A high-temperature alloy low-vortex disc U-shaped inner cavity ultrasonic shot peening control method

Abstract

This application discloses an ultrasonic shot peening control method for a U-shaped inner cavity of a high-temperature alloy low-vortex disk, relating to the field of ultrasonic shot peening. The method includes: dividing the inner cavity into three process areas based on its geometry: a straight groove section at the inlet, a rounded corner section at the bottom, and a rotary exit section; precisely positioning the workpiece on a high-precision rotary table of a multi-axis CNC machine tool using a dedicated fixture, and establishing a reliable workpiece coordinate system. Based on this, a differentiated CNC path is generated using the workpiece CAD model to drive the relative motion of the ultrasonic vibrating head, ensuring that different areas enter the effective impact range with trajectories matching their curvature characteristics. During shot peening, acoustic and vibration signals are acquired in real time and energy spectrum analysis is performed to dynamically determine coverage adequacy and energy distribution. The dwell time and dwell position are adaptively adjusted based on the analysis results and updated collaboratively with the CNC path, thereby achieving precise, uniform, and continuous strengthening of the complex inner cavity.

Description

Technical Field

[0001] This specification relates to the field of ultrasonic shot peening, and more specifically, this application relates to an ultrasonic shot peening control method for a U-shaped inner cavity of a high-temperature alloy low-vortex disk. Background Technology

[0002] High-temperature alloy low-pressure turbine disks, as crucial load-bearing structures in aero-engine turbine components, typically feature complex U-shaped internal cavities designed to balance structural lightweighting, flow channel stability, and thermal stress distribution control. Because these structures often operate in extreme environments such as high temperature, high pressure, vibration, and cyclic thermal shock, fatigue failure and microcrack initiation on the internal cavity surface are prone to occur in areas of abrupt morphological changes or curvature concentration. Therefore, surface strengthening techniques such as ultrasonic shot peening are urgently needed to improve their fatigue life and crack resistance.

[0003] However, traditional ultrasonic shot peening processes generally rely on the relative motion between a fixed shot peening head and a fixed trajectory. Their path planning is typically a regular straight-line scan or a simple fan-shaped reciprocating motion, which is difficult to effectively adapt to the continuously changing three-dimensional curvature within the U-shaped cavity of a low-volume vortex. Furthermore, traditional shot peening often uses pre-set fixed dwell times and trajectory parameters, lacking a process feedback mechanism based on real-time monitoring signals. This leads to problems such as strengthening blind zones, uneven energy distribution, or localized over-peening in complex curved cavities, particularly prominent in high-stress areas such as the bottom rounded corners and rotational transition sections. Traditional shot peening equipment typically does not perform spatial attitude calculations on the workpiece, and the machine tool's A / C axis linkage capability is not fully utilized. In strengthening cavities with multiple curvatures and transition sections, it is difficult to maintain the shot peening direction consistently aligned with the workpiece normal, affecting the depth and uniformity of the residual compressive stress layer.

[0004] Therefore, it is necessary to propose an ultrasonic shot peening control method for the U-shaped inner cavity of a high-temperature alloy low-vortex disk to at least solve some of the above-mentioned problems. Summary of the Invention

[0005] The summary section introduces a series of simplified concepts, which will be further explained in detail in the detailed description section. This summary section is not intended to limit the key and essential technical features of the claimed technical solution, nor is it intended to determine the scope of protection of the claimed technical solution.

[0006] This invention proposes a method for controlling ultrasonic shot peening of a U-shaped inner cavity of a high-temperature alloy low-vortex disk, comprising: The U-shaped inner cavity of the target low-volume disk workpiece is divided into three process areas according to its geometric shape: the straight groove section at the opening, the rounded corner section at the bottom, and the rotary exit section. The aforementioned low-volume vortex workpiece is fixed on the high-precision rotary table of a multi-axis CNC machine tool using a special fixture, and a precise workpiece coordinate system is established. After establishing the workpiece coordinate system, a CNC path is generated based on the CAD model of the target low-volume disk workpiece to drive the workpiece to move relative to the fixed ultrasonic vibrating head, so that different process areas can enter the effective impact range in a trajectory that matches their shape. Acoustic and vibration signals are collected in real time during ultrasonic shot peening, and the shot peening coverage effect and energy distribution are dynamically analyzed based on the acoustic and vibration signals. Based on the shot peening coverage effect and energy distribution, the dwell time is automatically adjusted so that the dwell time and dwell position of each area can be adaptively adjusted and coordinated with the CNC path control.

[0007] In one feasible implementation, the U-shaped inner cavity of the target low-volume disk workpiece is divided into three process areas according to its geometric shape: a straight groove section at the opening, a rounded corner section at the bottom, and a rotary exit section, including: Based on the curvature distribution, wall thickness variation and geometric transition characteristics of the U-shaped inner cavity, geometric feature recognition is performed on the inner cavity surface to obtain the geometric feature recognition results; Based on the above geometric feature recognition results, the curvature gradient abrupt change location and the continuity boundary of the geometric shape are determined; wherein, the linear groove surface with curvature less than a preset threshold and wall thickness change less than a preset difference is divided into the above groove straight groove segment, the curved surface with continuous curvature change and smooth transition trend is divided into the above bottom rounded corner segment, and the shape region with steep curvature change and tending towards the exit direction is divided into the above rotary exit segment.

[0008] In one feasible implementation, the above-mentioned CNC path for driving the workpiece to move relative to the fixed ultrasonic vibrating head is generated based on the CAD model corresponding to the target low-volume disk workpiece, so that different process areas can enter the effective impact range in a trajectory that matches their shape, including: Based on the CAD model corresponding to the target low-volume disk workpiece, the curvature, normal vector and reachability features of the straight groove section, the bottom rounded corner section and the rotary exit section are extracted respectively to obtain the geometric constraint information of each process area. A linear CNC path is planned for the above-mentioned straight groove section, which involves uniform rotation of the workpiece and coordinated constant axial feed, so that the above-mentioned straight groove section enters the effective impact range of the above-mentioned fixed ultrasonic vibrating head at a constant speed. Based on the continuous curvature variation characteristics of the bottom rounded corner segment, an A / C axis linked spatial interpolation path is generated, and a programmable dwell point is set in the spatial interpolation path to achieve deep reinforcement of the bottom rounded corner segment while maintaining normal alignment. A rotation compensation path is planned for the aforementioned rotary exit section in conjunction with its curvature change direction, and a regional transition section is set in the aforementioned rotation compensation path so that the aforementioned rotary exit section remains consistent with the impact direction of the aforementioned fixed ultrasonic vibrating head when transitioning to the exit direction.

[0009] In one feasible implementation, the above-mentioned real-time acquisition of acoustic and vibration signals during ultrasonic shot peening, and dynamic analysis of the shot peening coverage effect and energy distribution based on the acoustic and vibration signals, includes: After ultrasonic shot peening begins, the acoustic and vibration waveforms from the shot peening area are acquired in real time, and bandpass filtering and normalization are performed on the acoustic and vibration waveforms to obtain acoustic and vibration feature vectors characterizing local impact characteristics. Based on the energy density, dominant frequency distribution, and envelope morphology of the aforementioned acoustic eigenvectors, an energy spectrum analysis of the impact process is performed, and the shot peening energy distribution index is calculated from the analysis results. The shot peening energy distribution index and the historical feature data of the corresponding positions in the CNC path are compared to determine the shot peening coverage effect of the current shot peening area.

[0010] In one feasible implementation, the above-mentioned automatic adjustment of the dwell time based on the shot peening coverage effect and the energy distribution state, enabling adaptive adjustment of the dwell time and dwell position in each area and coordinated control with the CNC path, includes: Based on the curvature characteristics of different process regions and historical shot peening parameters, set adaptive coverage thresholds and energy uniformity thresholds for the region; The dwell deviation of the target shot peening position is calculated by combining the energy offset and coverage offset of the current shot peening point, so as to form the area dwell correction factor; The residence adjustment vector is calculated based on the residence correction factor of the above region, wherein the residence adjustment vector includes residence time increment, residence time decrement and residence point fine-tuning displacement; The above-mentioned dwell adjustment vector is synchronously mapped to the corresponding position of the above-mentioned CNC path to realize the coordinated update of dwell strategy and machine tool multi-axis trajectory; The updated residence time is written into the real-time shot peening control command to drive the shot peening system to perform delayed residence, early detachment, or local micro-relocation actions.

[0011] In one feasible implementation, the above-mentioned setting of the adaptive coverage threshold and energy uniformity threshold based on the curvature characteristics of different process regions and historical shot peening parameters includes: Based on the curvature characteristics of different process regions, the curvature value, curvature change rate and normal reach angle of each process region are calculated respectively. Based on the above curvature value, curvature change rate and normal reach angle, a geometric difficulty coefficient reflecting the shot peening difficulty of the region is constructed to form a set of regional geometric feature parameters. Based on the above historical shot peening parameters, energy spectrum distribution, coverage efficiency, dwell response curve and surface roughness index after shot peening, a regional shot peening response model is established through a regional data weighting mechanism, and corresponding shot peening sensitivity weights are generated for different regions. The above-mentioned set of regional geometric feature parameters, the above-mentioned shot peening sensitivity weights, and the above-mentioned regional shot peening response model are jointly solved to generate a multi-factor threshold calculation formula that includes regional geometric difficulty, shot peening sensitivity, and historical convergence trend. The adaptive coverage threshold and adaptive energy uniformity threshold of each region are then calculated.

