A curved flow channel additive and subtractive hybrid manufacturing system based on curvature self-adaption and occlusion determination

Abstract

The application provides a curved flow channel additive and subtractive hybrid manufacturing system based on curvature self-adaptation and shielding determination, which comprises the following steps: dividing the curvature interval, matching the corresponding effective incident angle to set the adaptive additive layer thickness; when the additive is carried out, the light path accessibility prediction is simultaneously carried out, that is, based on the current formed entity, the light path accessibility of the predicted generated geometry is predicted after the preset additive layer thickness is simulated and superimposed, the light shielding is predicted, then the femtosecond laser subtractive process preparation is simultaneously completed, after the single round of additive part is completed, the femtosecond laser subtractive is immediately stopped and triggered; during the femtosecond laser subtractive process, the inner wall region which will be shielded is removed with a reserved margin; and the formed geometry is updated and circulated. Through the closed-loop control of "curvature-layer thickness-shielding determination-alternating interface", the unprocessed risk caused by the shielding of the inner wall of the curved flow channel is reduced, the machining precision and efficiency are considered, and the application is suitable for efficient and precise manufacturing of complex curved flow channels.

Description

Technical Field

[0001] This invention belongs to the field of additive manufacturing and laser precision machining, specifically relating to a hybrid additive and subtractive manufacturing system for curved flow channels based on curvature adaptation and occlusion determination. Background Technology

[0002] Curved flow channels are widely used in aerospace, automotive, and high-end equipment fields. The surface quality and dimensional accuracy of their inner walls directly affect fluid transfer efficiency, heat dissipation, and reliability. Powder bed additive manufacturing (such as laser powder bed fusion molding) can achieve the integrated molding of complex curved flow channels, but the inner walls usually have problems such as residual powder adhesion, step effect, large surface roughness, and local dimensional deviations, which require subsequent finishing to meet the application requirements.

[0003] Femtosecond lasers, as an ultrafast laser processing method, offer advantages such as non-contact processing, a small heat-affected zone, and high processing precision, making them suitable for precision subtractive machining of curved flow channel inner walls. However, in machining curved flow channel inner walls, the interaction between curvature differences and limited incident windows can easily lead to typical problems of optical path obstruction. For example, in scenarios such as "machining the right inner wall from the left side," as the additive manufacturing process progresses, due to engineering constraints such as the laser's reachable space, incident angle range, working distance, and beam diameter, some areas of the inner wall may become unprocessable due to the loss of the laser's direct line of sight, leading to the risk that "subsequent subtractive machining of that area is impossible." This problem is more pronounced in high curvature regions.

[0004] Existing hybrid manufacturing methods employ traditional contact-based subtractive processing instead of femtosecond laser precision machining. They utilize a fixed layer thickness or a fixed alternating cycle of additive and subtractive machining throughout the process. Furthermore, traditional subtractive machining often fails to guarantee machining accuracy, resulting in extremely high heat-affected zones and increasing the risk of equipment collisions. It is difficult to balance forming efficiency and equipment accessibility in high-curvature areas. Considering the characteristics of femtosecond laser machining, current technologies have not yet integrated femtosecond lasers with curvature-adaptive layer thickness, nor have they proposed an alternating mechanism of "stopping additive machining and switching to subtractive machining" triggered by occlusion detection. Therefore, there is an urgent need in this field for a hybrid manufacturing system that integrates curvature-adaptive layer thickness setting, occlusion detection-triggered alternating additive and subtractive machining, femtosecond laser precision subtractive machining, and dynamic incident point transformation to fundamentally resolve the efficiency and accessibility contradictions of traditional hybrid manufacturing. Summary of the Invention

[0005] The technical problem to be solved by this invention is: during femtosecond machining of the inner wall of a curved flow channel, the problem of optical path obstruction and inaccessibility caused by curvature difference and limited coupling of the incident window, as well as the contradiction between efficiency and accessibility caused by fixed layer thickness and fixed alternation period. This invention provides a hybrid manufacturing system based on curvature-incident angle linkage adaptive layer thickness and using obstruction determination to trigger material addition and subtraction alternation, so as to reduce the risk of unprocessed inner wall due to obstruction and improve machining stability.

