool and the workpiece. The surface emissivity is calibrated in advance under the cutting fluid environment to eliminate the influence of the liquid film on the temperature measurement accuracy. The acoustic emission sensor is fixed to the rigid support part of the workpiece fixture through a high-temperature resistant mounting base, avoiding the area directly washed by the cutting fluid and impacted by chips. The three-dimensional cutting force data and the real-time coordinate data of the tool are directly read from the machine tool CNC system and the built-in force measurement module, including the cutting force components in the three directions, the tool spatial coordinates, the feed rate of each axis, the spindle speed and other CNC parameters.

[0029] (2) Multi-sampling rate signal time synchronization Using the scan cycle of the machine tool's programmable logic controller (PLC) as a unified timestamp reference, signals with different sampling rates are aligned along the time axis. The acoustic emission signal, with its higher sampling frequency, is segmented and marked according to fixed time windows, with each segment corresponding to a unified timestamp. The infrared temperature measurement device, with its lower frame rate, supplements the temperature data at intermediate moments through inter-frame linear interpolation, achieving continuity in the time dimension. Cutting force and CNC coordinate data are read synchronously according to the controller's scan cycle. The time synchronization error of all signals is controlled within milliseconds, ensuring the time correspondence of various physical quantities during the cutting transient process and avoiding identification deviations caused by time misalignment.

[0030] (3) Cutting path spatial coordinate mapping Through coordinate system transformation calculations, the real-time coordinates of the tool in the machine tool coordinate system are converted to its position in the workpiece coordinate system, corresponding to the specific machining points in the part's CAD model. The infrared temperature field is mapped from pixel coordinates to spatial coordinates, and combined with the contact contour between the tool cutting edge and the workpiece, the workpiece surface position corresponding to the temperature field data is determined. For the acoustic emission signal, acoustic signal propagation attenuation correction is performed based on the spatial distance between the current cutting position and the sensor, eliminating the influence of distance differences on the amplitude of characteristic quantities. Ultimately, each spatial coordinate point on the cutting path corresponds to a set of matched temperature, cutting force, and acoustic emission characteristic data, completing the spatial registration of multi-source signals.

[0031] S3 Signal Preprocessing and Feature Extraction In actual machining environments, various interferences exist, such as chip splashing, cutting fluid atomization, and machine tool vibration. The raw signals contain a large amount of noise and cannot be directly used for heat-affected zone identification. This step performs noise reduction, correction, and effective feature extraction on various signals to provide reliable input data for subsequent inversion calculations.

[0032] (1) Infrared temperature field interference removal and temperature correction To address the localized occlusion and anomalies in the temperature field caused by flying chips and cutting fluid atomization, inter-frame difference calculations are first performed on consecutive temperature field images to identify anomalous regions with abrupt temperature changes, which are then determined to be interference occlusion regions. Subsequently, using Fourier's law of heat conduction as constraints and the normal temperature data surrounding the occluded region as boundary values, radial basis function interpolation is employed to complete the temperature field in the occluded region, thus restoring the true temperature distribution of the workpiece surface.

[0033] The completed temperature field still needs to be corrected for emissivity: the surface temperature measurement value is corrected by combining the pre-calibrated emissivity coefficient of the cutting fluid film, eliminating the temperature measurement error caused by the absorption and reflection of infrared radiation by the liquid film, and finally obtaining an accurate workpiece surface temperature field.

[0034] (2) Acoustic emission signal denoising decomposition and phase transition feature extraction Acoustic emission signals generated during the cutting process are a mixture of multiple sources, including material plastic deformation, tool friction, machine tool vibration, and thermally induced phase transitions. Direct use of these signals would result in significant noise interference. This embodiment employs an adaptive weighted variational mode decomposition algorithm to decompose the raw acoustic emission signal. This algorithm uses the fluctuation amplitude of the real-time cutting force as a weighting factor and can automatically adjust the number of mode decompositions according to the severity of the cutting conditions, avoiding over-decomposition or under-decomposition problems caused by a fixed number of modes, and adapting to the signal characteristics under different cutting parameters.