[0012] In one feasible implementation, the above-mentioned calculation of the dwell deviation of the target shot peening position based on the energy offset and coverage offset of the current shot peening point includes: Based on the above adaptive coverage threshold and the above adaptive energy uniformity threshold, a dual threshold deviation function is constructed, and the deviation function is used to perform weighted amplification or suppression processing on different deviation sources to form a regional weighted deviation amount. By coupling the weighted deviation of the above-mentioned regions with the shot peening sensitivity weight of the above-mentioned regions, the deviation of the highly sensitive regions is given a higher adjustment priority, thereby forming a residence deviation that is more adapted to the physical characteristics of the regions.

[0013] In one feasible implementation, the calculation of the residence adjustment vector based on the aforementioned regional residence correction factor includes: A multivariate nonlinear mapping model based on the regional geometric difficulty coefficient, the aforementioned shot peening sensitivity weight, and real-time acoustic and vibration characteristics is constructed, and the aforementioned regional dwelling correction factor is projected onto the three-dimensional dwelling control space. In the aforementioned three-dimensional dwell control space, the dwell time increment, dwell time decrement, and dwell point fine-tuning displacement are solved by using a multivariable coupling function, so that the adjustment amounts of different physical dimensions can be automatically balanced according to the regional characteristics. The solution weights are dynamically adjusted based on the changing trends of each component in the three-dimensional dwell control space, thereby forming a dwell adjustment vector that can adapt to the processing difficulty of different regions.

[0014] In one feasible implementation, the above-mentioned synchronous mapping of the dwell adjustment vector with the corresponding position of the CNC path to achieve coordinated updating of the dwell strategy and the multi-axis trajectory of the machine tool includes: After obtaining the above-mentioned dwell adjustment vector, a multi-axis attitude compensation matrix is ​​constructed based on the machine tool A / C axis attitude, the spatial normal calculation results of the above-mentioned CNC path, and the real-time shot peening position. The time component in the dwell adjustment vector is coupled with the spatial attitude component in the compensation matrix so that the dwell adjustment can automatically compensate for attitude deviation while extending or shortening the dwell time, and maintain the normal alignment relationship between the fixed ultrasonic vibrating head and the workpiece surface. Based on the aforementioned multi-axis attitude compensation matrix, the spatial interpolation path is dynamically corrected, ensuring that the shot peening trajectory maintains continuity and coverage without blind spots after dwell adjustment.

[0015] In summary, this invention divides the U-shaped inner cavity into a straight groove section at the inlet, a rounded corner section at the bottom, and a rotary exit section. By establishing a precise coordinate system using a multi-axis CNC machine tool, it achieves differentiated shot peening path design for different geometric features of various regions. Based on curvature and normal analysis of the CAD model, a spatial trajectory adapted to complex surfaces is generated, effectively avoiding shot peening dead zones and improving energy distribution uniformity. Simultaneously, through real-time acquisition and energy spectrum analysis of acoustic and vibration signals, an online coverage and strengthening evaluation mechanism is constructed, dynamically identifying issues such as missed peening, over-peening, and uneven energy distribution, and adaptively adjusting the dwell time and position. Through attitude compensation and multi-axis trajectory collaborative control, the impact direction is always consistent with the workpiece normal, ensuring a continuous and stable shot peening trajectory. This method achieves uniform strengthening and continuous residual stress distribution in complex inner cavities under single-clamping conditions, significantly improving component fatigue performance and machining quality.

[0016] The ultrasonic shot peening control method for the U-shaped inner cavity of the high-temperature alloy low-vortex disk proposed in this application, other advantages, objectives and features of this application will be partly apparent from the following description, and partly understood by those skilled in the art through study and practice of this application. Attached Figure Description

[0017] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit this specification. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 A schematic flowchart of an ultrasonic shot peening control method for a high-temperature alloy low-vortex disk U-shaped inner cavity is provided for an embodiment of this application. Figure 2 This is a structural schematic diagram of a high-temperature alloy low-vortex disk U-shaped inner cavity part provided for an embodiment of this application. Detailed Implementation

[0018] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus. The technical solutions of the embodiments of this application will now be clearly and completely described in conjunction with the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them.

[0019] Ultrasonic shot peening is a surface strengthening method that utilizes ultrasonic vibration energy to drive projectiles to impact the surface of a workpiece at high frequencies. Its basic principle is to convert electrical energy into high-frequency mechanical vibration through an ultrasonic transducer, which is then transmitted to the shot peening head or cavity via an amplitude amplifier. This causes the projectile to acquire periodic kinetic energy under high-frequency action, continuously impacting the metal surface at an extremely high impact frequency. Unlike traditional shot peening, ultrasonic shot peening uses projectiles with lower velocities but extremely high impact frequencies, thereby forming a dense and uniform plastic deformation zone and a stable residual compressive stress field on the workpiece surface. This stress field effectively inhibits crack initiation and propagation, significantly improving the fatigue strength, stress corrosion resistance, and surface hardness of the parts.

[0020] During processing, ultrasonic shot peening eliminates the need for high-pressure airflow or centrifugal mechanisms, relying solely on the kinetic energy generated by ultrasonic vibration to accelerate and impact the shot. This results in higher energy efficiency and lower noise and pollution. Its concentrated and controllable peening energy allows for precise strengthening of complex curved surfaces, deep cavities, thin walls, and edge areas, making it particularly suitable for parts with complex internal structures such as drums, impellers, and mold cavities. Compared to traditional shot peening, ultrasonic shot peening can introduce a deeper compressive stress layer while maintaining surface quality. Furthermore, it allows for intelligent and repeatable processing control through real-time adjustment of amplitude and frequency, thus balancing surface strengthening effects with workpiece dimensional accuracy. It is an efficient, clean, and advanced surface strengthening technology suitable for precision structural parts.

[0021] Please see Figure 1 and Figure 2 , Figure 1 This is a flowchart illustrating an ultrasonic shot peening control method for a U-shaped inner cavity of a high-temperature alloy low-vortex disk, as provided in an embodiment of this application. Figure 2 A structural schematic diagram of a high-temperature alloy low-vortex disk U-shaped inner cavity component provided in this application embodiment may specifically include: S110. Divide the U-shaped inner cavity of the target low-volume disk workpiece into three process areas according to its geometric shape: the straight groove section at the opening, the rounded corner section at the bottom, and the rotary exit section. S120. Fix the above-mentioned low-volume chute workpiece onto the high-precision rotary table of a multi-axis CNC machine tool using a special fixture, and establish a precise workpiece coordinate system. S130. After the above workpiece coordinate system is established, a CNC path is generated based on the CAD model corresponding to the target low-volume disk workpiece to drive the workpiece to move relative to the fixed ultrasonic vibration head, so that different process areas can enter the effective impact range in a trajectory that matches their shape. S140. Acoustic and vibration signals are collected in real time during ultrasonic shot peening, and the shot peening coverage effect and energy distribution are dynamically analyzed based on the acoustic and vibration signals. S150. Based on the shot peening coverage effect and energy distribution, the dwell time is automatically adjusted so that the dwell time and dwell position of each area can be adaptively adjusted and coordinated with the CNC path control.

[0022] For example, to address the strengthening requirements of the complex U-shaped inner cavity of a high-temperature alloy low-volume disk, the entire ultrasonic shot peening process unfolds sequentially according to clearly defined technological steps. First, the U-shaped inner cavity of the target low-volume disk workpiece is finely divided according to its geometric morphology, and the overall inner cavity is divided into three process areas: the straight groove section 101, the rounded corner section 102, and the rotary exit section 103. This distinguishes the simple morphology and gentle curvature of the straight groove area from the continuously changing curvature and stress-sensitive rounded corner area, as well as the geometrically abrupt rotary exit section that flows towards the outlet, at the process planning level, laying the foundation for subsequent differentiated path planning and parameter control.

[0023] After the area division is completed, the aforementioned low-volume disk workpiece is reliably fixed on the high-precision turntable of a multi-axis CNC machine tool using a special fixture. Through the alignment process between the machine tool coordinate system and the workpiece reference surface, a precise workpiece coordinate system is established, so that the workpiece's posture and position in space can be accurately perceived and called by the CNC system. This ensures that the subsequently generated spatial trajectory corresponds one-to-one with the actual internal cavity position, avoiding shot peening blind spots or uneven strengthening caused by positioning errors.

[0024] After the workpiece coordinate system is established, the three-dimensional geometry of the U-shaped inner cavity is analyzed based on the CAD model of the target low-volume disk workpiece, generating a CNC path to drive the workpiece relative to the fixed ultrasonic vibrating head. By analyzing the curvature, normal direction, and accessibility of different process areas, the straight groove section preferentially adopts a trajectory of uniform rotation combined with axial feed, the bottom rounded corner section adopts a multi-axis linkage spatial interpolation trajectory, and the rotary exit section adopts a rotary path with transition compensation. This ensures that each process area can enter the effective impact range of the fixed ultrasonic vibrating head in a way that matches its own shape, achieving full coverage shot peening of the complex curved surface of the inner cavity.

[0025] In actual ultrasonic shot peening, acoustic and vibration signals from the shot-peening area are acquired in real time by arranged acoustic and vibration sensing units. These signals undergo preprocessing operations such as bandpass filtering and normalization to extract representative energy density, dominant frequency components, and envelope variation characteristics. Based on these acoustic and vibration characteristics, dynamic analysis of the current impact process is performed. This analysis serves two purposes: firstly, to determine whether the shot peening coverage is sufficient and whether there is any missed or over-peening; and secondly, to assess whether the shot peening energy is uniformly distributed across different areas and time periods, thereby obtaining quantitative results regarding the shot peening coverage effect and energy distribution.

