Aircraft engine thin-walled part sequential machining anti-deformation cutting method and system

Abstract

The application provides a deformation prevention cutting method and system for sequential machining of thin-walled parts of an aero-engine, and relates to the technical field of aero-engine part manufacturing. The method comprises the following steps: dividing a plurality of machining units of thin-walled parts on a blank to be machined, and pre-setting an annular material reservation area as a deformation prevention support belt at the inner wall connection between each thin-walled part and the next thin-walled part adjacent thereto; based on the deformation prevention support belt, performing rough machining on the current thin-walled part to obtain a rough machining surface; based on the rough machining surface, collecting radial runout data, and fitting to obtain a stiffness characteristic surface. Through stiffness characteristic surface modeling, dynamic offset compensation, path planning and closed-loop correction, the application realizes data connection and precision consistency control of sequential machining of thin-walled parts.

Description

Technical Field

[0001] This invention relates to the field of aero-engine component manufacturing technology, and in particular to a deformation-resistant cutting method and system for sequential processing of thin-walled aero-engine components. Background Technology

[0002] In the field of aero-engine component manufacturing, there are a large number of thin-walled, high-precision cylindrical and ring-shaped parts. To save precious metal materials, a nesting process is usually adopted to sequentially process multiple products from a single blank. However, these thin-walled parts have poor rigidity and require high machining accuracy. Deformation may occur during sequential machining, affecting the product qualification rate.

[0003] Traditional machining processes follow a process of finishing before cutting, which has the following drawbacks: In the cutting process, the part is only connected to the blank through an extremely thin connecting part, resulting in weak overall rigidity. The cutting force acts on this isolated and less rigid part, potentially causing elastic or plastic deformation, or uneven force distribution at the moment of cut, leading to tool wear and surface damage. In terms of data processing, machining is mainly based on geometric programming paths, without incorporating data acquisition and processing of the blank's original state. This may make it difficult to quantitatively obtain the stiffness distribution characteristics of the blank at different circumferential positions, requiring improvement in the matching between subsequent machining parameters and the actual blank state. Traditional finishing tools... The cutting path uses uniform cutting parameters, failing to fully consider the impact of the anisotropic distribution of blank stiffness on the machining process. The actual wall thickness distribution after finishing may be non-uniform, affecting the stability of subsequent cutting operations. In terms of cutting path planning, a fixed entry angle and feed rate are used without adjusting the cutting path according to the actual wall thickness distribution and stiffness changes. The matching between the cutting force direction and the workpiece stiffness distribution needs to be optimized. A data acquisition and feedback mechanism for residual stress release effects has not been established, making it difficult to obtain the actual change in the part contour after the support strip is removed. Error control in sequential machining mainly relies on single-piece assurance, and the consistency control capability of continuous machining needs to be improved. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method and system for preventing deformation during sequential machining of thin-walled parts for aero-engines. Through stiffness characteristic surface modeling, dynamic offset compensation, path planning and closed-loop correction, the method achieves data connectivity and accuracy consistency control during sequential machining of thin-walled parts.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0006] In a first aspect, a method for preventing deformation during sequential machining of thin-walled components for aero-engines, the method comprising:

[0007] The blank to be processed is divided into multiple processing units for thin-walled parts. At the inner wall connection between each thin-walled part and its next adjacent thin-walled part, an annular material retention area is pre-set as an anti-deformation support band. Based on the anti-deformation support band, the current thin-walled part is rough-machined to obtain a rough-machined surface. Based on the rough-machined surface, radial runout data is collected and fitted to obtain a stiffness characteristic surface.

[0008] The dynamic offset compensation amount is determined based on the stiffness characteristic surface, and the outer circle and inner hole of the current thin-walled part are precision machined to obtain a precision machined surface that retains the anti-deformation support band; based on the precision machined surface, the actual wall thickness distribution state and the actual wall thickness value are determined.

[0009] A dynamic cutting path mapping domain is constructed based on the actual wall thickness distribution and stiffness characteristic surface. Within the dynamic cutting path mapping domain, the feed path of the cutting tool is decomposed into multiple discretized segments according to the actual wall thickness value and the curvature change of the stiffness characteristic surface, and a corresponding entry angle and feed rate are set for each segment. Cutting is performed according to the entry angle and feed rate to maintain the finished contour without deformation and complete the cutting.

[0010] After the cutting is completed, the residual anti-deformation support band on the inner wall of the current thin-walled part is removed; during the cutting process, radial runout data is collected to obtain the released profile surface, and the profile correction value is obtained by comparing it with the stiffness feature surface; the profile correction value is used for radial runout data collection and stiffness feature surface fitting before the next thin-walled part is processed.

[0011] Secondly, a deformation-resistant cutting system for sequential processing of thin-walled components for aero-engines includes:

[0012] The stiffness feature module is used to divide the blank into multiple thin-walled parts processing units, and to pre-set an annular material retention area as an anti-deformation support band at the inner wall connection between each thin-walled part and its next adjacent thin-walled part; based on the anti-deformation support band, the current thin-walled part is rough-machined to obtain a rough-machined surface; based on the rough-machined surface, radial runout data is collected and fitted to obtain the stiffness feature surface;

[0013] The finishing module is used to determine the dynamic offset compensation amount based on the stiffness characteristic surface, and to finish the outer circle and inner hole of the current thin-walled part to obtain a finished surface that retains the anti-deformation support band; based on the finished surface, the actual wall thickness distribution state and the actual wall thickness value are determined.

[0014] The planning and execution module is used to construct a dynamic cutting path mapping domain based on the actual wall thickness distribution and stiffness feature surface. Within the dynamic cutting path mapping domain, the feed path of the cutting tool is decomposed into multiple discretized segments according to the actual wall thickness value and the curvature change of the stiffness feature surface, and a corresponding entry angle and feed rate are set for each segment. Cutting is performed according to the entry angle and feed rate to maintain the finished contour without deformation and complete the cutting.

[0015] The feedback module is used to remove the residual anti-deformation support band on the inner wall of the current thin-walled part after the cutting is completed; during the cutting process, radial runout data is collected to obtain the released profile surface, and the profile correction value is obtained by comparing it with the stiffness feature surface; the profile correction value is used for radial runout data collection and stiffness feature surface fitting before the next thin-walled part is processed.

[0016] Thirdly, a computing device includes:

[0017] One or more processors;

[0018] A storage device for storing one or more programs that, when executed by one or more processors, cause the one or more processors to implement the method.

[0019] Fourthly, a computer-readable storage medium storing a program that, when executed by a processor, implements the method.

[0020] The above-described solution of the present invention has at least the following beneficial effects:

[0021] By dividing the machining into processing units and pre-setting anti-deformation support strips, stable rigid support is provided for subsequent processing, reducing the risk of deformation during rough machining. Radial runout data is collected from the rough-machined surface and fitted with stiffness characteristic surfaces to achieve quantitative collection and concrete representation of the workpiece's rigidity characteristics, solidifying the initial data foundation and ensuring the authenticity and integrity of the data source. Dynamic offset compensation is determined based on the stiffness characteristic surfaces to adapt the compensation parameters to the actual rigidity of the workpiece, conforming to the workpiece's overall rigidity differences. Inner and outer circle finishing is performed while retaining anti-deformation support strips, balancing machining accuracy and workpiece structural stability. Simultaneously, the actual wall thickness distribution and values ​​are verified, constructing a standardized and traceable wall thickness data system. By constructing a dynamic cutting path mapping domain based on wall thickness distribution and stiffness characteristics, multi-dimensional machining data can be fused and correlated, and data operation standards can be unified. The cutting path is decomposed based on data features and segmented cutting parameters are matched to make the cutting strategy fit the actual working conditions of the workpiece. Data-driven path planning can effectively disperse cutting stress and maintain the structural stability of the finished contour. The anti-deformation support strip is accurately removed and the released contour data is collected to fully capture the contour features after the residual stress is released. Contour correction values ​​are extracted through hyperboloid comparison to form a closed-loop data feedback mechanism, which transforms the previous machining data into the basis for subsequent compensation, realizes the iterative reuse of data and the continuous stability of machining accuracy, and ensures the consistency of batch machining. Attached Figure Description

[0022] Figure 1 This is a schematic flowchart of an anti-deformation cutting method for sequential processing of thin-walled aero-engine parts according to an embodiment of the present invention.

[0023] Figure 2 This is a schematic diagram of an anti-deformation cutting system for sequential processing of thin-walled parts of an aero-engine, provided by an embodiment of the present invention. Detailed Implementation

[0024] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0025] like Figure 1 As shown, an embodiment of the present invention proposes a deformation-resistant cutting method for sequential processing of thin-walled parts for aero-engines, the method comprising the following steps:

[0026] Step 100: Divide the blank into multiple processing units for thin-walled parts, and pre-set an annular material retention area as an anti-deformation support strip at the inner wall connection between each thin-walled part and its adjacent next thin-walled part; based on the anti-deformation support strip, perform rough machining on the current thin-walled part to obtain a rough-machined surface; based on the rough-machined surface, collect radial runout data and fit to obtain a stiffness characteristic surface.

[0027] Step 200: Determine the dynamic offset compensation amount based on the stiffness characteristic surface, and perform finishing on the outer circle and inner hole of the current thin-walled part to obtain a finished surface that retains the anti-deformation support band; based on the finished surface, determine the actual wall thickness distribution state and the actual wall thickness value.

[0028] Step 300: Construct a dynamic cutting path mapping domain based on the actual wall thickness distribution and stiffness feature surface; within the dynamic cutting path mapping domain, decompose the feed path of the cutting tool into multiple discretized segments according to the actual wall thickness value and the curvature change of the stiffness feature surface, and set corresponding entry angle and feed rate for each segment; perform cutting according to the entry angle and feed rate to maintain the finished contour without deformation, and complete the cutting.

[0029] Step 400: After the cutting is completed, the residual anti-deformation support band on the inner wall of the current thin-walled part is removed; during the cutting process, radial runout data is collected to obtain the released state profile surface, and the profile correction value is obtained by comparing it with the stiffness feature surface; the profile correction value is used for radial runout data collection and stiffness feature surface fitting before the next thin-walled part is processed.