[0035] After decomposition, the correlation between each modal component and the cutting process is calculated using mutual information entropy. Low-frequency noise components corresponding to machine tool foundation vibration and tool entry impact are filtered out, while mid-to-high frequency effective components related to material plastic deformation and thermally induced phase transition are retained. Finally, parameters such as the center frequency, phase transition peak amplitude, and energy ratio of the phase transition characteristic frequency band are extracted from the effective components to form an acoustic emission feature set for phase transition identification.

[0036] (3) Cutting force signal processing and heat source parameter extraction The original three-dimensional cutting force signal is smoothed and denoised to extract stable real-time cutting force components. Combined with the feed rate components of each axis, the instantaneous power of the cutting process is calculated, providing basic data for subsequent heat source intensity calculation. Simultaneously, the specific energy parameter, i.e., the cutting work consumed per unit volume of material removed, is calculated to analyze the energy distribution pattern of the cutting process.

[0037] S4 Force-Thermo-Tissue Coupling Inversion Identification This step is the core of this method. It calculates the instantaneous heat source intensity by cutting force, combines heat accumulation correction and temperature field inversion, calculates the tissue transformation variables by phase change dynamics model, and introduces acoustic emission characteristics for correction, finally obtaining the three-dimensional spatial distribution and damage level of the heat-affected zone.

[0038] (1) Calculation of instantaneous heat source intensity During machining, the majority of the mechanical work consumed in cutting is converted into heat energy, and the heat source intensity is the core input parameter for temperature field calculation. Traditional methods often use fixed heat distribution coefficients, which are difficult to adapt to the effects of temperature and strain rate changes under different cutting conditions. This embodiment adopts a dynamically corrected instantaneous heat source intensity calculation method, and the calculation formula is as follows:

[0039] Formula parameter definition and derivation explanation: This formula, based on cutting power and incorporating the heat-to-work conversion efficiency and the ratio of heat transferred to the workpiece, calculates the heat transferred to the workpiece per unit time. Among these factors... for The instantaneous heat source intensity at a given moment represents the cutting heat power transferred to the workpiece per unit time. , , They are respectively The cutting force components in the three directions at any given time are acquired in real time by the machine tool force measurement system; , , These are the cutting speed components in the corresponding directions, obtained from the feed parameters of the CNC system; The heat-to-work conversion coefficient represents the proportion of cutting work converted into heat energy. For metal cutting processes, this coefficient is close to a constant value and is determined through standard cutting tests.

[0040] This is the heat distribution coefficient, representing the proportion of heat transferred to the workpiece to the total cutting heat. Unlike traditional fixed coefficients, this coefficient varies with workpiece temperature. With shear strain rate The dynamically changing function can be dynamically adjusted according to the cutting conditions, improving the accuracy of temperature field calculation.

[0041] The formula for calculating the heat distribution coefficient is as follows:

[0042] Formula parameter definition and derivation explanation: This formula corrects the baseline heat distribution coefficient from two dimensions: temperature and strain rate. Among them, The reference heat distribution coefficient is the heat distribution ratio under normal temperature and reference strain rate conditions, which is obtained through standard cutting tests. The initial ambient temperature, This refers to the melting point temperature of the corresponding material; The temperature influence coefficient reflects the influence of temperature rise on the heat distribution ratio. It is derived from the thermal properties of the material. Temperature changes will change the thermal conductivity of the workpiece, and thus change the proportion of cutting heat conducted to the workpiece. For reference strain rate, Both are strain rate sensitivity indices, calibrated through cutting tests or determined by material constitutive models. Changes in strain rate alter the heat generation rate and heat conduction conditions in the shear zone, affecting the heat distribution ratio.

[0043] By using this dynamically corrected heat distribution coefficient, the calculation of heat source intensity can better reflect the actual physical process under different cutting parameters, effectively improving the accuracy of subsequent temperature field inversion.