[0026] After obtaining the shot peening coverage effect and energy distribution, these analysis results are further compared with preset thresholds, historical process data, and the geometric characteristics of the current process area. The difference between the current dwell time and the target ideal dwell time for each area is automatically calculated. Based on this, by adjusting the dwell time of each dwell point on the shot peening trajectory, appropriately increasing or decreasing the local dwell time, and even making minor corrections to the dwell position, longer and denser dwell times are achieved in high-difficulty areas, areas with abrupt curvature changes, or areas with insufficient coverage, while the dwell time is appropriately shortened in areas with excessive coverage or excessive energy. Through this automatic adjustment strategy based on the shot peening coverage effect and energy distribution, adaptive adjustment of the dwell time and dwell position of each process area is achieved, while ensuring that the above adjustment process is coordinated with the original CNC path and does not disrupt the established spatial trajectory continuity and normal impact posture. Thus, refined and intelligent ultrasonic shot peening strengthening of the U-shaped inner cavity of the high-temperature alloy low-vortex disk is completed under single clamping conditions.

[0027] In summary, this invention precisely divides the U-shaped inner cavity into three process areas: a straight groove section, a rounded bottom section, and a rotary exit section. By establishing a precise workpiece coordinate system using a multi-axis CNC machine tool, it can formulate differentiated shot peening path strategies based on the geometric characteristics of different areas, fundamentally improving the shortcomings of traditional shot peening trajectories that cannot adapt to complex curved surfaces. Through three-dimensional curvature, normal, and reachability analysis based on the CAD model, this invention can generate CNC spatial trajectories that precisely match the shape of each process area. For example, a regular path with uniform rotation and axial feed is used in the straight groove area; an A / C axis linked spatial interpolation trajectory is used in the rounded bottom section; and a transition path with directional compensation is used in the rotary exit section. This effectively avoids shot peening dead zones and ensures the uniform application of strengthening energy on complex curved surfaces. By acquiring acoustic and vibration signals in real time and performing dynamic energy spectrum analysis, this invention establishes an online coverage and energy distribution assessment mechanism based on acoustic and vibration feedback. Compared with the traditional method that relies on experience to set parameters, this method can determine in real time whether there is missed peening, over-peening, or uneven energy in the shot peening coverage area, and automatically adjust the dwell time and dwell position accordingly. By constructing an adaptive dwell adjustment algorithm, the dwell time can be extended for areas with abrupt curvature changes, deep rounded corners, or areas with insufficient historical strengthening, while the dwell time can be appropriately shortened for areas with excessive strengthening. It can even perform micro-spatial repositioning of the dwell point, making the strengthening behavior more intelligent and precise. This invention synchronously maps the dwell adjustment command with the multi-axis trajectory of the machine tool, and corrects the A / C axis attitude in real time through the attitude compensation matrix, so that any dwell point can maintain the impact direction consistent with the workpiece normal, significantly improving the continuity and depth of the local residual stress layer. This integrated mapping method ensures that the shot peening trajectory remains smooth and continuous when adjusting the time or fine-tuning the position, without destroying the accessibility and stability of the original spatial motion, and realizing complete coordination between dwell control and CNC path. In summary, this invention not only solves the problem of insufficient strengthening ability of traditional ultrasonic shot peening for complex internal curved surfaces, but also realizes real-time adaptive strengthening based on acoustic vibration feedback. This enables high-temperature alloy low-vortex disk U-shaped internal cavity to obtain high-quality shot peening effect with uniform coverage, sufficient strengthening and continuous residual stress field under single clamping conditions, which significantly improves the fatigue life and service reliability of components and has outstanding engineering application value and creativity.

[0028] In one feasible implementation, the U-shaped inner cavity of the target low-volume disk workpiece is divided into three process areas according to its geometric shape: a straight groove section at the opening, a rounded corner section at the bottom, and a rotary exit section, including: Based on the curvature distribution, wall thickness variation and geometric transition characteristics of the U-shaped inner cavity, geometric feature recognition is performed on the inner cavity surface to obtain the geometric feature recognition results; Based on the above geometric feature recognition results, the curvature gradient abrupt change location and the continuity boundary of the geometric shape are determined; wherein, the linear groove surface with curvature less than a preset threshold and wall thickness change less than a preset difference is divided into the above groove straight groove segment, the curved surface with continuous curvature change and smooth transition trend is divided into the above bottom rounded corner segment, and the shape region with steep curvature change and tending towards the exit direction is divided into the above rotary exit segment.

[0029] For example, regarding the complex U-shaped internal cavity structure of a high-temperature alloy low-volume disk, the initial approach is not to simply rely on manual experience to visually distinguish straight grooves, rounded corners, and rotating sections. Instead, an automated identification process based on geometric parameters is introduced. Specifically, modeling software or a geometric analysis module is used to analyze the three-dimensional model of the U-shaped internal cavity, extracting information on the curvature distribution of the internal cavity surface in the axial and circumferential directions, the variation of wall thickness along the internal cavity path, and the morphological features of the geometric transitions between different regions. This forms a geometric feature dataset for subsequent judgment.

[0030] By analyzing the curvature variation with position, regions with curvature approaching zero and gradual change can be identified, as well as transition regions with continuous curvature changes but a relatively smooth trend. Regions where curvature exhibits significant abrupt changes near certain locations can also be identified. Simultaneously, by combining this with wall thickness variation information, it can be determined which regions are structurally regular with relatively constant wall thickness, and which regions experience changes in wall thickness due to structural transitions or flow channel bends. Furthermore, by incorporating geometric transition features, such as the transition of cross-sectional shape from a straight line to a circular arc, or from a circular arc towards the outlet direction, a comprehensive geometric feature identification result can be obtained.

[0031] After obtaining the geometric feature recognition results, the inner cavity is segmented based on the changes in curvature gradient and the continuity of the morphology. First, by detecting abrupt changes in curvature gradient, boundary points where significant changes in geometric characteristics occur can be found. These boundary points often correspond to the natural boundaries between straight grooves and fillets, and between fillets and the rotary exit section. Simultaneously, by judging the continuity of the geometric morphology, incorrect segmentation caused by minor local perturbations or modeling noise can be avoided.

[0032] In terms of specific classification criteria, when the curvature of a certain region within the cavity is less than a preset curvature threshold and the wall thickness variation is less than a preset wall thickness difference, the geometry of this region can be considered to be basically a linear groove surface with a regular and stable structure. Therefore, this region is classified as the aforementioned straight groove section. This region typically corresponds to the area near the inlet or a straight section structure, suitable for a relatively simple uniform rotation and axial feed shot peening strategy. For curved surface regions where the curvature is not close to zero but changes continuously along the cavity path with a smooth trend and no obvious sharp bends, they are identified as geometrically transitional arcs or composite arc regions. Structurally, these regions often correspond to the bottom fillet or the bending transition section of the cavity, thus they are classified as the aforementioned bottom fillet section. This region is usually a sensitive area of ​​stress concentration, requiring more refined normal control and dwell strategies. As for the areas where the curvature changes sharply and the geometric direction deviates significantly towards the outlet direction, it indicates that the structure gradually transitions from the bottom rounded corners to the turning outlet section, and the surface direction and flow characteristics change significantly. Therefore, these morphological areas are divided into the aforementioned turning outlet section for subsequent spraying and homogenization treatment of the outlet transition area.

[0033] By employing the aforementioned partitioning method based on curvature distribution, wall thickness variation, and geometric transition characteristics, this embodiment achieves automated process region division with engineering significance for the U-shaped inner cavity. Furthermore, the partitioning criteria for each region are supported by clear geometric judgments, independent of subjective experience. This provides a reliable geometric basis for planning ultrasonic shot peening trajectories that match the morphological characteristics of each subsequent process region. It also creates conditions for employing differentiated residence strategies, energy distribution control, and strengthening depth control in different regions, thereby enabling a more uniform, controllable, and repeatable surface strengthening effect for the low-vortex U-shaped inner cavity as a whole.

[0034] In one feasible implementation, the above-mentioned CNC path for driving the workpiece to move relative to the fixed ultrasonic vibrating head is generated based on the CAD model corresponding to the target low-volume disk workpiece, so that different process areas can enter the effective impact range in a trajectory that matches their shape, including: Based on the CAD model corresponding to the target low-volume disk workpiece, the curvature, normal vector and reachability features of the straight groove section, the bottom rounded corner section and the rotary exit section are extracted respectively to obtain the geometric constraint information of each process area. A linear CNC path is planned for the above-mentioned straight groove section, which involves uniform rotation of the workpiece and coordinated constant axial feed, so that the above-mentioned straight groove section enters the effective impact range of the above-mentioned fixed ultrasonic vibrating head at a constant speed. Based on the continuous curvature variation characteristics of the bottom rounded corner segment, an A / C axis linked spatial interpolation path is generated, and a programmable dwell point is set in the spatial interpolation path to achieve deep reinforcement of the bottom rounded corner segment while maintaining normal alignment. A rotation compensation path is planned for the aforementioned rotary exit section in conjunction with its curvature change direction, and a regional transition section is set in the aforementioned rotation compensation path so that the aforementioned rotary exit section remains consistent with the impact direction of the aforementioned fixed ultrasonic vibrating head when transitioning to the exit direction.

[0035] For example, for the U-shaped inner cavity of a high-temperature alloy low-volume disk, which has a complex geometry, drastic local curvature changes, and limited machining space, this method does not directly follow the trajectory on the machine tool based on experience. Instead, it first relies on the CAD three-dimensional model corresponding to the target low-volume disk workpiece to analyze the geometry of different regions of the inner cavity, thereby generating a CNC motion path that closely matches the process area.