[0030] In this embodiment of the invention, processing units are divided and anti-deformation support strips are preset to ensure rigid stability during the processing. Radial runout data is collected and stiffness feature surfaces are fitted to fully present the stiffness distribution characteristics of the blank. Dynamic offset compensation is determined based on the stiffness feature surfaces to improve the dimensional accuracy of the finishing process. The actual wall thickness distribution and wall thickness value are obtained by detecting the finishing surface to clearly understand the wall thickness state of the part. A dynamic cutting path mapping domain is constructed by combining the actual wall thickness distribution and the stiffness feature surfaces to adapt to the optimal cutting interval. The feed path is discretely segmented and the cutting angle and feed rate are matched to maintain the force balance during the cutting process and protect the finishing contour. Contour data of the support strip removal process is collected and compared to obtain contour correction values. The correction values ​​are iteratively used for subsequent processing to continuously improve the consistency and stability of batch part processing.

[0031] In a preferred embodiment of the present invention, step 100 involves dividing the blank to be processed into processing units for multiple thin-walled parts, and pre-setting an annular material retention area as an anti-deformation support strip at the inner wall connection between each thin-walled part and its adjacent next thin-walled part; based on the anti-deformation support strip, rough machining is performed on the current thin-walled part to obtain a rough-machined surface; based on the rough-machined surface, radial runout data is collected, and a stiffness characteristic surface is fitted to obtain the surface, including:

[0032] Step 101 involves dividing the blank into multiple thin-walled component processing units. Specifically, this includes: collecting the overall axial length parameters of the blank, retrieving the design axial length parameters of a single aero-engine thin-walled component, and combining the material utilization requirements and sequential processing layout rules of the nesting process. The axial length of the blank is divided by the design axial length of the single thin-walled component and rounded to the nearest integer to determine the total number of processing units that can be divided. According to the calculated number of processing units, the boundary positions are marked sequentially along the blank axis from one end to the other, and axial segmentation is completed at the boundary positions. At the same time, the circumferential processing boundary is determined using the blank inner hole reference surface to complete the circumferential definition. This results in the division of multiple independent and sequentially connected thin-walled component processing units. The axial start coordinates, axial end coordinates, and complete circumferential processing range of each processing unit are determined one by one, forming a multi-unit processing layout that adapts to the sequential processing flow, providing boundary basis for subsequent support strip setting and step-by-step processing.

[0033] Step 102: Based on the processing units of the multiple thin-walled parts, a pre-defined annular material retention area is set as an anti-deformation support band at the connection between the inner wall of each thin-walled part and its adjacent next thin-walled part. Specifically, this includes setting the radial width and axial height of the annular material retention area as the geometric parameters of the anti-deformation support band. Specifically, this includes: using each thin-walled part processing unit as a positioning reference, accurately positioning the interface plane between the inner wall of each thin-walled part and the adjacent subsequent thin-walled part along the blank axis, and extending it continuously in a 360° circumferential direction with the interface plane as a reference surface to form a complete closed annular area, which is the pre-defined annular material retention area. The annular material retention area is a ring-shaped solid material structure that is not cut and removed during the roughing and finishing stages, and is specifically used to provide continuous rigid support for the thin-walled part during the cutting process. This structure can effectively transfer the cutting force to the blank mother body, ensuring that the part does not deform during the cutting process. After the part is safely cut off, it is removed by a separate process.

[0034] The aforementioned annular material retention area is formally designated as the anti-deformation support strip. When pre-setting the annular material retention area, the axial setting position and circumferential closing range of the support strip are determined by the boundary position of the machining unit. Considering the wall thickness parameters of thin-walled parts of aero-engines, the force transmission direction during the cutting process, the overall rigidity support strength requirements, and the process convenience of subsequent turning removal of the support strip, for blanks of difficult-to-machine aerospace materials such as high-temperature alloys and titanium alloys, the support strength is calculated by combining the material strength and the cross-sectional dimensions of the parts. The radial width of the annular material retention area is determined to be 2mm to 5mm. Based on the contour dimensions of the machining unit and the cutting avoidance requirements, the axial height of the adapted machining trajectory is determined to be 3mm to 8mm. The aforementioned radial width and axial height parameters are solidified through size matching and process adaptation. The aforementioned radial width and axial height parameters are determined as the fixed geometric parameters of the anti-deformation support strip. These geometric parameters are directly embedded in the CNC programming logic to guide the compilation of subsequent roughing and finishing CNC programs.

[0035] Step 103: Based on the geometric parameters of the anti-deformation support strip, compile a CNC roughing program including the anti-deformation support strip; execute the CNC roughing program to rough machine the current thin-walled part and obtain a rough machined surface. Specifically, this includes: using the radial width and axial height geometric parameters of the anti-deformation support strip as the core basis for compilation, and combining the material properties of the thin-walled part to be machined, the blank allowance distribution, and the roughing process specifications, determining the basic cutting parameters such as the single-blade cutting depth, circumferential feed rate, spindle speed, and cooling method for the roughing stage; based on the design contour of the thin-walled part and the roughing allowance removal requirements, sequentially plan the layered tool path along the blank axial direction and the continuous feed path along the circumferential direction; throughout the path planning process, integrate the axial and radial intervals corresponding to the anti-deformation support strip... A reserved area is set as a no-cutting zone. When generating the tool path, the reserved area is automatically avoided to ensure that the support strip is not cut and is completely preserved during roughing. Based on the determined cutting parameters, tool movement trajectory and support strip retention constraints, a complete CNC roughing program with an integrated anti-deformation support strip retention strategy is written and integrated segment by segment according to the instruction format that the CNC system can recognize. The completed CNC roughing program is imported into the CNC machining equipment. After the program is verified and the tool is set, the equipment is driven to execute automatically. The tool completes the remaining material removal of the current thin-walled part layer by layer according to the path, cutting depth and feed parameters set in the program, without touching the anti-deformation support strip throughout the process. Finally, a rough-machined surface with dimensional accuracy and surface roughness that meet the requirements of subsequent online radial runout measurement and finishing is formed.

[0036] Step 104 involves evenly distributing multiple radial runout measurement points along the circumference of the rough-machined surface, and obtaining the radial runout value of each radial runout measurement point through an online probe to obtain radial runout data. Specifically, this includes: using the rough-machined surface as a reference measurement surface, and employing a sampling point layout optimization algorithm to complete the layout and determination of the measurement points. The sampling point layout optimization algorithm is a method for adaptively optimizing the number, location, and distribution density of measurement points based on the circumferential contour characteristics of the part, the uniformity requirements of stiffness distribution, and the measurement efficiency requirements. This algorithm can reduce redundant sampling and improve the representativeness and acquisition efficiency of radial runout data while ensuring measurement accuracy.

[0037] The total number of initial sampling points is determined based on the circumferential length and cross-sectional dimensions of the part. The total circumferential angle is then evenly distributed with the total number of initial sampling points to obtain the basic equal angle interval. A sampling point layout optimization algorithm is used to balance the circumferential segments, appropriately increasing the point density in areas prone to stiffness differences while maintaining a normal density in rigid and stable areas. This ultimately determines the circumferential angle position of each measurement point. Based on the optimized angle positions, radial runout measurement points are arranged circumferentially on the rough-machined surface, ensuring complete coverage and even distribution of the measurement points. The online probe mounted on the machining equipment is activated, controlling the probe to contact each measurement point sequentially along the radial direction with a constant contact force. Radial runout values ​​are collected at corresponding locations. The radial runout values ​​collected from all measurement points are recorded sequentially according to the circumferential angle and summarized in order to form a complete radial runout dataset covering the entire circumference and with an optimized layout. For example, for thin-walled cylindrical parts with the entire circumference, the initial total number of samples is set, and the basic interval is obtained by equal angular distribution. The circumference is divided into several uniform segments by the sampling point layout optimization algorithm. The number of sampling points is appropriately increased in the connecting segments near the anti-deformation support belt, and the standard interval is maintained in the stable segments far from the support belt. This makes the sampling layout more consistent with the actual stiffness variation characteristics of the blank and improves the reliability of subsequent data decomposition and surface fitting.

[0038] Step 105 involves performing harmonic decomposition on the radial runout data to extract the main harmonic components reflecting the anisotropic distribution of the blank stiffness. Specifically, this includes: using the optimized sampling radial runout dataset as the basic input data; regularizing and sorting the dataset according to the circumferential angle order to establish a one-to-one correspondence between the circumferential angle and the radial runout value; and performing harmonic decomposition operations point-by-point on the regularized radial runout data to decompose the continuously changing radial runout data into multiple harmonic components of different circumferential orders, different amplitudes, and different phase positions, ensuring that the signal components corresponding to various influencing factors are phased. Mutual separation: Among all harmonic components, high-frequency random disturbance components and noise components caused by measurement errors, equipment vibration, temperature fluctuations, and environmental interference are identified and eliminated one by one. At the same time, systematic error components unrelated to the stiffness characteristics of the blank itself are also eliminated. The dominant harmonic components that directly characterize the circumferential stiffness distribution law of the blank are retained, and the effective signal and interference signal are completely separated. The separated and screened dominant harmonic components are arranged in order of order to form a pure data set that can truly characterize the circumferential stiffness distribution characteristics of the blank, providing real and reliable data support for subsequent harmonic characteristic coefficient extraction and stiffness characteristic field construction.

[0039] Step 106, determining the amplitude, phase, and circumferential order of each major harmonic component as harmonic characteristic coefficients, specifically includes: taking each major harmonic component as the analytical object, performing feature analysis sequentially according to the circumferential order of the harmonic components from low to high; performing numerical analysis and feature extraction on each harmonic component, determining the amplitude of the component in the circumferential distribution, determining the phase position of the component in the circumferential direction, and determining the corresponding circumferential order of the component; and combining the three parameters of amplitude, phase, and circumferential order corresponding to the same harmonic component... The harmonic components are correlated and bound to form characteristic combinations of individual harmonic components. All characteristic combinations of major harmonic components are categorized, regulated, and integrated according to their circumferential order, establishing a one-to-one correspondence between harmonic components and characteristic parameters. The resulting set of correlated and integrated characteristic parameters is standardized to form harmonic characteristic coefficients that can be directly used to quantitatively describe the circumferential stiffness distribution of the blank. These harmonic characteristic coefficients are output in a standardized form as unified data input for the subsequent construction of the stiffness characteristic field, providing complete and standardized data support for the construction of the stiffness characteristic field.

[0040] Step 107: Construct a stiffness characteristic field with circumferential angle as the independent variable based on the harmonic characteristic coefficients. Specifically, this includes: using the standardized harmonic characteristic coefficients as the core calculation basis, setting the integer circumferential angles of the blank section as independent variables in a continuously increasing manner, fully covering the entire range from 0° to 360°; based on the amplitude, phase, and circumferential order parameters corresponding to each harmonic component, performing numerical matching and mapping calculations on each continuously changing circumferential angle, so that each circumferential angle position corresponds to a matching stiffness characteristic value; and completing the calculation for all circumferential angle positions... After numerical mapping, a continuous smooth fitting method is used to connect the discrete stiffness feature values, eliminating numerical abrupt changes between adjacent angular positions, so that the stiffness features change continuously and smoothly with the circumferential angle. Based on the above mapping and fitting results, a unique and stable correspondence between the circumferential angle and the corresponding stiffness feature is established, thereby constructing a single-section stiffness feature field with the circumferential angle as the independent variable. This stiffness feature field can completely present the stiffness differences and distribution trends of each circumferential position of the same section, realizing the continuous characterization and quantitative expression of the circumferential stiffness distribution of the blank section.