[0044] (2) Initial matrix temperature and thermal accumulation correction Before calculating the temperature field at the current cutting position, the initial temperature of the workpiece substrate is first determined. Residual data of the workpiece temperature field from the previous moment is read, and combined with the ambient cooling rate, the substrate temperature distribution at the current moment is calculated.

[0045] Then, the spatial distance between the current cutting position and the previous cutting path, as well as the tool path interval, are retrieved to determine whether the thermal accumulation correction condition is met: if the distance between the current cutting position and the previous tool path is less than 3 times the width of the heat-affected zone, and the interval is less than the characteristic time for the material to cool to a stable temperature, it indicates that the residual heat from the previous cutting has not completely dissipated, which will have a thermal superposition effect on the current machining. At this time, the thermal accumulation correction coefficient matrix obtained from the pre-calibration is called, and the thermal accumulation temperature increment is superimposed to correct the initial temperature of the substrate, so as to avoid the thermal affected zone identification result being too small due to residual heat superposition.

[0046] (3) Inversion of internal transient temperature field Using the corrected surface temperature field as boundary constraints and the calculated instantaneous heat source intensity as the internal heat source input, a fast solution method using the moving heat source Green's function is employed to solve for the transient temperature field inside the workpiece. This method, based on analytical solutions, offers computational efficiency improvements of more than two orders of magnitude compared to traditional finite element numerical simulation, meeting the real-time requirements of online identification.

[0047] By solving the calculations, the temperature change over time at different depths below the cutting zone is obtained, i.e., the temperature-time curve for each spatial location, which provides temperature input data for subsequent calculations of microstructure transformation.

[0048] (4) Calculation of tissue transformation variables and acoustic emission correction The temperature-time history at each spatial location is substituted into the phase transition kinetics model to calculate the initial value of the microstructure transformation variable at that location. Since pure temperature inversion is based on numerical calculations and has certain systematic errors, it cannot directly reflect the actual phase transition state inside the material. Therefore, acoustic emission phase transition characteristics are introduced for cross-correction.

[0049] When the acoustic emission phase transition characteristic energy at the current location exceeds the calibrated phase transition initiation threshold, it indicates that the degree of internal tissue transformation is higher than the calculation result of temperature inversion. The tissue transformation variable is corrected according to the multi-field coupling correction logic, and the boundary range of the heat-affected zone is expanded accordingly. If the acoustic emission characteristic energy is lower than the threshold, the tissue transformation variable is adjusted accordingly to achieve dual verification of temperature field numerical calculation and direct detection of phase transition signal, effectively improving the recognition accuracy.

[0050] (5) Determination of the boundary of the heat-affected zone and damage classification The isosurface where the microstructure transformation value equals the initial threshold is used as the physical boundary of the heat-affected zone (HAZ), thus defining its three-dimensional spatial range. Based on the magnitude of the microstructure transformation value, the HAZ is divided into three damage levels: a mild HAZ where the microstructure transformation value is between the initial and moderate thresholds, indicating minimal changes in microstructure properties that do not affect part performance; a moderate HAZ where the microstructure transformation value is between the moderate and severe thresholds, indicating a certain degree of performance degradation but still within design limits; and a severe HAZ where the microstructure transformation value exceeds the severe threshold, indicating a significant decrease in mechanical properties and fatigue strength beyond acceptable limits, making it a key area for quality control.

[0051] S5 Result Reconstruction and Output The calculated 3D distribution data of the heat-affected zone is mapped onto the original CAD model of the aerospace part. Different colors are used to distinguish heat-affected zones of different damage levels, enabling visualization. It supports cross-sectional viewing of any section, allowing observation of the thickness distribution of the heat-affected layer in the depth direction, and intuitively showing the positional relationship between the heat-affected zone and key features of the part.

[0052] Import preset process requirement thresholds, automatically scan the entire processing area, identify the locations where the heat-affected zone exceeds the standard, record information such as the coordinates, maximum depth, and damage level of the locations exceeding the standard, generate a standardized identification report, and store it in the part's processing quality file for subsequent process optimization and quality traceability.