[0006] To achieve the above objectives, the present invention provides the following solution: A hybrid additive and subtractive manufacturing system for curved flow channels based on curvature adaptation and occlusion determination includes: The curvature calculation module is used to obtain a three-dimensional model of the target curved flow channel, discretize the inner wall surface to obtain the inner wall point set and calculate the curvature distribution of the inner wall point set, divide the curvature interval, and synchronously match the effective incident angle range of the femtosecond laser corresponding to each curvature interval. The additive layer thickness setting module is used to set an adaptive additive layer thickness based on the curvature range and the corresponding effective incident angle range of the femtosecond laser. The optical path reachability prediction module is used to perform optical path reachability prediction on the expected generated geometry by using powder bed additive manufacturing technology to add materials layer by layer according to the adaptive additive layer thickness and simultaneously performing optical path reachability prediction during the additive process. Based on the currently formed entity, the module simulates and superimposes the preset additive layer thickness to predict the optical path reachability of the expected generated geometry. The subtractive material triggering module is used to simultaneously complete the preparation of the femtosecond laser subtractive material process when it is predicted that light occlusion will occur during the continuation of additive material. After the current additive part is completed, it immediately stops and triggers the femtosecond laser subtractive material process. The subtraction module is used to change the incident point in real time along the circumference or axis during the femtosecond laser subtraction process, dynamically adjust the incident attitude within the effective incident angle range of the corresponding curvature interval, and perform subtraction on the inner wall area that is currently accessible and will be blocked in the next stage under the constraints of incident angle and working distance, removing the reserved allowance. The repeat operation module is used to repeatedly trigger the optical path reachability prediction module and the subtractive material module after the optical path reachable geometry is formed, until the entire curved flow channel is manufactured.

[0007] Preferably, in the curvature calculation module, the process of dividing the curvature interval and matching the effective incident angle range includes: Based on the three-dimensional model of the target curved flow channel, the target curved surface of the inner wall of the flow channel is extracted and discretized based on the NURBS surface discretization algorithm to form a set of inner wall points for occlusion determination; The local curvature k of the inner wall is calculated based on the discrete surface, and the inner wall is divided into at least two curvature intervals according to the curvature magnitude, including a high curvature interval, a medium curvature interval, and a low curvature interval. Among them, the high curvature interval k≥18m -1 Medium curvature range 8m -1 <k<18m -1 Low curvature interval k≤8m -1 .

[0008] Preferably, in the additive layer thickness setting module, the layer thickness parameters used during additive manufacturing in each curvature range are related as follows: high curvature layer thickness < medium curvature layer thickness < low curvature layer thickness; the layer thickness in each range is linked and matched with the corresponding effective incident angle range.

[0009] Preferably, in the optical path reachability prediction module, the process of performing optical path reachability prediction includes: Based on the existing solid, the predicted geometry is generated by simulating the superposition of preset additive layers. A set of pre-defined femtosecond laser incident point locations is used, along with a defined effective incident angle range, working distance / attitude reachability constraint, and spot radius for the corresponding curvature interval. Optical path reachability prediction is performed on the set of inner wall points. When the optical path intersects with the predicted geometry, it is determined to be an obstruction. The optical path reachability prediction judgment also includes ensuring that the spot radius is not truncated, the incident attitude meets the allowable incident angle range of the corresponding curvature interval, and the working distance is within the constraint, and searching for the incident direction within the allowable incident angle range of the corresponding curvature interval.

[0010] Preferably, the optical path reachability prediction module is performed synchronously during the additive manufacturing process, and the number of continuous additive layers in a single round is adaptively set according to the inner wall curvature and the corresponding effective incident angle range; 1 to 3 layers are continuously added in the high curvature range, and 3 to 10 layers are continuously added in the low curvature range; the optical path reachability prediction is completed synchronously during the single round of additive manufacturing.

[0011] Preferably, the set of femtosecond laser incident points is distributed in the accessible space above and adjacent to the additive direction, the process window reserved during the additive process, and the preset incident trajectory; The real-time change of the femtosecond laser incident point position includes discrete switching or continuous movement, and the incident attitude is dynamically adjusted within the effective incident angle range of the corresponding curvature interval, subject to the constraints of equipment mechanics and working distance.