[0041] Step 108: Based on the stiffness feature field, the corresponding main harmonic components are repeatedly acquired at different cross-sectional positions along the axial direction of the blank and interpolated along the axial direction to construct a continuous stiffness feature surface. The stiffness feature surface is used to characterize the stiffness distribution characteristics of the blank in the processing section. Specifically, it includes: taking the single-section stiffness feature field as a reference standard, selecting multiple cross-sectional positions at different heights along the axial direction of the blank at equal intervals, and repeatedly performing radial runout data acquisition, harmonic decomposition processing, and main harmonic component extraction operations at each selected cross-sectional position to obtain the stiffness distribution information corresponding to each axial cross-section.

[0042] Based on axial multi-section data, a spatial interpolation algorithm is introduced. This algorithm utilizes known stiffness characteristic parameters of discrete sections to perform continuous numerical calculations between adjacent axial sections, smoothly connecting discrete stiffness information in three-dimensional space and forming a continuous distribution. The main harmonic components obtained from each axial section are correlated and organized with their corresponding axial coordinates. The spatial interpolation algorithm is then used to calculate stiffness parameters point-by-point at each axial position, filling in stiffness characteristic data at uncollected locations between sections. Finally, the interpolated global data undergoes smoothing fitting to eliminate... The abrupt changes and discontinuities in data between adjacent sections form a distribution pattern that is continuous in both the axial and circumferential directions. This results in a continuous stiffness feature surface that simultaneously covers both the circumferential and axial directions. This surface can completely and intuitively characterize the global stiffness distribution of the blank within the current processing section. For example, after selecting several equidistant sections along the blank's axial direction and obtaining the stiffness feature field of each section, the stiffness values ​​of the intermediate positions between sections are recursively extrapolated using a spatial interpolation algorithm. This allows the originally discrete section stiffness information to form a continuous transition in the axial direction, thereby constructing a stiffness feature surface that can completely reflect the global stiffness changes.

[0043] In this embodiment of the invention, the processing layout of the blank to be processed is regulated, and the processing boundaries of each thin-walled part are determined, laying the foundation for subsequent directional processing and precision control, and improving the overall orderliness of processing; the geometric parameters of the anti-deformation support belt are set, and a stable rigid support is formed by relying on the annular material retention area to constrain the deformation trend of the inner wall of the thin-walled part during processing, ensuring the structural stability of the blank processing section; a dedicated CNC program is compiled in combination with the support belt parameters to achieve standardized execution of the roughing process, taking into account both roughing efficiency and blank rigidity protection; circumferentially distributed measuring points are used in conjunction with online probes to collect data, achieving full-domain, real-time acquisition of radial runout of the roughing surface, ensuring the integrity of the original data and the timeliness of acquisition; redundancy is removed through harmonic decomposition. The data is used to extract the core distribution information of the anisotropy of the blank stiffness, filter random interference signals, and make the data more consistent with the actual stiffness characteristics of the blank. The harmonic characteristic coefficients are extracted quantitatively, transforming the abstract stiffness distribution into concrete characteristic parameters, simplifying the data representation form, and providing standardized parameter basis for subsequent field construction. A stiffness characteristic field is built with the circumferential angle as the independent variable to realize the visualization and continuous presentation of the circumferential stiffness distribution of the blank, clearly reflecting the stiffness differences in all directions and refining the stiffness representation dimension. Combined with axial multi-section data interpolation processing, a continuous stiffness characteristic surface is constructed to fully cover the axial and circumferential stiffness distribution of the processed section, realizing the integrated representation of the stiffness characteristics of the entire blank, and providing a rigidity reference basis for subsequent finishing.

[0044] In a preferred embodiment of the present invention, step 200 above involves determining the dynamic offset compensation amount based on the stiffness characteristic surface, and performing finishing on the outer circle and inner hole of the current thin-walled part to obtain a finished surface that retains the anti-deformation support strip; based on the finished surface, determining the actual wall thickness distribution state and the actual wall thickness value includes:

[0045] Step 201: Extract the stiffness difference value of the current thin-walled part at each circumferential position from the stiffness feature surface, and calculate the dynamic offset compensation amount opposite to the stiffness difference value based on the stiffness difference value. Specifically, this includes: retrieving the stiffness feature surface that was previously constructed and completely covers the machining section of the current thin-walled part. Taking a certain type of aerospace aluminum alloy thin-walled sleeve as an example, the workpiece has an outer diameter of 100mm and an axial length of 20mm for the machining section. The 20mm axial start and end interval and the 0° to 360° full circumferential range on the workpiece on the blank are used as the precision... In the preliminary extraction area, starting from the 0° reference position along the circumference, a stiffness characteristic value is read at equal intervals of 15°, for a total of 24 sets of circumferential point data. The arithmetic mean of the stiffness characteristic values ​​of the 24 circumferential positions within the axial section is summed to obtain the overall average stiffness value of the section. The independent stiffness value of each circumferential position is compared with the average stiffness value of the section one by one to determine the stiffness difference value of different circumferential positions of the current thin-walled component relative to the average stiffness state, and to determine the deviation range between the weak stiffness region and the strong stiffness region.

[0046] Based on the aforementioned stiffness differences, reverse adaptation calculations are performed to ensure that the direction of change of the dynamic offset compensation is completely opposite to the trend of machining deformation caused by stiffness deviation. Positive compensation is provided for undercutting deformation in weak rigidity areas and negative compensation is provided for overcutting deformation in strong rigidity areas, thereby offsetting the dimensional deviations caused by uneven stiffness. Combining the cutting parameters of carbide tools, the cutting characteristics of aluminum alloy materials, and the easily deformable and sensitive characteristics of thin-walled parts, the compensation amount is modified item by item, and the compensation amplitude and distribution gradient are finely adjusted to further optimize the adaptability of the compensation amount. Finally, a dynamic offset compensation amount that corresponds one-to-one with 24 circumferential measurement points and is fully adapted to the actual machining state of the workpiece is determined, providing data support for real-time compensation in the subsequent cutting process.

[0047] Step 202: Based on the dynamic offset compensation amount, a finishing toolpath including circumferential variable compensation amount is obtained. The finishing toolpath is executed to machine the outer circle and inner hole of the current thin-walled part to obtain a finished surface with a retaining anti-deformation support band. Specifically, this includes: arranging the 24 circumferential point dynamic offset compensation amounts in an orderly clockwise feed sequence from 0° to 360°. Taking a certain type of aerospace aluminum alloy thin-walled sleeve as an example, the workpiece has an outer diameter of 100mm and a wall thickness of 2mm to 3.5mm. The circumferential feed trajectory of the matching carbide precision turning tool is as follows. Twenty-four key nodes are set up, and the compensation amount corresponding to each trajectory node is matched with the node position one by one. The tool trajectory generation process for finishing external circles and internal holes is embedded point by point. During trajectory generation, the radial position of the tool is corrected in real time according to the compensation amount of each node. Radial micro-feed compensation for weak rigidity points and radial micro-retraction compensation for strong rigidity points enable the tool to synchronously and adaptively change the radial cutting position with the change of circumferential stiffness during continuous circumferential feed, forming a complete finishing toolpath that includes continuously changing circumferential compensation and takes into account both internal and external circle machining.

[0048] After the finishing toolpath is formed, the CNC system path simulation module is activated to perform full-process verification and a special check for cutting interference. The focus is on eliminating the contact risk between the tool and the inner wall annular anti-deformation support strip, locking the axial section where the support strip is located as a no-cutting area, and ensuring that the support strip structure is intact and undamaged throughout the finishing process. After the verification is passed, the toolpath is imported into the horizontal CNC turning center, and the workpiece coordinate system alignment and finishing tool setting are performed. After verifying that the coordinate reference is correct, the equipment is driven to run automatically. First, the outer cylindrical surface of the thin-walled part is finished according to the toolpath, and then the inner hole finishing tool is switched to complete the inner hole surface finishing. The cutting deformation is suppressed by the rigid constraint of the support strip throughout the process. Finally, a finished thin-walled part with full compliance of dimensional accuracy, geometric tolerance and surface quality is obtained, and the anti-deformation support strip is completely preserved.

[0049] Step 203: Select multiple wall thickness sampling points along the axial and circumferential directions on the finished surface. Obtain the actual coordinate values ​​of each wall thickness sampling point using an online probe. Calculate the actual wall thickness value at each sampling point based on the difference between the actual coordinate values ​​and the theoretical coordinate values. Specifically, this includes: taking the finished surface of a certain type of aerospace aluminum alloy thin-walled sleeve as the measurement object, the total axial length of the workpiece is 20mm. Using the intersection of the end faces of the workpiece and the annular anti-deformation support strip as the boundary, divide the axial direction into 5 wall thickness detection sections at equal intervals, ensuring that each section is evenly distributed and complete. It covers the entire processing area of ​​thin-walled parts, with no blind spots in detection. On each wall thickness detection section, with the center of rotation of the part as the reference, the 360° circumference is divided into equal 15° angles, and 24 wall thickness sampling points are arranged on a single section to ensure that the sampling points on the same section are evenly spaced and fully cover the circumferential range. The preset sampling order is the section order from the near end to the far end of the workpiece along the axis. Within the same section, the points are arranged clockwise according to the circumferential angle from small to large. All sampling points eventually form a regular and orderly grid of points, which can comprehensively and truly reflect the wall thickness distribution of the entire thin-walled part.

[0050] The high-precision contact-type online probe on the CNC equipment is activated. After the probe completes automatic homing and initial calibration, it moves sequentially to each wall thickness sampling point according to the preset point sequence. The actual radial coordinates of the inner hole surface and the outer circular surface at each sampling point are collected, and the coordinate data is transmitted to the CNC system in real time for synchronous recording. For the same sampling point, the radial distance difference between the inner and outer hole radial coordinates is calculated to directly obtain the initial measured wall thickness value at that point. The initial measured wall thickness value is compared with the theoretical design wall thickness (2.5mm) to eliminate deviations caused by probe contact error and tooling micro-displacement. After error correction, the actual wall thickness value at the sampling point that is true and reliable and can be used for subsequent analysis is finally obtained.