[0036] Specifically, based on the aforementioned CAD model, curvature, normal vector, and accessibility features are extracted from the previously defined straight groove section, bottom rounded corner section, and rotary exit section. Curvature characterizes the degree of bending and curvature variation trend of the internal cavity surface; the normal vector determines the normal impact direction at each point; and accessibility features reflect whether the ultrasonic vibrating head can approach the point and its adjacent area without interference under the multi-axis posture of the machine tool. Through these feature extraction operations, a set of geometric constraint information, including curvature constraints, normal constraints, and machining accessibility constraints, can be established for each process area, providing precise boundary conditions and guiding rules for subsequent CNC path planning.

[0037] After obtaining the aforementioned geometric constraint information, for straight groove sections with relatively regular shapes, small curvature, and nearly straight walls, this embodiment prioritizes a relatively simple yet highly efficient trajectory method. Specifically, it plans a motion that causes the workpiece to rotate uniformly around the turntable axis, while simultaneously applying a constant-speed axial feed, thereby forming a linear CNC path. Since both the turntable rotation speed and the axial feed speed remain constant, each section of the groove wall in this region can pass through the effective impact range of the fixed ultrasonic vibrator head at a stable and controllable linear velocity, thus ensuring the uniformity of shot peening coverage and the stability of impact energy. This path planning method is not only computationally simple but also beneficial for improving processing efficiency and path reusability, making it very suitable for areas with simple geometric features, such as straight groove sections.

[0038] For the bottom rounded corner section located at the bottom of the groove, which is in the shape of an arc or a composite arc, its curvature changes continuously and it is often a sensitive area for stress concentration and fatigue crack initiation. Therefore, this embodiment employs a more refined multi-axis linkage control strategy. Based on the curvature and normal features extracted from the CAD model, a spatial interpolation path with A / C axis linkage is generated. This ensures that, under the synergistic effect of turntable rotation (C-axis) and oscillation tilt (A-axis), each target point on the bottom rounded corner surface of the workpiece can be aligned with the impact direction of the fixed ultrasonic vibrator head in a near-ideal normal orientation during movement. To achieve deeper residual compressive stress and higher strengthening uniformity in these critical areas, several programmable dwell points are also set on the aforementioned spatial interpolation path. When the workpiece moves to the vicinity of these dwell points, the feed rate is actively reduced or paused for a period of time, allowing the ultrasonic vibrator head to maintain repeated impacts in the local area for a longer period, thereby achieving deep strengthening of the bottom rounded corner section while maintaining normal alignment.

[0039] For the rotary exit section where the geometry changes abruptly from the bottom of the tank to the outlet, and the curvature changes abruptly, this embodiment further considers the risks of drastic changes in morphology, discontinuous coverage, and uneven energy distribution. Therefore, a rotary compensation mechanism is introduced in the path planning. Specifically, a rotary compensation path with attitude compensation is planned by combining the curvature change direction and local normal change law of the rotary exit section. This allows the workpiece's attitude to be adjusted synchronously during the transition from the rounded corner to the outlet, thus counteracting the normal offset trend caused by the geometric change. At the same time, a transition section is set in the rotary compensation path to make the motion trajectory of the transition area smoother in space, thereby avoiding sharp fluctuations in shot peening energy or local impact blind spots caused by abrupt attitude changes. Through such path and attitude coordinated control, the rotary exit section can maintain a general consistency with the impact direction of the fixed ultrasonic vibrator head while gradually turning the overall flow direction towards the outlet, ensuring the strengthening continuity and smooth transition of the residual stress field in the transition area near the inner cavity outlet.

[0040] This embodiment extracts curvature, normal, and accessibility features of different process areas based on a CAD model, and tailors CNC path planning strategies for the straight groove section, the bottom rounded corner section, and the rotary exit section. The straight groove section adopts a linear path with uniform rotation and constant feed; the rounded corner section uses A / C axis linked spatial interpolation and superimposed programmed dwell; and the rotary exit section adopts a rotation compensation path combined with curvature abrupt change direction. This multi-region collaborative path planning method allows internal cavity areas with different geometric characteristics to enter the effective impact range of the fixed ultrasonic vibrating head with the trajectory most suitable for their own morphology. This significantly improves the uniformity of shot peening coverage, the consistency of reinforcement depth, and the overall continuity of the residual stress field in complex U-shaped internal cavities while ensuring processing efficiency.

[0041] In one feasible implementation, the above-mentioned real-time acquisition of acoustic and vibration signals during ultrasonic shot peening, and dynamic analysis of the shot peening coverage effect and energy distribution based on the acoustic and vibration signals, includes: After ultrasonic shot peening begins, the acoustic and vibration waveforms from the shot peening area are acquired in real time, and bandpass filtering and normalization are performed on the acoustic and vibration waveforms to obtain acoustic and vibration feature vectors characterizing local impact characteristics. Based on the energy density, dominant frequency distribution, and envelope morphology of the aforementioned acoustic eigenvectors, an energy spectrum analysis of the impact process is performed, and the shot peening energy distribution index is calculated from the analysis results. The shot peening energy distribution index and the historical feature data of the corresponding positions in the CNC path are compared to determine the shot peening coverage effect of the current shot peening area.

[0042] For example, in order to achieve real-time monitoring and intelligent evaluation of the ultrasonic shot peening process of the U-shaped inner cavity of the low-vortex disk of high-temperature alloy, this method introduces an online detection mechanism with dual acoustic and vibration channels during the processing. By extracting features and analyzing the energy spectrum of the above signals, the dynamic determination of the shot peening coverage effect and energy distribution state is achieved.

[0043] Specifically, after ultrasonic shot peening begins, acoustic and vibration waveforms from the shot peening area are acquired in real time using acoustic and vibration sensors positioned near the peening area. The acoustic waveforms reflect the acoustic emission characteristics generated when the shot peening medium impacts the workpiece surface, while the vibration waveforms reflect the local vibration response of the workpiece under impact load. Since background noise from the machine tool, environmental noise, and non-impact-related interference signals are inevitably superimposed in the environment, this embodiment performs bandpass filtering on the acoustic and vibration waveforms to remove low-frequency and high-frequency interference components unrelated to the ultrasonic shot peening process, retaining only the target frequency band highly correlated with the shot peening impact behavior. Subsequently, normalization processing is used to bring the signal amplitudes from different time periods and different sensor channels to a uniform level, eliminating scale differences caused by sensor sensitivity, amplification, or operating condition fluctuations, to obtain an acoustic-vibration feature vector characterizing the local impact characteristics. This acoustic-vibration feature vector integrates the basic response characteristics of the impact process in the time and frequency domains, serving as the fundamental input for subsequent energy analysis and coverage assessment.

[0044] After obtaining the aforementioned acoustic vibration characteristic vectors, this embodiment further performs energy spectrum analysis on the impact process based on the energy density, dominant frequency distribution, and envelope morphology of the acoustic characteristic vectors. Statistical analysis of the energy density reflects the overall impact intensity level of each impact cycle or each path segment. Analysis of the dominant frequency distribution identifies abnormal impact patterns during the current shot peening process, such as abnormally high single impact energy, abnormal impact rhythm, or unstable particle stream. Examining changes in the envelope morphology reveals the stability and uniformity of the impact process over time, such as whether the envelope is smooth, or whether periodic fluctuations or abrupt changes occur. Based on these multidimensional spectral characteristics, this embodiment constructs a shot peening energy distribution index. By comprehensively quantifying the concentration, volatility, and stability of the energy distribution, this index can intuitively characterize whether the energy distribution within the current shot peening area is uniform and whether there is a risk of insufficient strengthening due to excessively low energy or excessive impact due to excessively high energy.

[0045] After calculating the shot peening energy distribution index, the index is not judged in isolation, but rather compared with historical characteristic data of the corresponding position in the CNC path. Specifically, the CNC system records characteristic data of each spatial position in previous shot peening batches, including the shot peening energy distribution index, acoustic and vibration response level, and corresponding coverage assessment results of that position or area during previous processing. By comparing the current shot peening energy distribution index with historical characteristic data longitudinally, it is possible to determine the deviation of the current shot peening area from the predetermined process target, such as whether the current energy level is significantly lower than the level during previous good processing, or whether the energy fluctuation is significantly higher than the historical stable state; on the other hand, it is possible to identify the relative differences between different areas on the same path, thereby determining whether there is insufficient coverage, uneven coverage, or local reinforcement anomalies in certain areas.

[0046] Finally, this embodiment combines the comparison results of the above-mentioned shot peening energy distribution index with historical characteristic data to form a judgment on the shot peening coverage effect of the current shot peening area. When the judgment result shows that the energy distribution of the current area is basically consistent with the historical ideal state and the fluctuation is within the allowable range, the coverage effect can be considered good. If the energy distribution index is significantly low or the fluctuation is too large, it indicates that there may be insufficient strengthening or shot peening instability in the area, which needs to be corrected in subsequent steps by adjusting the residence time or local re-peening.

[0047] In one feasible implementation, the above-mentioned automatic adjustment of the dwell time based on the shot peening coverage effect and the energy distribution state, enabling adaptive adjustment of the dwell time and dwell position in each area and coordinated control with the CNC path, includes: Based on the curvature characteristics of different process regions and historical shot peening parameters, set adaptive coverage thresholds and energy uniformity thresholds for the region; The dwell deviation of the target shot peening position is calculated by combining the energy offset and coverage offset of the current shot peening point, so as to form the area dwell correction factor; The residence adjustment vector is calculated based on the residence correction factor of the above region, wherein the residence adjustment vector includes residence time increment, residence time decrement and residence point fine-tuning displacement; The above-mentioned dwell adjustment vector is synchronously mapped to the corresponding position of the above-mentioned CNC path to realize the coordinated update of dwell strategy and machine tool multi-axis trajectory; The updated residence time is written into the real-time shot peening control command to drive the shot peening system to perform delayed residence, early detachment, or local micro-relocation actions.