[0012] Preferably, in the subtractive material module, the femtosecond laser subtractive material includes one or more combinations of femtosecond laser ablation, femtosecond laser micro-removal, or femtosecond laser polishing.

[0013] Preferably, after femtosecond laser subtraction, the surface morphology or local residual material of the inner wall is detected. When the set conditions are not met, the incident point trajectory, incident attitude, femtosecond laser parameters and the execution frequency of the next occlusion determination are corrected.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) Effectively manage the risk of optical path occlusion: With the hard constraint of "switching when any inner wall point is unreachable", the occlusion risk is identified in advance through synchronous prediction during the additive manufacturing process, avoiding the accumulation of occlusion due to continued additive manufacturing, and reducing the risk of "a certain section of the inner wall becoming permanently unprocessable". (2) Curvature-incident angle-layer thickness multi-dimensional adaptive: The greater the curvature, the narrower the effective incident angle range, the smaller the layer thickness, and the more frequent the judgment, so that the high curvature area can complete the material reduction of the reachable area before the shading occurs, thus improving the feasibility of the process; (3) Determine no missed detection: By establishing a quantitative relationship between the inner wall dispersion step length and the radius of curvature, the high curvature area is automatically encrypted for sampling, and combined with the global direction search within the corresponding incident angle range, the reliability of occlusion determination is improved.

[0015] (4) More comprehensive inner wall processing: Femtosecond subtractive processing adopts dynamic incident point and incident angle adaptive adjustment within the corresponding curvature range, expands the reachable space, and improves the range of machinable inner wall area and surface quality stability; (5) Balancing efficiency and quality: In low curvature areas, a larger layer thickness and a lower decision frequency can be used to improve efficiency, while in high curvature areas, accessibility can be ensured by a smaller layer thickness and a higher decision frequency, thus achieving a balance between overall efficiency and quality. Attached Figure Description

[0016] To more clearly illustrate the technical solution of the present invention, the drawings used in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram illustrating the prediction of occlusion during the processing of curved flow channels according to an embodiment of the present invention; Figure 2 This is a schematic diagram of curvature partitioning according to an embodiment of the present invention; Figure 3 This is a flowchart of the closed-loop control of "additive manufacturing - occlusion determination - femtosecond subtraction" in this invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0020] Example 1: like Figure 1 , Figure 2 , Figure 3As shown, a hybrid additive and subtractive manufacturing system for curved flow channels based on curvature adaptation and occlusion determination includes: The curvature calculation module is used to obtain the three-dimensional model of the target curved flow channel, discretize the inner wall surface to obtain the inner wall point set and calculate the curvature distribution of the inner wall point set, divide the curvature interval, and synchronously match the effective incident angle range of the femtosecond laser corresponding to each curvature interval.

[0021] A further implementation method involves, in the curvature calculation module, the process of dividing the curvature interval and matching the effective incident angle range includes: Based on the three-dimensional model of the target curved flow channel, the target curved surface of the inner wall of the flow channel is extracted and discretized based on the NURBS surface discretization algorithm to form the inner wall point set P={pi} for occlusion determination, with a discretization chord error ≤0.003mm.

[0022] The local curvature k or radius of curvature ρ (k=1 / ρ) of the inner wall is calculated based on the discrete surface, and the inner wall is divided into at least two curvature intervals according to the curvature magnitude, including a high curvature interval, a medium curvature interval and a low curvature interval; Among them, the high curvature interval k≥18m -1 Medium curvature range 8m -1 <k<18m -1 Low curvature interval k≤8m -1 .

[0023] The effective incident angle range of femtosecond lasers corresponding to each curvature range narrows as the curvature increases, that is: high curvature incident angle range < medium curvature incident angle range < low curvature incident angle range; The additive layer thickness setting module is used to set an adaptive additive layer thickness based on the curvature interval and the corresponding effective incident angle range. A further implementation involves setting the layer thickness relationship for each curvature interval during additive processing in the additive layer thickness setting module as follows: high curvature layer thickness < medium curvature layer thickness < low curvature layer thickness, with each interval's layer thickness and corresponding effective incident angle range exhibiting a one-to-one linkage matching relationship. Specifically, an adaptive additive layer thickness h is set for different curvature intervals, synchronously matching the corresponding effective incident angle range, so that the larger the curvature, the smaller the corresponding layer thickness, and the narrower the effective incident angle range; and an initial value for the stage additive thickness is given, making the stage additive thickness smaller in the high curvature region and the occlusion detection more frequent, thereby reducing the risk of occlusion accumulation. A segmented model is preferably used: High curvature region: h H Medium curvature region: h M Low curvature region: h L And satisfy h H <h M <h L The corresponding effective incident angle range satisfies the following condition: high curvature range < medium curvature range < low curvature range.