[0051] Step 204: Arrange the actual wall thickness values ​​at each sampling point according to the circumferential angle and axial position to form an actual wall thickness distribution dataset. Specifically, this includes: assigning a unified position code to all wall thickness sampling points of a certain type of aerospace aluminum alloy thin-walled sleeve; assigning consecutive axial position numbers from S01 to S05 according to the arrangement order of the 5 axial detection sections; assigning consecutive circumferential angle numbers from A01 to A24 according to the clockwise arrangement order of the 24 circumferential sampling points within a single section; and using the combination of axial and circumferential numbers to ensure that each sampling point has an independent and unique combination code, such as S01-A01 and S03-A15; accurately binding the verified actual wall thickness value of each sampling point one-to-one with the corresponding axial position number and circumferential angle number; constructing a 5-row, 24-column two-dimensional data array with the axial position number as the row identifier and the circumferential angle number as the column identifier; and filling all actual wall thickness values ​​one by one into the corresponding array cells to complete the two-dimensional ordered classification and arrangement of the full wall thickness data.

[0052] A closed-loop verification process is performed row by row and column by column on the already arranged two-dimensional data array. Duplicate wall thickness data is checked and removed row by row, and abnormal values ​​deviating from the normal wall thickness range of 2.0mm to 3.0mm are identified column by column. These abnormal values ​​are mostly caused by probe skipping or coordinate offset. After removal, the average values ​​of adjacent points are used to make reasonable supplements to ensure that the data is complete and without gaps. The verified two-dimensional data without redundancy or abnormalities are standardized according to the data format of the CNC system, and the numerical accuracy and storage encoding are unified. The data is then archived to a dedicated machining data module to form a unified, dimensionally complete actual wall thickness distribution dataset that can be directly retrieved and called by subsequent processes. This provides standardized data support for subsequent stiffness correlation analysis and cutting parameter optimization.

[0053] Step 205: Interpolate the actual wall thickness distribution dataset to obtain a continuous wall thickness distribution field, and extract the actual wall thickness values ​​corresponding to each circumferential angle from the wall thickness distribution field as the actual wall thickness distribution state and actual wall thickness value of the thin-walled part before cutting. Specifically, this includes: using a normalized actual wall thickness distribution dataset of a certain type of aerospace aluminum alloy thin-walled sleeve part, after deduplication, verification, and outlier removal, as the basis for calculation. This dataset is a two-dimensional data array of 5 rows and 24 columns. The axial section position parameters of S01-S05 and the circumferential angle parameters of A01-A24 in the array are defined as mutually independent distributions. A bidirectional data computation model with axial and circumferential correlation is constructed to unify the computational benchmark and numerical accuracy of the two dimensions. Based on this model, cubic spline smoothing interpolation is performed simultaneously on the discrete wall thickness sampling data in the axial and circumferential dimensions. Relying on the actual wall thickness values ​​of adjacent sampling points, the blank wall thickness data of the unmeasured area between sampling points is gradually calculated and filled. At the same time, Gaussian filtering algorithm is used to smooth the numerical jumps between points, weaken the small data fluctuations of local points, eliminate abrupt changes and discontinuities between adjacent values, and ensure that the wall thickness value of the whole domain presents a uniform and continuous transition state as the axial and circumferential positions change.

[0054] Based on the complete data after interpolation and filtering, a continuous wall thickness distribution field is constructed that fully covers the 20mm axial machining range and the 360° circumferential range of the thin-walled sleeve, realizing three-dimensional visualization and quantitative representation of the wall thickness status across the entire domain. Using a fixed 1° circumferential angle as the extraction benchmark, the corresponding coordinate positions are located one by one within the continuous wall thickness distribution field, and the wall thickness value at that point is extracted. The extracted values ​​are then arranged in a regular order according to the increasing circumferential angle from 0° to 360°, forming a continuous and smooth wall thickness distribution curve without discontinuity or fluctuation. This curve can intuitively reflect the differential distribution law of wall thickness across the entire circumferential domain of the workpiece, completely and realistically representing the actual wall thickness distribution state and precise wall thickness values ​​at each point of the thin-walled part before the cutting process, providing reliable data support for subsequent dynamic cutting path planning and cutting parameter optimization.

[0055] In this embodiment of the invention, stiffness difference values ​​are extracted based on stiffness characteristic surfaces, and reverse dynamic offset compensation is matched according to the stiffness difference to ensure that the compensation parameters are consistent with the actual rigidity state of the blank, allowing the cutting load and rigidity distribution to be mutually adapted. A circumferential variable compensation finishing toolpath is used for machining, allowing the tool trajectory to adaptively adjust with rigidity differences, stabilizing the machining shape while retaining the anti-deformation support band, and maintaining rigid support during the machining process. The actual wall thickness value is directly calculated by sampling multiple points in the axial and circumferential directions and obtaining coordinates using an online probe, ensuring the wall thickness data is authentic and reliable, and fully reflecting the actual state after machining. The actual wall thickness data is regularized according to circumferential angles and axial positions to form a structured actual wall thickness distribution dataset, improving data readability and subsequent processing efficiency. Interpolation processing of the wall thickness data yields a continuous wall thickness distribution field, fully presenting the global wall thickness variation law, providing a stable and reliable wall thickness basis for the cutting process.

[0056] In a preferred embodiment of the present invention, step 300 involves constructing a dynamic cutting path mapping domain based on the actual wall thickness distribution and stiffness characteristic surface; within the dynamic cutting path mapping domain, the feed path of the cutting tool is decomposed into multiple discretized segments according to the actual wall thickness value and the curvature change of the stiffness characteristic surface, and a corresponding entry angle and feed rate are set for each segment; cutting is performed according to the entry angle and feed rate to maintain the finished contour from deformation, thus completing the cutting, including:

[0057] Step 301: Based on the actual wall thickness distribution state and stiffness characteristic surface, determine the wall thickness distribution function and stiffness distribution function of the current thin-walled part at the cutting section. Specifically, this includes: retrieving the actual wall thickness distribution dataset after verification and interpolation, and the stiffness characteristic surface covering the entire machining section. Taking a thin-walled cylindrical part of a certain type of aero-engine as an example, the annular cutting section with a diameter of 120mm is locked as the core analysis area. Redundant wall thickness and stiffness data on both sides of the axial direction of the cutting section are eliminated through data filtering to remove data interference from non-machining areas. For the filtered wall thickness distribution field, the measured wall thickness values ​​corresponding to 24 measurement points arranged at 15° intervals in the circumferential direction within the cutting section are selected. The wall thickness value range is 2.0mm to 3.5mm. The discrete wall thickness data is processed by a polynomial continuous smooth fitting method to construct a wall thickness distribution function that can reflect the continuous change of wall thickness at the cutting section with the circumferential angle. This function is used to quantitatively characterize the wall thickness difference at each position in the circumferential direction. The continuous wall thickness values ​​can directly provide a basis for subsequent force calculation of cutting units and feed parameter matching.

[0058] Stiffness data of the region coinciding with the annular cutting section in the stiffness feature surface are extracted synchronously. The stiffness feature values ​​corresponding to 24 circumferential points are smoothed and fitted point by point to construct a stiffness distribution function that can reflect the continuous change of stiffness at the cutting section with the circumferential angle. This function is used to intuitively reflect the differences in rigidity bearing capacity at various positions of the workpiece. The continuous stiffness values ​​can be used to define the cutting force safety threshold and optimize the cutting angle direction. Through the above dual data fitting process, the scattered wall thickness and stiffness information are transformed into a continuous and computable function form, unifying the data dimension and calculation standard, eliminating the discontinuity defects of discrete data, and building a solid data foundation for the subsequent dynamic cutting path mapping domain.

[0059] Step 302: Based on the wall thickness distribution function and stiffness distribution function, establish a dynamic cutting path mapping domain with circumferential angle and axial position as independent variables. This dynamic cutting path mapping domain describes the correspondence between the cutting force experienced by the cutting tool at each position and the workpiece stiffness. Specifically, this includes: after completing the construction of the wall thickness distribution function and stiffness distribution function, taking a thin-walled cylindrical component of a certain type of aero-engine as an example, uniformly calibrating the data accuracy and polar coordinate system of the wall thickness distribution function and stiffness distribution function, aligning the coordinate references of the wall thickness data and stiffness data to the cutting section. The rotation center unifies the data sampling interval and numerical accuracy, eliminating calculation deviations caused by data dimensions and coordinate offsets. The circumferential angle of the workpiece's cutting section and the axial positioning position of the tool are used as two independent variables. The circumferential angle is set to a value range of 0° to 360°, and the axial positioning position is based on the center plane of the 120mm diameter cutting section, with a narrow range of ±2mm. The circumferential wall thickness distribution function and the radial stiffness distribution function are simultaneously incorporated into the same polar coordinate calculation framework to carry out data fusion and correlation calculations, thereby building a dynamic cutting path mapping domain adapted to the workpiece.

[0060] This dynamic cutting path mapping domain, through built-in coordinate matching and numerical association rules, precisely binds the operating parameters of the cutting tool, such as feed rate and entry angle, at different circumferential and axial positions, to the measured wall thickness and local stiffness values ​​of the workpiece at corresponding points. For 24 key points at 15° intervals along the circumference, a one-to-one correlation is established between the parameters. Numerical calculations clearly depict the quantitative correspondence between the magnitude and direction of the cutting force and the local stiffness of the workpiece. Small feed rates and high-rigidity entry angles are matched for thin-walled regions, while conventional feed parameters are adapted for thick-walled regions. This achieves the matching of cutting condition parameters with the workpiece's own rigidity and wall thickness properties, providing a dedicated analysis platform for subsequent cutting unit division, segmented entry angle, and feed rate settings.

[0061] Step 303: Within the dynamic cutting path mapping domain, the section to be cut is discretized into multiple cutting units along the circumferential direction. For each cutting unit, based on the actual wall thickness and curvature value of the stiffness characteristic surface, the instantaneous cutting force direction at that cutting unit is determined. Specifically, this includes: entering the established dynamic cutting path mapping domain; taking a thin-walled cylindrical part of a certain type of aero-engine as an example, reading the total circumferential length of the 120mm diameter cutting section of the workpiece; combining the micron-level machining accuracy requirements of the part with the 2mm cutting edge width of the cutting tool, and setting a preset 15° as a fixed circumference. The angular interval division standard divides the entire circumferential section to be cut into 24 uniform and non-overlapping cutting elements in a clockwise direction along the circumference. The number of elements is inversely proportional to the circumferential angular interval, thereby realizing the discretization of the entire cutting area, refining the granularity of local analysis, and ensuring the accuracy of the stress analysis of a single cutting element. For each independent cutting element, the actual wall thickness value of the corresponding circumferential position of the cutting element is retrieved from the dynamic cutting path mapping domain, and the stiffness value of the stiffness feature surface at the center point and surrounding neighborhood of the cutting element is extracted simultaneously.