[0053] Model self-iterative optimization To accommodate the impact of material batch fluctuations and tool wear during production, this method also includes a model self-iterative optimization step. Offline sampling is periodically performed at several identification locations in non-critical areas such as process lugs and machining allowances on the parts. The actual measured values ​​of the heat-affected zone are obtained through metallographic inspection and microhardness testing.

[0054] The actual measured values ​​were compared with the online recognition results, and the least squares method was used to correct the thermal conductivity coefficient and phase transformation kinetic parameters in the inversion model. Simultaneously, a tool wear compensation mechanism was introduced. Based on the cutting mileage or flank wear of the tool, a corresponding tool wear compensation coefficient was set, dynamically adjusting the calculated value of the heat source intensity and iteratively updating the threshold database and model parameters. After multiple iterations, the recognition accuracy can be gradually improved, ensuring recognition stability throughout the entire tool life cycle and under different material batches.

[0055] Example 2 This embodiment is a verification example in the actual production scenario of aerospace parts. The data is taken from the first-line test data of the precision milling process of titanium alloy blades of an aerospace manufacturing enterprise, and is used to verify the recognition accuracy and application effect of the method of the present invention in the actual industrial environment.

[0056] First, the test conditions and test plan.

[0057] The test subject was the precision milling of the blade profile of the TC4 titanium alloy aero-engine compressor. The test was conducted on a mass-production five-axis CNC machining center, using a solid carbide coated end mill. The test parameters covered three typical cutting conditions in actual production in the workshop, and the remaining cooling and clamping conditions were consistent with the mass production process.

[0058] The experiment used the depth of the heat-affected zone obtained by offline metallographic microscopy as the baseline true value and set up two control groups: the control group used the industry-standard "surface infrared temperature threshold method" to determine the boundary of the heat-affected zone with the material phase transformation temperature as a fixed threshold; the experimental group used the force-heat-structure coupling identification method described in this invention to compare the identification accuracy of the two methods; at the same time, the actual effect of the multi-pass cutting heat accumulation correction mechanism was verified separately.

[0059] Then, a comparative test of the accuracy of heat-affected zone identification was conducted.

[0060] Three sets of typical cutting parameters were selected for testing. Each test was repeated three times, and the average value was taken. The metallographic measured values, the values ​​identified by the traditional method, and the values ​​identified by the method of this invention were recorded respectively. The test results are shown in the table below: Table 1 Comparison of accuracy in identifying depth of heat-affected zone Precision milling conditions 62 87 40.3% 66 6.5% Semi-finish milling condition 95 124 30.5% 99 4.2% Large cutting depth working conditions 148 196 32.4% 141 4.7% The test results show that the traditional single temperature threshold method is affected by the limitations of surface temperature measurement and the deviation of internal temperature gradient estimation, and the relative error of identification generally exceeds 30%. However, the method of this invention, through internal temperature field inversion, phase transition dynamics calculation and acoustic emission characteristic correction, can stably control the relative error of identification within 7%, and the identification accuracy is significantly improved, which can meet the quality control requirements of precision machining of aerospace parts.

[0061] Next, the effect of multi-stage heat accumulation correction was verified.

[0062] For typical scenarios of multi-pass milling of aerospace parts, continuous cutting experiments with different toolpath spacings were set up to compare the recognition accuracy before and after enabling the heat accumulation correction method, verifying the actual effect of the residual heat superposition effect correction. The test results are shown in the table below: Table 2 Comparison of the effects of multi-channel heat accumulation correction 0.8 112 89 20.5% 107 4.5% 1.2 98 82 16.3% 94 4.1% 2.0 76 70 7.9% 73 3.9% Test results show that without thermal accumulation correction, the smaller the pass spacing, the more significant the residual heat superposition effect and the greater the recognition deviation. The thermal accumulation correction mechanism introduced in this invention can effectively compensate for the recognition error caused by the superposition of residual heat from multiple passes. The accuracy improvement is particularly obvious in the scenario of small-pitch, close-tooth milling, which is suitable for the multi-pass machining characteristics of complex curved surface parts in aerospace.