[0048] For example, the process described above, which automatically adjusts the residence time based on shot peening coverage and energy distribution, can be characterized by a set of explicit mathematical relationships. Firstly, in the step of setting adaptive coverage thresholds and energy uniformity thresholds for different process regions based on their curvature characteristics and historical shot peening parameters, each process region can... (For example, the straight groove section at the opening, the rounded corner section at the bottom, and the rotary exit section) Define a geometric difficulty coefficient. This coefficient takes into account the average curvature of the region, the rate of change of curvature, and normal accessibility. For example, it can be expressed in the following form:

[0049] in, Indicates the first The average curvature of the region, It is the maximum value of the average curvature across all regions; Indicates the first The rate of change of the curvature of the region along the path direction. It is the maximum value among the absolute values ​​of the rate of change of curvature in all regions; This represents the average angle between the normal to this region and the ideal strike normal; , which is a weighting coefficient used to balance the contribution of curvature magnitude, curvature variation, and normal reachability to the difficulty of the region. The larger the value, the more complex the geometry of the region, the more difficult it is to align the normals, and the greater the difficulty in strengthening control.

[0050] After obtaining the geometric difficulty coefficient, an adaptive coverage threshold can be set for each region by combining historical shot peening parameters (such as historical energy spectrum, historical coverage, and historical quality score). and adaptive energy uniformity threshold ,For example:

[0051] in, and These are the basic coverage threshold and basic energy uniformity threshold set by the system, respectively. This is an adjustment coefficient. The greater the geometric difficulty, the... The higher the value (the more comprehensive the coverage required), the better. The lower the value (the smaller the allowable energy fluctuation), the stricter the shot peening control requirements for complex areas.

[0052] In the step of jointly calculating the dwell deviation of the target shot peening position based on the energy offset and coverage offset of the current shot peening point to form the area dwell correction factor, the area where the current shot peening point is located can be set as follows: Its current energy index is The target or historical ideal energy index is The current coverage indicator is The target or historical ideal coverage indicator is Then, energy offset and coverage deviation can be defined as normalized bias:

[0053] In the formula, This indicates that the energy level is too high. This indicates that the energy level is low. This indicates excessive coverage. This indicates insufficient coverage. Based on this, the two can be linearly or weightedly fused into a residence bias. :

[0054] in, and , where is the weighting coefficient, used to adjust the relative importance of energy bias and coverage bias in residency regulation. If Overall, the situation leans more towards insufficient energy / coverage, and a longer stay should be preferred; if If the spraying is too dense, it indicates overspraying or excessive coverage, and the dwell time should be reduced.

[0055] After obtaining the residency deviation, the region's geometric difficulty can be combined with this deviation to form a region residency correction factor. :

[0056] in, This is an adjustment coefficient used to amplify the effects of biases in complex regions. When the region's geometric difficulty... When the deviation is large, the same amount of deviation This will be amplified, giving the region a higher priority in subsequent residency adjustments, thus reflecting the strategy of prioritizing the correction of regions that are difficult to process and prone to problems.

[0057] In the step of calculating the residence adjustment vector based on the aforementioned residence correction factor, where the residence adjustment vector includes residence time increment, residence time decrement, and residence point fine-tuning displacement, the residence adjustment vector can be denoted as:

[0058] in, Indicates dwell time, used to extend dwell time when coverage is insufficient or energy is low; This indicates a reduction in dwell time, used to shorten dwell time when energy is too high or coverage is too dense. This represents the fine-tuning displacement of the dwell point along the CNC path or normal direction, used for slightly shifting the dwell position in areas of local geometric complexity. A feasible solution method is:

[0059] in, These are the proportional coefficients for time increment, time decrement, and displacement adjustment, respectively; when... hour, 0 indicates that the length of stay should be increased. When hour, This indicates that the length of stay should be reduced. Then according to Symbols and regional difficulty The direction and magnitude of the fine-tuning displacement are determined in order to achieve more precise spatial enhancement control in geometrically complex regions.

[0060] By synchronously mapping the aforementioned dwell adjustment vector to the corresponding position of the aforementioned CNC path, the dwell switch and the machine tool can be synchronized. In the collaborative update of axis trajectories, it can be understood as... The correction is for the feed rate or local dwell time at that path point. This is mapped to a small displacement correction of the path point on the interpolation trajectory. By embedding these corrections in the CNC interpolation module, a coordinated control effect is achieved that allows the dwell adjustment to not disrupt the continuity of the original trajectory and the normal attitude.

[0061] Finally, in the step of writing the updated residence time into the real-time shot peening control command to drive the shot peening system to perform delayed residence, early disengagement, or local micro-dwelling actions, the control system calculates the residence time according to the above formula. Update the dwell time and dwell position for the current and subsequent shot peening cycles: when When it is the dominant state, a delayed dwell time is executed accordingly. When the dominant force is in a dominant position, the corresponding action should be to detach early; when... When the area is large and the geometry is complex, a local micro-dwelling action is performed. In this embodiment, the automatic adjustment of dwell time based on shot peening coverage and energy distribution is no longer an empirical manual adjustment, but an intelligent dwelling control process that can be quantitatively solved and executed in real time.

[0062] In one feasible implementation, the above-mentioned setting of the adaptive coverage threshold and energy uniformity threshold based on the curvature characteristics of different process regions and historical shot peening parameters includes: Based on the curvature characteristics of different process regions, the curvature value, curvature change rate and normal reach angle of each process region are calculated respectively. Based on the above curvature value, curvature change rate and normal reach angle, a geometric difficulty coefficient reflecting the shot peening difficulty of the region is constructed to form a set of regional geometric feature parameters. Based on the above historical shot peening parameters, energy spectrum distribution, coverage efficiency, dwell response curve and surface roughness index after shot peening, a regional shot peening response model is established through a regional data weighting mechanism, and corresponding shot peening sensitivity weights are generated for different regions. The above-mentioned set of regional geometric feature parameters, the above-mentioned shot peening sensitivity weights, and the above-mentioned regional shot peening response model are jointly solved to generate a multi-factor threshold calculation formula that includes regional geometric difficulty, shot peening sensitivity, and historical convergence trend. The adaptive coverage threshold and adaptive energy uniformity threshold of each region are then calculated.

[0063] For example, for the three process areas—the straight groove section at the inlet, the rounded corner section at the bottom, and the rotary exit section—the curvature characteristics of each area are extracted based on a 3D CAD model. For the first... Each process area (e.g.) Corresponding to the straight groove section, Corresponding to the bottom rounded corner segment, (corresponding to the exit segment of the revolution), its average curvature can be calculated. Rate of curvature change along the path direction And the average angle between the normal vector of this region and the ideal impact normal vector. And calculate the maximum curvature globally. and the maximum rate of change of curvature Based on this, a geometric difficulty coefficient reflecting the geometric difficulty of shot peening in the region is constructed. For example, it can be represented as:

[0064] in, Indicates the first The average curvature of the region, The maximum value among the average curvatures of each region is used for normalization; Indicates the first The rate of change of curvature of the region along the path direction. This represents the maximum absolute value of the rate of change of curvature in each region; This is the average deflection angle between the normal to this region and the ideal impact normal; These are weighting coefficients for three geometric factors, used to balance the influence of curvature magnitude, curvature variation, and normal accessibility on shot peening difficulty. Therefore, when a region has large curvature, drastic curvature variations, or a large normal deflection angle, its geometric difficulty coefficient... The increase will be significant, reflecting the high difficulty in shot peening coverage and attitude control in this area.

[0065] After forming the aforementioned set of regional geometric feature parameters, a regional shot peening response model is further established based on historical shot peening parameters, integrating energy harmonic distribution, coverage efficiency, residence response curves, and post-peening surface roughness indices. Corresponding shot peening sensitivity weights are then generated for different regions. Specifically, statistical analysis can be performed on the energy spectrum characteristics, final coverage, residence-strengthening response relationship, and roughness changes of each process region across multiple historical batches to extract fluctuations related to stability. For example, energy fluctuation can be defined. Coverage efficiency fluctuation or convergence speed index and the normalized index of the roughness change rate after shot peening. And a shot peening sensitivity weight is constructed by weighting. :

[0066] in, It can represent the first The normalized standard deviation of the energy spectrum characteristics in each batch of processing in the region reflects whether the energy distribution is stable; This reflects the speed or fluctuation of the area's coverage efficiency converging from its initial state to the target coverage level. It reflects the fluctuation of crude viscosity index between different batches of processing; These are the weighting coefficients for the three types of indicators, used to express the importance of different quality indicators in sensitivity evaluation. When a certain area experiences large energy fluctuations, uncontrollable coverage, and inconsistent roughness changes during historical processing, its shot peening sensitivity weight is... The corresponding increase indicates that the region is more sensitive to disturbances in process parameters and requires stricter threshold control.

[0067] To reflect the time-dimensional characteristic of historical convergence trends, this embodiment can also define a convergence trend index for each region. For example, the quality score sequence (including residual stress depth, defect rate, or overall quality score) of a certain region in several early batches can be denoted as... The scoring sequence in the most recent batches is denoted as The convergence of processing quality in this region is measured by the decay of variance.

[0068] in, Represents variance operation; when When the value approaches 1, it indicates that the recent quality score of the region is significantly more stable, and the historical convergence trend is good; when When the value is small or even close to 0, it indicates that the region has experienced significant fluctuations during long-term processing, which can easily lead to unstable shot peening results.