[0062] After retrieving the numerical values, the rise and fall and deflection trends of the surface curvature are analyzed using numerical difference. Combined with the curvature trend, the optimal force direction for the local stiffness of the cutting unit is determined. Regions with gentle curvature represent uniform stiffness distribution, while regions with abrupt curvature changes represent significant stiffness differences, thus determining the optimal path for force transmission. Based on this, considering the wall thickness bearing capacity and local stiffness characteristics of the cutting unit, and combining the cutting angle and feed direction of the cutting tool, the direction of the cutting force at the moment of cutting is finally determined through numerical comparison and attitude adaptation calculations. For units with weak stiffness, the cutting force direction is adjusted to the side of the cutting unit with a thicker wall, stronger support, and better rigidity, ensuring that the force analysis fully matches the actual rigidity state of the workpiece and avoiding workpiece contour deformation and tool wear caused by force deviations.

[0063] Step 304: Based on the instantaneous cutting force direction, calculate the cutting angle of the cutting tool at the cutting unit that matches the normal of the stiffness feature surface. Specifically, this includes: based on the instantaneous cutting force direction corresponding to each cutting unit, introducing a tangent equation solving algorithm to assist in the cutting angle calculation. This algorithm uses the circular outline of the thin-walled part's cut section as the reference circle, takes the corresponding point of the cutting unit as the tangent point, and calculates the direction of the tangent and the normal at the tangent point through geometric relationships to determine the optimal cutting posture of the tool. Using the normal of the stiffness feature surface at the position of the cutting unit as the reference, and combining the above algorithm, perform a reverse calculation operation. Using the center of the thin-walled part's cross-section as the reference, locate the coordinates of the tangent point of the current cutting unit on the circular cross-section, solve the tangent and normal orientation at the tangent point through the algorithm, calibrate the stiffness normal and the cross-section normal to coincide, adjust the tool's cutting posture, and keep the cutting angle highly compatible with the double normal direction to determine the optimal cutting angle of the cutting tool in the cutting unit.

[0064] For example, for a certain cutting unit, after locating the tangent point with the center of the thin-walled part's cross-section as the reference, the algorithm for solving the tangent equation of the circle is used to determine that the normal of the tangent point points to the center of the circle. Simultaneously, the stiffness normal of the cutting unit is matched, and finally the tool is determined to enter the cutting unit at a small angle offset from the normal. This not only conforms to the direction of optimal stiffness but also avoids cutting vibration. In the process of determining the entry angle, the rigidity distribution characteristics of the workpiece are fully considered, so that the cutting force is transmitted along the direction of higher workpiece stiffness. This avoids overload of local weak rigidity positions due to the deviation of the entry angle, further optimizing the cutting force transmission path, reducing the risk of thin-walled contour deformation from the source, and ensuring the integrity of the finished surface.

[0065] Step 305: Based on the entry angle and actual wall thickness of each cutting unit, and combined with the cutting parameter model, calculate the feed rate corresponding to each cutting unit to control the peak cutting force not to exceed the limit allowed by the workpiece stiffness at that cutting unit. Specifically, this includes: constructing a dedicated cutting parameter model based on the machining characteristics of aerospace thin-walled sleeve parts, using the workpiece local stiffness value, actual wall thickness, optimal tool entry angle, and cutting tool material specifications as core modeling parameters, while simultaneously incorporating the cutting resistance characteristics of difficult-to-machine aerospace materials such as high-temperature alloys or titanium alloys, and combining the critical threshold of plastic deformation of thin-walled parts with the cutting tool's cutting force. Process parameters such as cutting durability index are corrected and optimized item by item to build the basic calculation framework of the model. After the model is built, actual test training is carried out. Multiple sets of on-site machining data under different wall thickness gradients, different stiffness distributions, and different cutting angles are collected, including measured cutting force values, workpiece deformation detection values, and machining surface quality data. The above data are used as training samples and input into the model. By repeatedly comparing the sample data with the theoretical calculation values, the weight parameters and compensation coefficients inside the model are gradually corrected to eliminate the deviation between theoretical calculation and actual machining until the model output parameters meet the requirements of high-precision thin-walled part machining.

[0066] By combining the fully trained cutting parameter model, the optimal entry angle and actual wall thickness of each cutting unit are imported into the model as core input parameters. Combined with the stiffness distribution value of the corresponding unit, the cutting load threshold of the cutting unit at the current entry angle is calculated to determine the maximum allowable cutting force upper limit of the workpiece stiffness. Based on the actual wall thickness of the unit, a graded adaptation calculation is performed. For units with thicker walls, the feed rate is appropriately increased in combination with the load threshold, while for units with thinner walls, the feed rate is simultaneously reduced to reduce the cutting load. Through multiple rounds of iterative calculations within the model, the optimal parameter values ​​are gradually approximated, and finally, a dedicated feed rate adapted to each cutting unit is obtained. The peak cutting force is checked in real time throughout the entire calculation process to control it within the limit that the workpiece stiffness of the cutting unit can withstand, taking into account both cutting efficiency and the safety of thin-walled part forming. From the parameter level, the problem of workpiece contour deformation and surface tool damage caused by excessive feed is avoided, and the cutting parameters are matched with the local working conditions.

[0067] Step 306: Arrange the approach angle and feed rate of each cutting unit in circumferential order to obtain a continuous cut-off tool feed path composed of multiple discretized segments. Each discretized segment has an independent approach angle and feed rate. Specifically, this includes: extracting the three sets of core parameters corresponding to each cutting unit: circumferential angle identifier, optimal approach angle, and suitable feed rate; establishing a fixed binding relationship between a single cutting unit and the three sets of parameters; classifying and arranging all units in circumferential angle order starting from 0° and increasing clockwise to 360° to form an ordered discrete parameter sequence; after the arrangement is completed, traversing adjacent cutting units sequentially and calculating the difference in their approach angles and feed rates respectively. The difference in feed rate is pre-set with a preset stable threshold, taking into account the machining accuracy level of thin-walled parts, tool cutting performance, and workpiece rigidity characteristics. The stable threshold for the difference in the entry angle is 3 degrees, and the stable threshold for the difference in the feed rate is 10 mm per minute. The obtained difference is compared with the above preset stable threshold one by one. For adjacent points where the difference in the entry angle exceeds 3° or the difference in the feed rate exceeds 10 mm per minute, a linear smooth transition algorithm is activated to perform numerical correction. The median value of the parameters of adjacent units is taken as the transition parameter, and the parameter difference between adjacent points is gradually reduced until the difference is controlled within the corresponding stable threshold range, so as to completely eliminate the parameter abrupt breakpoints and prevent the tool from having sudden changes in posture, cutting vibration, or workpiece ripples during operation.

[0068] After completing the parameter smoothing correction, all discretized parameters are sequentially connected end-to-end in the clockwise circumferential cutting direction. A corresponding number of transition parameter nodes are added between adjacent segments to ensure seamless connection between each segment and the next, ultimately forming a continuous, complete, uninterrupted, and abrupt cut-off tool feed path. This path retains the independent entry angle and feed rate of each discretized segment, and the parameters of adjacent segments transition smoothly without abrupt fluctuations. This ensures that during the circumferential cutting process, the operating parameters can match the dynamic changes in the workpiece wall thickness and stiffness in real time, perfectly conforming to the actual machining state of the workpiece. This provides reliable path support for the subsequent smooth execution of the cut-off process and ensures the precision of the finishing contour.

[0069] Step 307: Perform the cutting operation according to the continuous cutting tool feed path, so that the cutting force is transmitted sequentially to the anti-deformation support belt and the blank body along the direction with higher stiffness corresponding to each segment, thereby maintaining the geometric accuracy of the finished contour during the cutting process to complete the cutting. Specifically, this includes: converting the continuous cutting tool feed path into a G-code machining program that can be recognized by the CNC system, importing it into the matching horizontal CNC turning machining center, taking a certain type of aero-engine titanium alloy thin-walled cylindrical part as an example, the workpiece has an outer diameter of 120mm and a wall thickness of 2mm to 3.5mm, starting the built-in VERICUT path simulation module of the equipment, carrying out full-stroke path interference verification, checking the collision risk between the cutting tool and the blank body, the three-jaw tooling fixture, and the inner wall annular anti-deformation support belt one by one, and performing tool setting verification at the same time, using an edge finder to calibrate the relative position of the cutting tool tip coordinates and the workpiece cutting section, confirming that there is no interference or collision on the path, no coordinate deviation, and no parameter input errors, locking the machining parameters such as spindle speed and feed rate and starting the cutting command.

[0070] After receiving the cutting command, the equipment drives the carbide cutting tool to run along a predetermined continuous feed path, performing the cutting operation step by step in a clockwise direction around the circumference. During the cutting process, the tool matches the entry angle and feed rate of each 15° segment in real time, so that the cutting force is transmitted along the high rigidity direction corresponding to each segment, and is sequentially transmitted to the annular anti-deformation support belt and the blank body. Relying on the rigid bearing effect of the support belt and the overall rigidity of the blank body, the cutting load and residual stress are quickly dispersed, avoiding elastic or plastic deformation of the thin-walled contour after finishing. During the cutting process, the built-in infrared monitoring unit monitors the tool's running posture and cutting status in real time, keeping the spindle speed stable and the feed without fluctuations until the full circumference cutting action is completed. After the tool exits the cutting area along the retraction path, the separation of the thin-walled part and the blank body is completed. The geometric accuracy and surface quality of the finished contour are guaranteed throughout the process, and the nesting cutting process is successfully completed.

[0071] In this embodiment of the invention, a dual distribution function is determined based on the actual wall thickness distribution and stiffness characteristic surface to achieve normalized integration of wall thickness and stiffness data, laying a solid foundation for subsequent data processing. A multi-variable dynamic cutting path mapping domain is established to achieve quantitative correlation between cutting force and workpiece stiffness, improving the matching degree between cutting parameters and the actual state of the workpiece. The circumferential section of the cutting section is discretized into cutting units to refine the local working condition analysis dimension and achieve accurate determination of the stress state of a single cutting unit. The cutting force transmission direction is optimized by combining the stiffness normal back calculation of the cutting angle to ensure that the force transmission path conforms to the stiffness distribution of the workpiece itself. The segmented feed rate is calculated based on the cutting parameter model to achieve controllable constraint of the cutting force peak and avoid deformation risks caused by local overload. The segmented parameters are integrated by circumferential sorting to construct a discretized continuous feed path, achieving orderly connection and global adaptation of cutting parameters. The cutting operation is executed along the dynamic path to guide the cutting force to be efficiently transmitted to the blank parent body, stabilize the geometric accuracy of the finishing contour, and ensure the smooth completion of the cutting process.