[0063] Finally, the overall application benefits were verified.

[0064] In addition to improved recognition accuracy, this method also offers the following advantages in actual production: Non-destructive testing: No need to cut or sample parts; all machined surfaces of finished parts can be inspected, avoiding the scrap-worthy damage of high-value aerospace parts caused by traditional metallographic testing, and significantly reducing the cost of quality verification.

[0065] Online real-time detection: The identification process is carried out simultaneously with the cutting process. The single-segment toolpath identification delay is in the second range. No additional machine stop detection process is required, and it does not affect the production cycle, adapting to the rhythm requirements of mass production lines.

[0066] Highly adaptable to engineering applications: Sensor deployment requires no modification to the main structure of the machine tool, can be directly connected to existing CNC systems, is compatible with mainstream five-axis machining equipment in workshops, and has good conditions for mass production and promotion.

[0067] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for identifying the heat-affected zone during machining of aerospace parts, characterized in that, Includes the following steps: S1 Pre-calibration: For the corresponding materials of the aerospace parts to be processed, a database of critical thresholds for thermal effects based on phase transformation dynamics is constructed, and the quantitative mapping relationship of temperature-thermal action time-microstructure transformation under different cutting conditions is calibrated. At the same time, the thermal accumulation correction coefficient of multi-pass cutting and the acoustic emission characteristic threshold corresponding to the solid phase transformation of the material are calibrated. S2 Online Acquisition and Spatiotemporal Registration: During the cutting process, infrared temperature field signals, acoustic emission signals, and three-dimensional cutting force signals of the workpiece cutting area are acquired simultaneously, as well as real-time tool coordinate data output by the machine tool CNC system; Using the CNC coordinates of the machine tool as a unified reference, the time axis synchronization and spatial position mapping of multi-source signals are completed, so that each spatial coordinate point on the workpiece cutting path corresponds to a set of multi-physical quantity signals. S3 Signal Preprocessing and Feature Extraction: Interference removal and temperature correction are performed on the infrared temperature field, noise reduction and decomposition of the acoustic emission signal are performed and phase transition related features are extracted, and the cutting force signal is processed to obtain cutting specific energy and instantaneous heat source intensity parameters. S4 Force-Heat-Organization Coupling Inversion Identification: The instantaneous heat source intensity is calculated based on the triaxial cutting force. Based on the instantaneous heat source intensity and the initial temperature of the workpiece matrix, the heat accumulation correction coefficient is called to correct the residual heat superposition effect of multi-pass cutting. The transient temperature field inside the workpiece is inverted by the moving heat source solution. The temperature field data is substituted into the phase transformation dynamics model to calculate the microstructure transformation variables at each location. When the acoustic emission phase transformation characteristic quantity exceeds the calibration threshold, the boundary range of the heat-affected zone is expanded accordingly to correct the systematic error of pure temperature inversion, and the three-dimensional spatial distribution and damage level of the heat-affected zone are obtained. S5 Result Reconstruction and Output: Automatically determine the out-of-standard areas based on preset process thresholds and output an identification report.

2. The method for identifying the heat-affected zone during machining of aerospace parts according to claim 1, characterized in that, In step S1, the mapping relationship between temperature, heat treatment time, and microstructure transformation is fitted; and three levels of heat-affected zone critical thresholds are set in combination with the performance requirements of aerospace parts: the initial threshold corresponds to a microstructure transformation of ≥5% or a decrease in microhardness of ≥3%, the moderate threshold corresponds to a microstructure transformation of 15%~30% or a decrease in hardness of 5%~10%, and the severe threshold corresponds to a microstructure transformation of ≥30% or a decrease in hardness of >10%.