[0069] In obtaining the geometric difficulty coefficient of the region Shot peening sensitivity weight Historical convergence trend indicators Subsequently, this embodiment jointly solves the aforementioned regional geometric feature parameter set, the aforementioned shot peening sensitivity weights, and the aforementioned regional shot peening response model to construct a multi-factor threshold calculation formula, which is used to calculate the adaptive coverage threshold and adaptive energy uniformity threshold for each region. For example, the adaptive coverage threshold can be... Defined as:

[0070] in, This is the system's basic coverage threshold. Adjustable coefficient; geometric difficulty and shot peening sensitivity The larger, the better A higher value indicates a higher minimum coverage requirement for the area; while historical convergence trends... The better, the lower the additional coverage requirement can be, so it appears in the formula as a minus sign.

[0071] Accordingly, the adaptive energy uniformity threshold can be set. Designed as follows:

[0072] in, The threshold for the system's basic energy uniformity; This is an adjustable coefficient. The higher the geometric difficulty and shot peening sensitivity, the smaller the allowable energy fluctuation range should be. and The threshold is reduced by a minus sign; however, when a region has historically shown good convergence, the tolerance for energy fluctuations can be appropriately relaxed. right It has a positive corrective effect.

[0073] The multi-factor threshold calculation formula in this embodiment is a specific mathematical expression of the combined effects of three dimensions: regional geometric difficulty, shot peening sensitivity, and historical convergence trend. Based on these calculation results, different adaptive coverage thresholds and energy uniformity thresholds are generated for the straight groove section, the bottom rounded corner section, and the rotary exit section, respectively. This allows for stricter coverage and energy control in regions with high curvature, stress sensitivity, and historical instability, while the control conditions can be appropriately relaxed in regions with simple geometry and stable history. The entire U-shaped inner cavity achieves regional differentiation, parameter adaptation, and time-evolution optimization in shot peening control, which is beneficial for obtaining a more uniform and reliable residual stress field distribution and surface strengthening quality in complex inner cavity structures.

[0074] In one feasible implementation, the above-mentioned calculation of the dwell deviation of the target shot peening position based on the energy offset and coverage offset of the current shot peening point includes: Based on the above adaptive coverage threshold and the above adaptive energy uniformity threshold, a dual threshold deviation function is constructed, and the deviation function is used to perform weighted amplification or suppression processing on different deviation sources to form a regional weighted deviation amount. By coupling the weighted deviation of the above-mentioned regions with the shot peening sensitivity weight of the above-mentioned regions, the deviation of the highly sensitive regions is given a higher adjustment priority, thereby forming a residence deviation that is more adapted to the physical characteristics of the regions.

[0075] For example, the process of jointly calculating the dwell deviation of the target shot peening position based on the energy offset and coverage deviation of the point can be implemented using a mathematical framework that first constructs a double-threshold deviation function and then combines it with regional sensitivity weighted amplification, thereby quantifying the originally abstract deviation magnitude and adjustment priority into a calculable dwell deviation.

[0076] For those currently in the th For each process area (such as the straight groove section at the inlet, the rounded corner section at the bottom, or the rotary exit section), the shot peening point is first obtained from the online monitoring results to determine the current coverage index of that shot peening point. (For example, the current estimated coverage rate is a coverage adequacy indicator), and the current energy uniformity indicator. (For example, a measure of the uniformity of the shot peening energy distribution index). Meanwhile, an adaptive coverage threshold has already been calculated for this region in the preceding steps. and adaptive energy uniformity threshold Based on these two adaptive thresholds, this embodiment constructs a dual-threshold deviation function to simultaneously measure coverage deviation and energy deviation. For example, normalized coverage deviation and energy deviation can be defined first:

[0077] in, This indicates that the current coverage is below the adaptive coverage threshold, and there is a risk of insufficient coverage. This indicates that coverage has reached or exceeded the target level; This indicates that the energy fluctuation or unevenness is higher than the permissible level. This indicates that the energy distribution is relatively stable.

[0078] To construct the dual-threshold bias function, this embodiment applies different weights and nonlinear scaling to coverage bias and energy bias. For example, a region-weighted bias is defined. :

[0079] in, and These are the weighting coefficients for coverage bias and energy bias, used to control the relative influence of the two types of bias in the overall bias. and These are deviation functions designed for different physical meanings. For example, to more effectively amplify insufficient coverage, one could... Designed as a piecewise function sensitive to negative bias:

[0080] in, This is a coverage deviation amplification factor, so that when coverage is insufficient... It is a negative value and is scaled up proportionally, but no further negative bias is added when the coverage has reached or exceeded the standard; correspondingly, in order to suppress excessively high energy or excessively uneven energy distribution, it will... Designed as a piecewise function sensitive to positive deviation:

[0081] in, This is the energy deviation amplification factor. When the energy uniformity exceeds the threshold (i.e., the fluctuation is too large), Positive values ​​are amplified, while no additional positive bias is introduced when the energy remains within acceptable limits. This dual-threshold function design allows for differentiated reinforcement of insufficient coverage and energy anomalies: insufficient coverage is biased towards producing larger negative biases, and energy anomalies are biased towards producing larger positive biases, thus reflecting in subsequent dwell time adjustments areas that require extended dwell time and areas that require shortened dwell time or reduced impact.

[0082] After obtaining the weighted deviation of the above-mentioned region Subsequently, this embodiment does not directly... Instead of using it as a basis for adjusting the residence status, it is further combined with the previously calculated regional shot peening sensitivity weight. Perform coupling calculations. Area shot peening sensitivity weights. It is constructed based on characteristics such as historical energy spectrum fluctuations, coverage efficiency convergence speed, and roughness fluctuations after shot peening, and is used to characterize the first... The sensitivity of the region to parameter disturbances and changes in shot peening conditions. When A larger value indicates that this area is prone to quality fluctuations or abnormal results in historical processing, requiring higher adjustment priority in the current shot peening process. For this purpose, a residence deviation can be defined. for:

[0083] in, This is the sensitivity amplification factor, used to control the strength of the sensitivity weight's amplification of the bias; when hour, and Equal; when When it is large, A value significantly greater than 1 amplifies the same physical deviation into a larger residence deviation in sensitive regions. This results in a more aggressive response to deviations in geometrically complex, historically unstable, or quality-sensitive areas.

[0084] From a physical perspective, When this occurs, it indicates that the results based on the dual threshold function and sensitivity coupling in this area are biased towards the side with insufficient coverage / low energy. The subsequent dwell control module will tend to increase the dwell time or perform local spraying. When the coverage is too dense / energy too high, the system tends to shorten the dwell time or weaken the local enhancement. Through this joint solution mechanism of first constructing a dual-threshold deviation function and then overlaying regional shot peening sensitivity weights, this embodiment transforms the energy offset and coverage deviation of the current shot peening point into a dwelling deviation quantity that combines geometric characteristics, historical stability, and regional sensitivity. This provides a quantitative and feasible computational basis for subsequently solving the regional dwelling correction factor and dwelling adjustment vector based on the dwelling deviation quantity.

[0085] In one feasible implementation, the calculation of the residence adjustment vector based on the aforementioned regional residence correction factor includes: A multivariate nonlinear mapping model based on the regional geometric difficulty coefficient, the aforementioned shot peening sensitivity weight, and real-time acoustic and vibration characteristics is constructed, and the aforementioned regional dwelling correction factor is projected onto the three-dimensional dwelling control space. In the aforementioned three-dimensional dwell control space, the dwell time increment, dwell time decrement, and dwell point fine-tuning displacement are solved by using a multivariable coupling function, so that the adjustment amounts of different physical dimensions can be automatically balanced according to the regional characteristics. The solution weights are dynamically adjusted based on the changing trends of each component in the three-dimensional dwell control space, thereby forming a dwell adjustment vector that can adapt to the processing difficulty of different regions.

[0086] For example, the process of calculating the dwell adjustment vector based on the aforementioned regional dwell correction factor is not simply multiplying the dwell correction factor by a coefficient to obtain the time increment or decrement. Instead, it involves constructing a multivariate nonlinear mapping model that integrates regional geometric difficulty, shot peening sensitivity, and real-time acoustic and vibration characteristics. This model projects the regional dwell correction factor into a three-dimensional dwell control space, and then solves for three components—dwell time increment, dwell time decrement, and dwell point fine-tuning displacement—within this space. The solution weights are dynamically adjusted based on the trends of these components as the working conditions change, thereby forming a dwell adjustment vector that can adapt to the processing difficulty of different regions.

[0087] Specifically, the first The area retention correction factor for a process region (e.g., the straight groove section at the opening, the rounded corner section at the bottom, or the rotary exit section) within the current shot peening cycle is denoted as... The geometric difficulty coefficient, which reflects the geometric complexity of the region, has been calculated in the aforementioned steps. And shot peening sensitivity weights obtained based on historical energy spectrum fluctuations, coverage efficiency, and roughness fluctuations. Furthermore, several features such as energy density, dominant frequency shift, and envelope stability can be extracted from real-time acoustic vibration monitoring, and these real-time acoustic vibration features can be organized into a vector. Based on this, this embodiment first constructs the input feature vector for solving the problem. :

[0088] in, This represents a scalar or low-dimensional vector obtained after normalizing, principal component extraction, or combining several features of real-time acoustic vibration features. This vector is used to compress the dimensionality of acoustic vibration features, avoiding the direct use of the high-dimensional original signal. Thus, This comprehensively reflects the extent to which adjustments are needed in the current region. How difficult is characterization and processing? Characterization, "How sensitive to process disturbances" is determined by The characterization and whether the impact state is stable at this moment are determined by Characterization.