[0072] In a preferred embodiment of the present invention, step 400, after the cutting is completed, removes the residual anti-deformation support strip on the inner wall of the current thin-walled part; during the cutting process, radial runout data is collected to obtain the released contour surface, and compared with the stiffness feature surface to obtain the contour correction value; the contour correction value is used for radial runout data collection and stiffness feature surface fitting before the next thin-walled part is processed, including:

[0073] Step 401: After completing the cutting of the current thin-walled part, the remaining anti-deformation support band on the inner wall of the thin-walled part is removed by a turning operation to obtain the inner surface of the thin-walled part after releasing residual stress. Specifically, this includes: after completing the cutting operation of the current thin-walled part and the tool exiting the cutting area, positioning the remaining annular anti-deformation support band on the inner wall of the thin-walled part, marking the axial start and end range and radial thickness of the support band, and retrieving the preset precision turning cutting parameters. These preset precision turning cutting parameters include: single-layer cutting depth, axial feed rate, and spindle speed. The parameter presets are determined comprehensively based on the material characteristics of the aerospace thin-walled part, the radial width of the support band, the rigidity and strength of the workpiece, and the protection requirements of the finishing surface, taking into account both cutting efficiency and workpiece anti-deformation requirements; and selecting a tool tip circle. The carbide precision turning tool with arc adaptability employs a layered micro-turning method to gradually remove the annular anti-deformation support strip. During the turning process, the depth of cut and axial feed rate of each layer are controlled, avoiding the already machined inner and outer contour surfaces of the thin-walled part throughout the process. Micro-feeding is used to prevent excessive cutting force from causing secondary deformation of the workpiece. At the same time, the cutting status is monitored in real time to ensure that only the support strip body is removed without damaging the inner wall reference surface of the workpiece. After the support strip is completely removed, it is left to stand for a preset time to allow the residual stress inside the thin-walled part to be fully released, eliminating the rigid constraint effect of the support strip. Finally, a real inner surface of the thin-walled part with no external force restraint and complete release of residual stress is obtained, providing a flat, interference-free, and actual measured surface for subsequent online probe acquisition of radial runout data.

[0074] Step 402: During the removal of the anti-deformation support strip, radial runout data from multiple measurement points along the circumferential and axial directions are collected using an online probe to obtain a radial runout dataset during the removal process. Specifically, this includes: simultaneously activating the high-precision online probe mounted on the CNC machining equipment while starting the anti-deformation support strip removal process, completing the initial tool alignment between the probe and the inner wall of the workpiece, and eliminating acquisition errors caused by probe installation and coordinate deviations; retrieving a preset measurement path, which is a grid-like measurement trajectory covering the entire inner wall of the thin-walled part. The preset path is determined by combining the axial length of the thin-walled part, the circumferential contour, the residual stress release law, and the removal range of the support strip. Measurement sections are divided along the axial direction, and measurement points are arranged along the circumferential direction to ensure that the points are evenly distributed and there are no blind spots, so as to fully capture the contour deformation characteristics; performing the acquisition operation according to the preset measurement path, arranging measurement points at equal angles along the entire circumference of the thin-walled part, and dividing multiple measurement sections at fixed intervals along the axial direction of the workpiece, so that the measurement points are evenly covered in a grid pattern covering the entire inner wall of the thin-walled part, taking into account the full-area characterization of residual stress release.

[0075] During the cutting process of the support strip, the online probe moves synchronously with the cutting process. Using a constant force contact acquisition method, it collects the radial runout values ​​of each measurement section and each circumferential point one by one. After each data point is collected, it is immediately temporarily stored and verified locally. After invalid data is removed, it is transmitted in real time to the dedicated data module of the CNC system for classification and storage. After the support strip is completely removed, all valid data are organized and integrated according to the axial section order and the circumferential angle order, and finally a radial runout dataset covering the entire circumferential and axial cutting process is formed, which completely preserves the dynamic deformation data of the contour during the gradual release of residual stress.

[0076] Step 403: Perform surface fitting on the radial runout dataset during the cutting process to obtain a released state contour surface that reflects the actual contour after the residual stress is released. Specifically, this includes: retrieving the collected and normalized full radial runout dataset. Taking a certain type of thin-walled cylindrical part of aerospace as an example, the dataset contains radial runout values ​​at circumferential intervals of 15° and axial intervals of 5mm. Perform the first round of preprocessing on the dataset, compare the radial runout values ​​at each point with the overall data mean, and filter out abnormal collected values ​​that deviate from the normal value range. Such abnormal values ​​are mostly caused by instantaneous interference of the probe and cutting vibration. Data purification is completed by direct rejection, retaining valid runout data that fits the actual deformation state.

[0077] Based on this, a smooth surface fitting algorithm suitable for thin-walled rotating parts is adopted to perform a global fitting operation on the effective runout data after purification. During the operation, the circumferential angle and axial positioning position corresponding to each measurement point are used as two-dimensional coordinate parameters, and the measured radial runout values ​​matched at each point are used as surface height characterization parameters. The fitting operation logic is substituted point by point, and the contour data between points is gradually iterated to complete the data, eliminate the discontinuity defects of discrete data, and finally fit to generate a continuous, smooth surface without abrupt changes. This surface is the release state contour surface that can truly reflect the actual contour of the workpiece after the residual stress is completely released, completing the visualization transformation of discrete runout data into a continuous three-dimensional contour, laying the foundation for subsequent surface comparison deviation calculation.

[0078] Step 404 involves comparing the released contour surface with the stiffness feature surface point by point, calculating the normal deviation value at each corresponding position, and constructing a deviation distribution dataset from all normal deviation values. Specifically, this includes: retrieving the stiffness feature surface fitted during the pre-processing stage from the CNC system's surface database, simultaneously retrieving the processed released contour surface, using the rotation centerline of the thin-walled part itself as a unified common reference, and combining it with the workpiece end face positioning origin to establish a globally unified three-dimensional machining coordinate system; importing the two surfaces into this coordinate system, performing a reference coincidence calibration operation, and eliminating surface modeling and data... The relative positional deviation caused by transmission ensures that the hyperboloids are under the same spatial reference. After the coordinate calibration is completed, the measurement points of the two surfaces are locked and matched one by one according to the principle that the circumferential angle and axial position are completely consistent. The point index relationship is established in advance to ensure that each measured release contour point can correspond to the unique theoretical point on the stiffness characteristic surface, completely avoiding the problems of mismatch, omission, and repeated matching of points. Taking a certain type of aerospace thin-walled ring part as an example, the inner wall of the workpiece is arranged with regular grid points at 15° intervals in the circumferential direction and 5mm intervals in the axial direction. These grid points are used as fixed comparison points to achieve accurate correspondence of the contour points in the entire domain without blind spots.

[0079] Based on the aforementioned one-to-one matching points, the special calculation logic for normal deviation is initiated. For each matching point, the tangential normal vector of the released profile surface and the theoretical tangential normal vector of the stiffness characteristic surface at that point are calculated. Through spatial distance calculation, the difference in perpendicular distance between the two sets of normal vectors is obtained. This difference is the normal deviation value for that point, and the deviation value is marked with a positive or negative sign to represent the profile offset direction. After performing the above deviation calculation for all matching points one by one, all normal deviation values ​​are systematically categorized and arranged according to a fixed order in which the circumferential angle increases from 0° to 360° and the axial position extends from the end closer to the blank to the end closer to the free end. During the arrangement process, values ​​are simultaneously filtered to remove random interference values ​​below a preset small deviation threshold, which is set to 2.0μ. The threshold m is not a fixed value. Its determination is based on three main aspects: firstly, it conforms to the dimensional tolerance zone requirements marked on the design drawings of thin-walled aerospace parts, selecting 1 / 10 of the lower limit of the tolerance zone as the benchmark threshold to avoid the mis-screening of effective machining deviations; secondly, it combines the measurement accuracy limit of the online measurement unit of the CNC system, matching the repeatability accuracy of the probe with the data acquisition noise range to eliminate the measurement error of the equipment itself; and thirdly, it adapts to the residual stress release law of difficult-to-machine materials such as titanium alloys, distinguishing between random measurement fluctuations and contour deviations caused by real residual stress, retaining only the effective deviation data reflecting the release of residual stress, and finally forming a regular and orderly deviation distribution dataset covering the entire circumference and axial direction of the workpiece, fully realizing the full-dimensional quantitative comparison between the actual released contour and the theoretical stiffness contour, providing reliable data support for the subsequent extraction of contour correction values.

[0080] Step 405 involves statistically analyzing the deviation distribution dataset to extract the main deviation components reflecting the residual stress release effect as contour correction values. Specifically, this includes: retrieving the deviation distribution dataset, initiating a specialized statistical analysis process for residual stress deformation, performing a full-domain statistical analysis of the previously validated effective normal deviation values ​​in the dataset, sequentially calculating the arithmetic mean, sample variance, and population standard deviation of all deviation values, and accurately classifying systematic deviations and random errors based on three types of statistical characteristic values; and determining discrete values ​​that deviate significantly from the overall mean and have an excessively large variance. The minor component of random error caused by external interference is identified as the core component of systematic deviation caused by residual stress release. The stable values ​​that are concentrated near the mean and have very small fluctuations are identified as the core component of systematic deviation caused by residual stress release. Taking a certain type of thin-walled cylindrical part of aerospace as an example, the discrete deviations of 3μm to 5μm that appear sporadically in the deviation distribution data of this workpiece are random errors caused by cutting vibration and instantaneous poor contact of the probe, accounting for less than 5%. The remaining stable deviations of 0.8μm to 1.2μm that are concentrated are the core deviation components caused by residual stress release after the support strip is removed. Based on this, the two types of deviation components can be efficiently and accurately distinguished.