3. The method for identifying the heat-affected zone during machining of aerospace parts according to claim 1, characterized in that, In step S1, the thermal accumulation correction coefficient is calibrated through cutting tests under different pass spacing and different tool pass interval times. The correction coefficient has a corresponding mapping relationship with the pass spacing, tool pass interval time, and initial substrate temperature. In step S4, if the distance between the current cutting position and the previous cutting path is less than 3 times the width of the heat-affected zone and the interval time is less than the material cooling characteristic time, the thermal accumulation temperature increment is superimposed to correct the initial substrate temperature.

4. The method for identifying the heat-affected zone during machining of aerospace parts according to claim 1, characterized in that, In step S2, the 1ms scanning cycle of the machine tool PLC is used as a unified timestamp reference to align the time axis of signals with different sampling rates, and the time synchronization error is ≤1ms. The tool coordinates are mapped to the workpiece coordinate system through coordinate system transformation, and the pixel-space coordinate transformation of the infrared temperature field and the propagation attenuation correction of the acoustic emission signal are completed, so as to realize the one-to-one spatial correspondence between the signal and the cutting position.

5. The method for identifying the heat-affected zone during machining of aerospace parts according to claim 1, characterized in that, In step S3, the acoustic emission signal is decomposed into noise using an adaptive weighted variational mode decomposition algorithm. The number of mode decompositions is automatically adjusted using the real-time cutting force fluctuation amplitude as a weighting factor. The low-frequency noise components of machine tool vibration and tool impact are removed by mutual information entropy screening, while the mid-to-high frequency components corresponding to plastic deformation and thermal phase transition are retained. The center frequency, phase transition peak amplitude, and energy ratio of the characteristic frequency band are extracted as phase transition characteristic quantities.

6. The method for identifying the heat-affected zone during machining of aerospace parts according to claim 1, characterized in that, In step S3, the abnormal regions of chip obstruction and cutting fluid atomization are identified in the infrared temperature field through inter-frame difference. The temperature of the obstructed region is completed by radial basis function interpolation, constrained by Fourier's law of heat conduction. The surface temperature measurement value is corrected by combining the pre-calibrated cutting fluid film emissivity coefficient to obtain the real workpiece surface temperature field.

7. The method for identifying the heat-affected zone during machining of aerospace parts according to claim 1, characterized in that, In step S4, the calculation of the instantaneous heat source intensity based on the triaxial cutting force is specifically achieved through the following formula: ； Where Q(t) is the instantaneous heat source intensity at time t, Fx(t), Fy(t), and Fz(t) are the triaxial cutting forces at time t, vx, vy, and vz are the cutting velocity components in the corresponding directions, and η is the heat-work conversion coefficient. The heat distribution coefficient is the ratio of temperature T to strain rate. The function; The heat distribution coefficient The calculation formula is: ； in, The baseline heat distribution coefficient was obtained through cutting tests. For ambient temperature, The melting point of the material; The temperature influence coefficient is determined through the material's thermophysical parameters; For reference strain rate, It is a strain rate sensitivity index, determined through cutting test calibration or material constitutive model; The moving heat source solution is a fast solution method using the moving heat source Green's function. It uses the surface temperature field measured by infrared as the boundary constraint and the instantaneous heat source intensity as the internal input to solve the temperature-time history at different depths inside the workpiece. The temperature-time history at each location is substituted into the phase transformation dynamics model to calculate the microstructure transformation variables at each point in space. The isosurface where the microstructure transformation variables are equal to the initial threshold is determined as the physical boundary of the heat-affected zone.

8. The method for identifying the heat-affected zone during machining of aerospace parts according to claim 1, characterized in that, In step S4, the heat-affected zone is divided into three damage levels: mild, moderate, and severe, based on the tissue transformation variables.

9. The method for identifying the heat-affected zone during machining of aerospace parts according to claim 1, characterized in that, It also includes a model self-iterative optimization step: periodically select the part's margin area for offline metallographic inspection and hardness testing, compare the actual measured values ​​with the online identification values, correct the thermal conductivity coefficient and phase transformation kinetic parameters using the least squares method, introduce a tool wear compensation coefficient, dynamically correct the heat source intensity based on the tool wear amount, and iteratively update the threshold database.

Images