[0089] After constructing the above input feature vectors, the process of calculating the residence adjustment vector based on the above-mentioned region residence correction factor is abstracted into a multivariate nonlinear mapping model, namely:

[0090] in, This is a dwell time increment used to extend dwell time when coverage is insufficient or energy is low; To reduce dwell time, used to shorten dwell time when energy is too high or coverage is too dense; It allows for fine-tuning of the dwell point displacement, enabling small-scale spatial adjustments to the dwell position in complex geometric regions. This refers to the dwell adjustment vector in the three-dimensional dwell control space. To reflect the characteristics of multivariable nonlinear mapping, this embodiment adopts the following form of nonlinear mapping:

[0091] Where M is the weight matrix, used to linearly combine the input feature vectors into the three-dimensional control space; and is a hyperbolic tangent nonlinear function, used to saturate the output so that the dwell time increment, decrement, and fine-tuning displacement do not amplify indefinitely. This is a diagonal gain matrix for scaling the three components, where the diagonal elements correspond to the amplification factors of the time increment, time decrement, and displacement components, respectively. For example, it can be written as:

[0092] so, This determines the upper limit and sensitivity of the dwell time increment under the same input features. Control the magnitude of the reduction in dwell time. The maximum range of fine-tuning displacement of the dwell point is controlled. In this way, the regional dwell correction factor... Geometric difficulty level Shot peening sensitivity weight The output in the three-dimensional dwell control space is determined together with the real-time acoustic and vibration characteristics. This allows the stay adjustment to reflect both the magnitude of the current deviation and the differentiated requirements of regional geometry and historical behavior.

[0093] In the aforementioned three-dimensional dwell control space, to ensure that the adjustment amounts of different physical dimensions can be automatically balanced according to regional characteristics, this embodiment employs directional explicit coupling function forms for each of the three components. For example, for dwell time increments and decrements, a half-wave activation mechanism corresponding to the sign of the dwell deviation can be used to prevent them from being simultaneously non-zero, thus avoiding the contradictory situation of simultaneous addition and subtraction; for dwell point fine-tuning displacement, the displacement direction can be determined based on the regional geometric accuracy and the sign of the dwell deviation. A feasible construction method is to further decompose the time component based on the aforementioned nonlinear mapping, for example:

[0094] in These represent the first and second components of the nonlinear mapping output vector, respectively. Ensure that both time increments and decrements are non-negative; when the current deviation is insufficiently covered / energy is low, the first component will be larger while the second component will approach zero, and vice versa, the second component will dominate. For fine-tuning the displacement of the dwell point... The magnitude and direction of the displacement can be directly given using the third component, for example:

[0095] in The third component of the nonlinear mapping output, after... After the limit is set, then by By controlling the maximum displacement, this is effective in geometrically complex areas with large deviations. It will be slightly larger, thus allowing the dwell point to shift slightly towards more critical stress-sensitive locations; while in geometrically simpler or less biased areas, It will automatically converge to near zero, avoiding over-correction.

[0096] To adapt this multivariate nonlinear mapping model to variations in different regions and batch processing conditions, this embodiment also introduces a mechanism for dynamically adjusting the solution weights based on the changing trends of each component in the aforementioned three-dimensional dwell control space. Specifically, the three-dimensional control vectors can be statistically analyzed within each region or time window. Historical output sequences, for example, examining and The mean and variance of batches or over time, and their correlation with final quality indicators (such as residual stress level, roughness, and crack defect rate). If a certain area is found to still have an insufficient coverage trend after multiple processing cycles, and the quality does not improve significantly even after increasing the dwell time multiple times, the corresponding area can be appropriately enlarged. Or adjust the pair in matrix M The weighting of the time increments allows for larger increments under similar biases; conversely, if over-enhancement or excessive surface damage is observed in a certain area, the weighting can be reduced. Increase Alternatively, relevant parameters can be adjusted to make the decrease in residence time more sensitive to the same deviation. Mathematically, this can be described using a simple adaptive update form, for example:

[0097] Among them, superscript Indicates the first Each processing batch or control cycle It can represent the first The region in Coverage bias residuals in the batch This represents the residual of the energy distribution deviation. This is the step size or learning rate function. Through this dynamic adjustment based on output trends and quality feedback, the solution weights of each component in the three-dimensional dwell control space will gradually converge to reasonable values ​​suitable for the processing difficulty and historical performance of different regions over time, thus forming a dwell adjustment vector solution mechanism that is adaptive to regional differences.

[0098] In summary, this embodiment constructs a multivariate nonlinear mapping model to uniformly model the regional dwell correction factor, regional geometric difficulty, shot peening sensitivity, and real-time acoustic and vibration characteristics. This model is then projected onto a three-dimensional dwell control space, where dwell time increment, dwell time decrement, and dwell point fine-tuning displacement are solved. The solution weights are dynamically adjusted based on the historical trends of each component, thereby forming a dwell adjustment vector that can reflect the current deviation and adapt to the processing difficulty of different regions. This provides a precise, adjustable, and self-learning control foundation for the subsequent collaborative updating of dwell strategies and multi-axis CNC paths.

[0099] In one feasible implementation, the above-mentioned synchronous mapping of the dwell adjustment vector with the corresponding position of the CNC path to achieve coordinated updating of the dwell strategy and the multi-axis trajectory of the machine tool includes: After obtaining the above-mentioned dwell adjustment vector, a multi-axis attitude compensation matrix is ​​constructed based on the machine tool A / C axis attitude, the spatial normal calculation results of the above-mentioned CNC path, and the real-time shot peening position. The time component in the dwell adjustment vector is coupled with the spatial attitude component in the compensation matrix so that the dwell adjustment can automatically compensate for attitude deviation while extending or shortening the dwell time, and maintain the normal alignment relationship between the fixed ultrasonic vibrating head and the workpiece surface. Based on the aforementioned multi-axis attitude compensation matrix, the spatial interpolation path is dynamically corrected, ensuring that the shot peening trajectory maintains continuity and coverage without blind spots after dwell adjustment.

[0100] For example, the process of synchronously mapping the above-mentioned dwell adjustment vector with the corresponding position of the above-mentioned CNC path to realize the collaborative update of dwell strategy and machine tool multi-axis trajectory can be understood as mapping the dwell adjustment in time and the attitude compensation and position fine adjustment in space to the motion commands of machine tool A / C axis and feed axis without disrupting the overall continuity of the original CNC spatial interpolation trajectory, thereby realizing the integrated collaboration of dwell control and trajectory control.

[0101] Specifically, for any point on the CNC path, parameters can be used. Its position on the path can be represented by the spatial position and orientation, which can be denoted as: and .in, Let be the spatial position vector of the workpiece relative to the fixed ultrasonic vibrating head in the machine tool coordinate system. The attitude rotation matrix, calculated from the machine tool's A-axis and C-axis angles, reflects the relationship between the workpiece surface normal and the machine tool coordinate system at that location. In the preceding steps, the dwell adjustment vector for the region to which this path point belongs has been calculated based on the shot peening coverage deviation and energy distribution.

[0102] in, To increase the length of stay, To reduce the length of stay, To fine-tune the displacement of the dwell point along the CNC path or normal direction.

[0103] After obtaining the aforementioned dwell adjustment vector, firstly based on the machine tool's current A-axis angle... C-axis angle And the spatial normal solution derived from the CNC path. Construct a multi-axis attitude compensation matrix. The current attitude rotation matrix can be represented as:

[0104] in, Let C be the rotation matrix about the C-axis. Let A be the rotation matrix about the A-axis. Combining the real-time shot reading position and normal deviation evaluation results, the minute attitude correction amount required to accurately align the workpiece surface normal with the impact direction of the fixed ultrasonic vibrating head can be calculated. . Based on this, a multi-axis attitude compensation matrix is ​​constructed:

[0105] This compensation matrix automatically compensates for normal offsets caused by dwell adjustments, micro-dwellings, or path disturbances by fine-tuning the A / C axis attitude.

[0106] Couple the time component of the above-mentioned dwell adjustment vector with the spatial attitude component of the above-mentioned compensation matrix: On the one hand, the combined quantity of increases and decreases in residence time

[0107] Adjust the feed rate or dwell time at that path point, for example, by adjusting the original feed rate. Revised to

[0108] in, Adjust the scaling factor for time, when When an extended stay is required, The corresponding reduction leads to local deceleration or even a brief pause; when When a shorter stay is required, The corresponding increase allows for earlier departure. On the other hand, while adjusting the time, an attitude compensation matrix is ​​used... The attitude fine-tuning is superimposed onto the original attitude matrix to obtain the updated attitude:

[0109] In this way, while the dwell time is extended or shortened, the workpiece posture is also adjusted synchronously, thereby maintaining the normal alignment between the fixed ultrasonic vibrating head and the workpiece surface, and avoiding the reduction of the strengthening effect due to the impact direction deviating from the normal due to local dwell adjustment.

[0110] For fine-tuning the displacement of the dwelling point In this embodiment, it is mapped to CNC path parameters. Or, at the corresponding spatial location, for example, the updated path parameters can be set as:

[0111] in, This is a scaling factor used to map the scalar displacement of the dwell point to the changes in path parameters. When When the indicator should slightly shift forward or backward to the desired position, adjust accordingly. This alters the precise landing point of the dwell point on the spatial interpolation path, bringing it closer to areas of local stress concentration or geometric weakness. Spatially, this corresponds to:

[0112] That is, the position is slightly offset along the original spatial interpolation path, so as to achieve local repositioning of the dwell point while maintaining the smoothness and continuity of the overall trajectory.