[0081] After qualitatively distinguishing the deviation components, all minor deviation values ​​corresponding to random errors in the dataset are screened and removed one by one, while the core deviation components that can centrally reflect the residual stress release effect are fully retained. The retained core deviation components are then subjected to refined numerical normalization, uniformly retaining two significant decimal places and removing redundant tails that have no practical processing significance. The normalized core deviation values ​​are then weighted and averaged, and the proportion is calculated according to the circumferential and axial weight coefficients of each point to eliminate the interference of small fluctuations at individual points and weaken the impact of extreme small deviations on the overall result. Finally, the unique value after weighted average processing is determined as the contour correction value for this round of processing. This value can fully characterize the overall contour offset trend and offset caused by residual stress release, serving as the basis for systematic error compensation in subsequent processing and completely eliminating the impact of residual stress deformation on the accuracy of continuous processing.

[0082] Step 406: The contour correction value is stored in the compensation parameter set of the CNC device as a benchmark calibration value for radial runout data acquisition and stiffness feature surface fitting before the next thin-walled part is processed. Specifically, this includes: extracting the contour correction value determined after weighted averaging, and performing encoding processing according to the standardized parameter encoding rules built into the CNC device. The encoding rules are formulated by combining the workpiece model, processing sequence number, and compensation parameter type, and assigning a globally unique digital code to the contour correction value. At the same time, the numerical accuracy and data format of the correction value are double-checked to check for numerical overflow and format incompatibility issues, ensuring that the parameter can be normally recognized and read by the CNC system compensation module. After the verification is passed, the encoded contour correction value is entered into the cutting compensation parameter set built into the CNC device and archived in the special compensation folder for thin-walled part nesting processing. It is stored in order of processing batch. Taking a thin-walled cylindrical part of a certain type of aero-engine as an example, the 1.0μm contour correction value obtained in this round of processing is encoded as a special compensation parameter with the corresponding workpiece model and batch number, completing the parameter solidification storage and one-click retrieval binding to prevent parameter loss or confusion.

[0083] After completing parameter storage and archiving, a unique association identifier is established between the contour correction value and the next thin-walled part of the same model to be processed. The association identifier is synchronously embedded into the CNC machining task sheet and process flow card, ensuring that the correction value is only applicable to the preliminary preparation process of the next workpiece. The contour correction value is set as the benchmark calibration value and pre-associated with the radial runout data acquisition module and stiffness feature surface fitting module of the next workpiece through the internal link of the CNC system. The automatic calling permission of the module is set to ensure that when entering the machining process of the next workpiece, the correction value can be directly retrieved for benchmark calibration, realizing the closed-loop transmission and efficient reuse of machining error data, and providing a preliminary calibration basis for subsequent radial runout data correction and stiffness feature surface fitting.

[0084] Step 407: When machining the next thin-walled part, the contour correction value is called to correct the radial runout data, obtaining the corrected radial runout data. Specifically, this includes: entering the machining preparation stage for the next thin-walled part of the same model and specification; completing the tooling fixture reset, blank positioning and clamping, and online probe return operation; initiating the pre-machining calibration process built into the CNC device; automatically reading the previously established association identifier; and retrieving the contour correction value for the corresponding batch and model from the thin-walled part-specific compensation catalog of the cutting compensation parameter set through the identifier index. After the parameters are retrieved, the system automatically verifies their validity to ensure that the correction values ​​are not lost or tampered with. Simultaneously, the online probe is started and, according to the predetermined grid point layout rules, it collects radial runout data of the inner wall of the next workpiece to be processed. Taking a thin-walled cylindrical part of a certain type of aero-engine as an example, the system retrieves the 1.0μm contour correction value obtained from the previous round of processing and simultaneously collects the original radial runout data of the workpiece at circumferential intervals of 15° and axial intervals of 5mm. Throughout the process, it ensures that the collected points correspond one-to-one with the correction value matching points, without misalignment or omission.

[0085] After completing the precise parameter retrieval and complete acquisition of raw data, the numerical calibration operation logic is automatically initiated. A one-to-one correspondence between points is established, and the measured values ​​of each point in the original radial runout dataset are directionally superimposed with the contour correction values ​​for compensation. Systematic deviations caused by residual stress release in the raw data are precisely calibrated item by item to correct contour offset errors. After the operation is complete, a secondary data verification process is automatically initiated to compare the data fluctuation amplitude before and after correction, eliminate minor anomalies caused by point interference during the operation, and thoroughly filter out errors caused by residual stress release. After successful verification, the corrected data is standardized and archived as calibrated precise radial runout data. This data is then directionally stored in the dedicated machining data module of the CNC system, providing reliable and unbiased raw data support for subsequent harmonic decomposition, stiffness feature field construction, and surface fitting.

[0086] Step 408: Based on the corrected radial runout data, harmonic decomposition is performed sequentially to extract the main harmonic components. The corresponding harmonic characteristic coefficients are determined according to the main harmonic components. A stiffness characteristic field is constructed based on the harmonic characteristic coefficients, and interpolation is performed along the axial direction to obtain a new stiffness characteristic surface. This achieves consistent accuracy control in continuous sequential processing. Specifically, this includes: retrieving the corrected radial runout data; performing format normalization preprocessing on the data set to unify the data dimension and accuracy; starting the harmonic decomposition operation process; using Fourier series decomposition logic to decompose the one-dimensional discrete radial runout data into harmonic components of different orders, frequencies, and amplitudes; classifying and separating low-order and high-order harmonic components according to their order; performing amplitude quantization comparison and sorting on various harmonic components; and setting the amplitude... The threshold screening rule retains low-order main harmonic components with amplitudes exceeding the threshold that play a dominant role in workpiece stiffness and deformation, while eliminating high-order harmonic components with amplitudes below the threshold that are merely interference signals. Combining the measured amplitude and phase parameters of the screened main harmonic components, the characteristic coefficients of each harmonic group are determined through least-squares numerical fitting calculations. Taking a thin-walled cylindrical component of a certain type of aero-engine as an example, the corrected radial runout data of 24 measurement points spaced 15 degrees circumferentially and 8 measurement sections spaced 5 mm axially are used. After harmonic decomposition, first- and second-order low-order main harmonic components are screened out, while third-order and higher-order interference harmonics are eliminated. The amplitude and phase coefficients of the first and second-order harmonics are obtained through fitting calculations, thus completing the accurate extraction of core harmonic characteristic parameters.

[0087] After extracting the harmonic characteristic coefficients, a mapping relationship model between the harmonic characteristic coefficients and the circumferential and axial stiffness of the thin-walled part is established. Based on this mapping relationship and combined with the workpiece's geometric dimensions, a global stiffness characteristic field covering the entire circumferential and axial directions of the workpiece is constructed to intuitively represent the distribution of stiffness strength at various positions of the workpiece. Piecewise linear interpolation is initiated along the axial direction of the thin-walled part. Using the points where the axial stiffness data has been collected as reference nodes, the slope of stiffness change between adjacent nodes is calculated. The stiffness data at the intermediate missing positions is filled in segment by segment to ensure that the axial stiffness data is continuous and uninterrupted, and transitions without abrupt changes. Based on the complete stiffness data of the full dimensions after interpolation, a smooth surface fitting algorithm is used for global fitting to remove data fluctuations and burrs, resulting in a new stiffness characteristic surface that is suitable for the next workpiece to be processed. This surface can match the original wall thickness and anisotropic stiffness distribution characteristics of the workpiece in real time. Relying on the contour correction value passed by the closed loop mentioned above, the processing data is linked and iterated to effectively avoid the accuracy deviation caused by residual stress deformation and ensure the consistency of dimensional accuracy and processing stability of multi-piece nesting sequential processing.

[0088] In this embodiment of the invention, the residual anti-deformation support strip on the inner wall is removed by turning, releasing the residual stress inside the thin-walled part and obtaining the true workpiece state without rigid constraints, laying the foundation for subsequent data acquisition. Radial runout data is collected from multiple points and dimensions using an online probe, forming a complete runout dataset that retains all deformation information during the removal process, improving the comprehensiveness of data acquisition. Surface fitting processing is performed on the radial runout dataset, transforming discrete data into a continuous contour surface, intuitively representing the actual contour shape after residual stress release, and optimizing the data presentation. The released contour surface is compared point-by-point with the stiffness characteristic surface to obtain a standardized deviation distribution dataset, realizing the correlation between the actual contour and the theoretical shape. This paper discusses the quantitative comparison of contours and refines the dimensions of data processing. Statistical analysis is conducted on the deviation dataset to extract core deviation components, eliminate redundant and interfering data, and form accurate contour correction values ​​to improve data utilization efficiency. The contour correction values ​​are stored in the CNC compensation parameter set to establish a closed-loop transmission mechanism for machining data, providing calibration basis for subsequent workpiece machining and realizing the reuse and inheritance of machining data. The radial runout data of the next workpiece is calibrated using the contour correction values ​​to eliminate the error influence caused by the previous stress release and ensure the reliability of the initial data. Based on the correction data, harmonic decomposition and interpolation are performed to construct a new stiffness feature surface, realizing iterative optimization of machining parameters and improving the accuracy consistency of continuous sequential machining.

[0089] like Figure 2 As shown, embodiments of the present invention also provide a deformation-resistant cutting system for sequential processing of thin-walled aero-engine parts, comprising:

[0090] The stiffness feature module is used to divide the blank into multiple thin-walled parts processing units, and to pre-set an annular material retention area as an anti-deformation support band at the inner wall connection between each thin-walled part and its next adjacent thin-walled part; based on the anti-deformation support band, the current thin-walled part is rough-machined to obtain a rough-machined surface; based on the rough-machined surface, radial runout data is collected and fitted to obtain the stiffness feature surface;

[0091] The finishing module is used to determine the dynamic offset compensation amount based on the stiffness characteristic surface, and to finish the outer circle and inner hole of the current thin-walled part to obtain a finished surface that retains the anti-deformation support band; based on the finished surface, the actual wall thickness distribution state and the actual wall thickness value are determined.

[0092] The planning and execution module is used to construct a dynamic cutting path mapping domain based on the actual wall thickness distribution and stiffness feature surface. Within the dynamic cutting path mapping domain, the feed path of the cutting tool is decomposed into multiple discretized segments according to the actual wall thickness value and the curvature change of the stiffness feature surface, and a corresponding entry angle and feed rate are set for each segment. Cutting is performed according to the entry angle and feed rate to maintain the finished contour without deformation and complete the cutting.

[0093] The feedback module is used to remove the residual anti-deformation support band on the inner wall of the current thin-walled part after the cutting is completed; during the cutting process, radial runout data is collected to obtain the released profile surface, and the profile correction value is obtained by comparing it with the stiffness feature surface; the profile correction value is used for radial runout data collection and stiffness feature surface fitting before the next thin-walled part is processed.

[0094] It should be noted that this system is a system corresponding to the above method. All implementation methods in the above method embodiments are applicable to this embodiment and can achieve the same technical effect.

[0095] Embodiments of the present invention also provide a computing device, including: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0096] Embodiments of the present invention also provide a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the method described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0097] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0098] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0099] In the embodiments provided by this invention, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0100] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0101] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0102] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0103] Furthermore, it should be noted that in the apparatus and method of the present invention, it is obvious that the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered equivalent solutions of the present invention. Moreover, the steps performing the above series of processes can naturally be executed in the order described, but are not necessarily required to be executed in chronological order; some steps can be executed in parallel or independently of each other. Those skilled in the art will understand that all or any step or component of the method and apparatus of the present invention can be implemented in any computing device (including processors, storage media, etc.) or network of computing devices, in hardware, firmware, software, or a combination thereof. This can be achieved by those skilled in the art using basic programming skills after reading the description of the present invention.

[0104] Therefore, the object of the present invention can also be achieved by running a program or a set of programs on any computing device. The computing device can be a known general-purpose device. Therefore, the object of the present invention can also be achieved simply by providing a program product containing program code implementing the method or apparatus. That is, such a program product also constitutes the present invention, and the storage medium storing such a program product also constitutes the present invention. Obviously, the storage medium can be any known storage medium or any storage medium developed in the future. It should also be noted that in the apparatus and method of the present invention, it is obvious that the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered equivalent to the present invention. Furthermore, the steps performing the above series of processes can naturally be performed in the order described, but are not necessarily required to be performed in chronological order. Some steps can be performed in parallel or independently of each other.

[0105] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method of preventing deformation and cutting of an aeroengine thin-walled part during sequential machining, characterized in that, The method includes: Step 100: Divide the blank into multiple processing units for thin-walled parts, and pre-set an annular material retention area as an anti-deformation support strip at the inner wall connection between each thin-walled part and its adjacent next thin-walled part; based on the anti-deformation support strip, perform rough machining on the current thin-walled part to obtain a rough-machined surface; based on the rough-machined surface, collect radial runout data and fit to obtain a stiffness characteristic surface. Step 200: Determine the dynamic offset compensation amount based on the stiffness characteristic surface, and perform finishing on the outer circle and inner hole of the current thin-walled part to obtain a finished surface that retains the anti-deformation support band; based on the finished surface, determine the actual wall thickness distribution state and the actual wall thickness value. Step 300: Construct a dynamic cutting path mapping domain based on the actual wall thickness distribution and stiffness feature surface; within the dynamic cutting path mapping domain, decompose the feed path of the cutting tool into multiple discretized segments according to the actual wall thickness value and the curvature change of the stiffness feature surface, and set corresponding entry angle and feed rate for each segment; perform cutting according to the entry angle and feed rate to maintain the finished contour without deformation, and complete the cutting. Step 400: After the cutting is completed, the residual anti-deformation support band on the inner wall of the current thin-walled part is removed; during the cutting process, radial runout data is collected to obtain the released state profile surface, and the profile correction value is obtained by comparing it with the stiffness feature surface; the profile correction value is used for radial runout data collection and stiffness feature surface fitting before the next thin-walled part is processed.

2. The method of claim 1, wherein Step 100 includes: Divide the blank into multiple thin-walled part processing units; According to the processing unit of the plurality of thin-walled parts, at the connection between the inner wall of each thin-walled part and the next adjacent thin-walled part, an annular material retention area is pre-set as an anti-deformation support strip. Specifically, the radial width and axial height of the annular material retention area are set as the geometric parameters of the anti-deformation support strip. Based on the geometric parameters of the anti-deformation support strip, a CNC roughing program containing the anti-deformation support strip is compiled; the CNC roughing program is executed to rough machine the current thin-walled part and obtain a rough machined surface.

3. The method of claim 2, wherein, Step 100 further includes: Multiple radial runout measurement points are evenly distributed along the circumference on the rough-machined surface. The radial runout value of each radial runout measurement point is obtained through an online probe to obtain radial runout data. Harmonic decomposition was performed on the radial runout data to extract the main harmonic components that reflect the anisotropic distribution of the blank stiffness. Based on the main harmonic components, the amplitude, phase, and circumferential order of each component are determined as harmonic characteristic coefficients. A stiffness characteristic field with circumferential angle as the independent variable is constructed based on the harmonic characteristic coefficients. Based on the stiffness characteristic field, the corresponding main harmonic components are repeatedly obtained at different cross-sectional positions along the axial direction of the blank and interpolated along the axial direction to construct a continuous stiffness characteristic surface. The stiffness characteristic surface is used to characterize the stiffness distribution characteristics of the blank in the processing section.

4. The method of claim 3, wherein, Step 200 includes: Extract the stiffness difference value of the current thin-walled component at each circumferential position from the stiffness feature surface, and calculate the dynamic offset compensation amount that is opposite to the stiffness difference value based on the stiffness difference value. Based on the dynamic offset compensation amount, a finishing toolpath containing circumferential variable compensation amount is obtained. The finishing toolpath is executed to process the outer circle and inner hole of the current thin-walled part, and a finishing surface with anti-deformation support band is obtained. Multiple wall thickness sampling points are selected along the axial and circumferential directions on the finished surface. The actual coordinate values ​​of each wall thickness sampling point are obtained by an online probe. The actual wall thickness value at each sampling point is calculated based on the difference between the actual coordinate values ​​and the theoretical coordinate values. The actual wall thickness values ​​at each sampling point are arranged according to the circumferential angle and axial position to form an actual wall thickness distribution dataset; Interpolation processing is performed on the actual wall thickness distribution dataset to obtain a continuous wall thickness distribution field. The actual wall thickness value corresponding to each circumferential angle is extracted from the wall thickness distribution field as the actual wall thickness distribution state and actual wall thickness value of the thin-walled part before cutting.

5. The method of claim 4, wherein, Step 300 includes: Based on the actual wall thickness distribution state and stiffness characteristic surface, determine the wall thickness distribution function and stiffness distribution function of the current thin-walled component at the cut section; Based on the wall thickness distribution function and stiffness distribution function, a dynamic cutting path mapping domain is established with circumferential angle and axial position as independent variables. The dynamic cutting path mapping domain is used to describe the correspondence between the cutting force on the cutting tool at each position and the workpiece stiffness.

6. The method of claim 5, wherein, Step 300 further includes: Within the dynamic cutting path mapping domain, the section to be cut is discretized into multiple cutting units along the circumferential direction. For each cutting unit, the instantaneous cutting force direction at that cutting unit is determined based on the actual wall thickness value and the curvature value of the stiffness characteristic surface. Based on the instantaneous cutting force direction, the cutting angle of the cutting tool at the cutting unit that matches the normal of the stiffness characteristic surface is calculated in reverse. Based on the entry angle and actual wall thickness of each cutting unit, and combined with the cutting parameter model, the feed rate corresponding to each cutting unit is calculated to control the peak cutting force to not exceed the limit allowed by the workpiece stiffness at that cutting unit. Arrange the entry angle and feed rate of each cutting unit in circumferential order to obtain a continuous cut-off tool feed path consisting of multiple discretized segments, where each discretized segment has an independent entry angle and feed rate. The cutting operation is performed according to the continuous cutting tool feed path, so that the cutting force is transmitted sequentially to the anti-deformation support belt and the blank body along the direction with higher stiffness corresponding to each segment, thereby maintaining the geometric accuracy of the finished contour during the cutting process to complete the cutting.

7. The method of claim 6, wherein, Step 400 includes: After the current thin-walled part is cut, the remaining anti-deformation support band on the inner wall of the thin-walled part is removed by turning process to obtain the inner surface of the thin-walled part after releasing the residual stress. During the process of cutting off the anti-deformation support strip, radial runout data of multiple measurement points along the circumferential and axial directions are collected by an online probe to obtain a radial runout dataset during the cutting process; Surface fitting is performed on the radial runout dataset during the resection process to obtain the released state profile surface that reflects the actual profile after the residual stress is released; The released state profile surface is compared point by point with the stiffness feature surface, the normal deviation value at each corresponding position is calculated, and the deviation distribution dataset is formed by all the normal deviation values. Statistical analysis was performed on the deviation distribution dataset to extract the main deviation components that reflect the residual stress release effect, which were then used as profile correction values.

8. The method of claim 7, wherein, Step 400 further includes: The contour correction value is stored in the compensation parameter set of the CNC device and used as a benchmark calibration value for radial runout data acquisition and stiffness characteristic surface fitting before the next thin-walled part is processed. When machining the next thin-walled part, the contour correction value is called to correct the radial runout data, and the corrected radial runout data is obtained. Based on the corrected radial runout data, harmonic decomposition is performed sequentially to extract the main harmonic components. The corresponding harmonic characteristic coefficients are determined according to the main harmonic components. A stiffness characteristic field is constructed based on the harmonic characteristic coefficients, and interpolation is performed along the axial direction to obtain a new stiffness characteristic surface, thereby achieving consistent accuracy control for continuous sequential processing.

9. A deformation prevention cutting-off system for the sequential machining of thin-walled parts of an aeroengine, which implements the method according to any one of claims 1 to 8, characterized in that, include: The stiffness feature module is used to divide the blank into multiple thin-walled parts processing units, and to pre-set an annular material retention area as an anti-deformation support strip at the inner wall connection between each thin-walled part and its adjacent next thin-walled part. Based on the anti-deformation support strip, the current thin-walled part is rough-machined to obtain a rough-machined surface; Based on the rough-machined surface, radial runout data is collected, and stiffness characteristic surfaces are obtained by fitting. The finishing module is used to determine the dynamic offset compensation amount based on the stiffness characteristic surface, and to finish the outer circle and inner hole of the current thin-walled part to obtain a finished surface that retains the anti-deformation support band; based on the finished surface, the actual wall thickness distribution state and the actual wall thickness value are determined. The planning and execution module is used to construct a dynamic cutting path mapping domain based on the actual wall thickness distribution and stiffness characteristic surfaces; Within the dynamic cutting path mapping domain, based on the actual wall thickness and the curvature variation of the stiffness characteristic surface, the feed path of the cutting tool is decomposed into multiple discretized segments, and a corresponding entry angle and feed rate are set for each segment; the cutting is performed according to the entry angle and feed rate to maintain the finished contour without deformation and complete the cutting. The feedback module is used to remove the residual anti-deformation support band on the inner wall of the current thin-walled part after the cutting is completed; during the cutting process, radial runout data is collected to obtain the released profile surface, and the profile correction value is obtained by comparing it with the stiffness feature surface; the profile correction value is used for radial runout data collection and stiffness feature surface fitting before the next thin-walled part is processed.

10. A computing device, comprising: include: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the method as described in any one of claims 1 to 8.

Images