[0113] After completing the mapping of the time and space dimensions described above, this embodiment further dynamically corrects the entire spatial interpolation path based on the aforementioned multi-axis attitude compensation matrix. This is achieved by adjusting the attitude matrices of several key nodes along the path. Overlay It uses interpolation to smoothly transition between adjacent nodes, generating an updated spatial interpolation path. In this way, even if delayed dwell, early disengagement, or local micro-dwelling actions are performed in certain areas, the entire shot peening trajectory remains continuous in three-dimensional space, without any abrupt changes in attitude or trajectory interruption. At the same time, through the synergistic effect of attitude compensation and dwell position fine-tuning, the shot peening trajectory of the complex U-shaped inner surface is guaranteed to have no dead angle coverage. This ensures that the straight groove section at the inlet, the rounded corner section at the bottom, and the rotating exit section can still maintain uniform, stable, and normal-consistent impact conditions after dwell adjustment, ultimately achieving a more continuous and consistent residual stress field distribution and strengthening effect.

[0114] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for controlling ultrasonic shot peening of a U-shaped inner cavity of a high-temperature alloy low-vortex disk, characterized in that, include: The U-shaped inner cavity of the target low-volume disk workpiece is divided into three process areas according to its geometric shape: the straight groove section at the opening, the rounded corner section at the bottom, and the rotary exit section. The low-volume vortex workpiece is fixed on the high-precision rotary table of a multi-axis CNC machine tool using a special fixture, and a precise workpiece coordinate system is established. After the workpiece coordinate system is established, a CNC path is generated based on the CAD model corresponding to the target low-volume disk workpiece to drive the workpiece to move relative to the fixed ultrasonic vibrating head, so that different process areas can enter the effective impact range in a trajectory that matches their shape. Acoustic and vibration signals are collected in real time during ultrasonic shot peening, and the shot peening coverage effect and energy distribution are dynamically analyzed based on the acoustic and vibration signals. The dwell time is automatically adjusted based on the shot peening coverage effect and the energy distribution state, so that the dwell time and dwell position of each area can be adaptively adjusted and coordinated with the CNC path control.

2. The ultrasonic shot peening control method for the U-shaped inner cavity of a high-temperature alloy low-vortex disk according to claim 1, characterized in that, The process of dividing the U-shaped inner cavity of the target low-volume disk workpiece into three process areas based on its geometric shape—a straight groove section at the inlet, a rounded corner section at the bottom, and a rotary exit section—includes: Based on the curvature distribution, wall thickness variation, and geometric transition characteristics of the U-shaped inner cavity, geometric feature recognition is performed on the inner cavity surface to obtain the geometric feature recognition results; Based on the geometric feature recognition results, the curvature gradient abrupt change location and the continuity boundary of the geometric shape are determined; wherein, the linear groove surface with curvature less than a preset threshold and wall thickness change less than a preset difference is divided into the groove opening straight groove segment, the curved surface with continuous curvature change and smooth transition trend is divided into the bottom rounded corner segment, and the shape region with steep curvature change and tending towards the exit direction is divided into the rotary exit segment.

3. The ultrasonic shot peening control method for the U-shaped inner cavity of a high-temperature alloy low-vortex disk according to claim 1, characterized in that, The process of generating a CNC path based on the CAD model corresponding to the target low-vortex disk workpiece to drive the workpiece to move relative to the fixed ultrasonic vibrating head, enabling different process areas to enter the effective impact range with trajectories matching their morphology, includes: Based on the CAD model corresponding to the target low-volume disk workpiece, the curvature, normal vector and reachability features of the straight groove section, the bottom rounded corner section and the rotary exit section are extracted respectively to obtain the geometric constraint information of each process area; A linear CNC path is planned for the straight groove section of the slot, in which the workpiece rotates at a constant speed and is fed axially at a constant speed, so that the straight groove section of the slot enters the effective impact range of the fixed ultrasonic vibrating head at a constant speed. Based on the continuous curvature variation characteristics of the bottom rounded corner segment, an A / C axis linked spatial interpolation path is generated, and a programmable dwell point is set in the spatial interpolation path, so that the bottom rounded corner segment can achieve deep reinforcement while maintaining normal alignment. A rotation compensation path is planned for the rotary exit section in combination with its curvature change direction, and a regional transition section is set in the rotary compensation path so that the rotary exit section remains consistent with the impact direction of the fixed ultrasonic vibrating head when transitioning to the exit direction.

4. The ultrasonic shot peening control method for the U-shaped inner cavity of a high-temperature alloy low-vortex disk according to claim 1, characterized in that, The method of real-time acquisition of acoustic and vibration signals during ultrasonic shot peening, and dynamic analysis of shot peening coverage and energy distribution based on the acoustic and vibration signals, includes: After ultrasonic shot peening begins, the acoustic and vibration waveforms from the shot peening area are acquired in real time, and bandpass filtering and normalization are performed on the acoustic and vibration waveforms to obtain acoustic-vibration feature vectors characterizing local impact properties. Energy spectrum analysis of the impact process is performed based on the energy density, dominant frequency distribution, and envelope morphology of the acoustic feature vector, and the shot peening energy distribution index is calculated from the analysis results. The shot peening energy distribution index is compared with the historical feature data of the corresponding position in the CNC path to determine the shot peening coverage effect of the current shot peening area.

5. The ultrasonic shot peening control method for the U-shaped inner cavity of a high-temperature alloy low-vortex disk according to claim 1, characterized in that, The automatic adjustment of the dwell time based on the shot peening coverage effect and the energy distribution state, enabling adaptive adjustment of the dwell time and dwell position in each area and coordinated control with the CNC path, includes: Based on the curvature characteristics of different process regions and historical shot peening parameters, set adaptive coverage thresholds and energy uniformity thresholds for the region; The dwell deviation of the target shot peening position is calculated by combining the energy offset and coverage offset of the current shot peening point, so as to form the area dwell correction factor; The residence adjustment vector is calculated based on the residence correction factor of the region, wherein the residence adjustment vector includes residence time increment, residence time decrement and residence point fine-tuning displacement; The dwell adjustment vector is synchronously mapped to the corresponding position of the CNC path to achieve coordinated updating of the dwell strategy and the multi-axis trajectory of the machine tool; The updated residence time is written into the real-time shot peening control command to drive the shot peening system to perform delayed residence, early detachment, or local micro-relocation actions.

6. The ultrasonic shot peening control method for the U-shaped inner cavity of a high-temperature alloy low-vortex disk according to claim 5, characterized in that, The adaptive coverage threshold and energy uniformity threshold for the region are set based on the curvature characteristics of different process regions and historical shot peening parameters, including: Based on the curvature characteristics of different process regions, the curvature value, curvature change rate and normal reachable angle of each process region are calculated respectively. Based on the curvature value, curvature change rate and normal reachable angle, a geometric difficulty coefficient reflecting the shot peening difficulty of the region is constructed to form a set of regional geometric feature parameters. Based on the historical shot peening parameters, energy spectrum distribution, coverage efficiency, dwell response curve and surface roughness index after shot peening, a regional shot peening response model is established through a regional data weighting mechanism, and corresponding shot peening sensitivity weights are generated for different regions. The set of regional geometric feature parameters, the shot peening sensitivity weights, and the regional shot peening response model are jointly solved to generate a multi-factor threshold calculation formula that includes regional geometric difficulty, shot peening sensitivity, and historical convergence trend. The adaptive coverage threshold and adaptive energy uniformity threshold for each region are then calculated.

7. The ultrasonic shot peening control method for the U-shaped inner cavity of a high-temperature alloy low-vortex disk according to claim 5, characterized in that, The calculation of the dwell deviation of the target shot peening position based on the energy offset and coverage offset of the current shot peening point includes: A dual-threshold deviation function is constructed based on the adaptive coverage threshold and the adaptive energy uniformity threshold, and the deviation function is used to perform weighted amplification or suppression processing on different deviation sources to form a regional weighted deviation amount. The weighted deviation of the region and the shot peening sensitivity weight of the region are coupled and calculated to give the deviation of the highly sensitive region a higher adjustment priority, thereby forming a residence deviation that is more adapted to the physical characteristics of the region.

8. The ultrasonic shot peening control method for the U-shaped inner cavity of a high-temperature alloy low-vortex disk according to claim 5, characterized in that, The step of calculating the residence adjustment vector based on the regional residence correction factor includes: A multivariate nonlinear mapping model based on the regional geometric difficulty coefficient, the shot peening sensitivity weight, and the real-time acoustic and vibration characteristics is constructed, and the regional residence correction factor is projected onto the three-dimensional residence control space. In the three-dimensional dwell control space, the dwell time increment, dwell time decrement and dwell point fine-tuning displacement are solved by using a multivariable coupling function, so that the adjustment amounts of different physical dimensions can be automatically balanced according to the regional characteristics. The solution weights are dynamically adjusted based on the changing trends of each component in the three-dimensional dwell control space, thereby forming a dwell adjustment vector that can adapt to the processing difficulty of different regions.

9. The ultrasonic shot peening control method for the U-shaped inner cavity of a high-temperature alloy low-vortex disk according to claim 5, characterized in that, The step of synchronously mapping the dwell adjustment vector to the corresponding position of the CNC path to achieve coordinated updating of the dwell strategy and the multi-axis trajectory of the machine tool includes: After obtaining the dwell adjustment vector, a multi-axis attitude compensation matrix is ​​constructed based on the machine tool A / C axis attitude, the spatial normal calculation result of the CNC path, and the real-time shot peening position. The time component in the dwell adjustment vector is coupled with the spatial attitude component in the compensation matrix so that the dwell adjustment automatically compensates for attitude deviation while extending or shortening the dwell time, maintaining the normal alignment between the fixed ultrasonic vibrating head and the workpiece surface. The spatial interpolation path is dynamically corrected based on the multi-axis attitude compensation matrix, so that the shot peening trajectory maintains continuity and coverage without dead angles after the dwell adjustment.

Images