[0024] This embodiment provides a specific numerical range for adaptive additive layer thickness: High curvature region (k ≥ 18 m) -1 The effective incident angle range is ±20° to ±35°, and the layer thickness is taken as the minimum value, ranging from 0.02 to 0.04 mm. Medium curvature range (8 m) -1 <k<18 m -1 The effective incident angle range is ±35° to ±50°, and the layer thickness is taken as the median value, ranging from 0.04 to 0.06 mm. Low curvature range (k ≤ 8 m) -1 ): The effective incident angle range is ±50° to ±65°, and the layer thickness is taken as the maximum value, ranging from 0.06 to 0.10 mm.

[0025] To avoid missed detections due to insufficient sampling in high curvature regions, a quantitative relationship is established between the inner wall dispersion step length and the local radius of curvature. Assume the allowable geometric approximation chord height error is e. max If the local radius of curvature is ρ, then the arc length step ΔS of adjacent sampling points is approximated using the constant sine height error: (1) This formula is applicable to sampling the inner wall of curved flow channels discretized by NURBS surfaces, and the applicable condition is e. max Take a value of 0.002~0.01mm, where ρ and e max A consistent unit (in mm) must be used, therefore the unit of ΔS is mm. When ρ decreases (curvature increases), ΔS automatically decreases to achieve sampling refinement in high-curvature areas. Preferably, e max Use a value of 0.002~0.01mm to balance the accuracy of the judgment with the amount of calculation.

[0026] The optical path reachability prediction module is used to perform optical path reachability prediction simultaneously during the additive manufacturing process by using powder bed additive manufacturing technology to add layers according to adaptive additive layer thickness. Based on the currently formed entity, after simulating and superimposing the preset additive layer thickness, the optical path reachability prediction is performed on the expected generated geometry.

[0027] A further implementation method is that the process of performing optical path reachability prediction in the optical path reachability prediction module includes: During the additive manufacturing process, optical path reachability prediction is performed simultaneously. Based on the existing solid Ω, the expected generated geometry Ω' is obtained by simulating the superposition of a preset additive layer thickness h. The set of femtosecond laser incident point positions E is preset, and the effective incident angle range of the corresponding curvature interval is defined. The working distance / attitude reachability constraint W and the spot radius r are used to predict and judge the optical path reachability of the inner wall point set. When the optical path intersects with the predicted geometry Ω', it is judged as occlusion. The real-time change of the femtosecond laser incident point position includes discrete switching or continuous movement, and the incident attitude is dynamically adjusted within the effective incident angle range of the corresponding curvature interval, subject to the mechanical constraints of the equipment and the working distance.

[0028] The optical path reachability prediction judgment also includes ensuring that the spot radius is not truncated, the incident attitude meets the allowable incident angle range of the corresponding curvature interval, and the working distance is within the constraint, and searching for the incident direction within the allowable incident angle range of the corresponding curvature interval.

[0029] A further implementation involves the femtosecond laser incident point location set being distributed above the additive direction, and in the accessible space, the process window reserved during the additive process, and the preset incident trajectory.

[0030] A further implementation method is that the optical path reachability prediction module is performed synchronously during the additive manufacturing process, and the number of continuous additive layers in a single round is adaptively set according to the curvature of the inner wall; 1 to 3 layers are continuously added in the high curvature range, and 3 to 10 layers are continuously added in the low curvature range; the optical path reachability prediction is completed synchronously during the single round of additive manufacturing.

[0031] Specifically, during each stage of the additive manufacturing process, occlusion determination is performed on the currently formed geometry and the predicted geometry. Let the predicted solid be Ω', and for any inner wall point... ∈P, define its reachability indicator function: (2) Among them, e This represents the straight light path (or ray segment) from the incident point e along the incident direction θ to the inner wall point p; when the light path intersects with the predicted forming entity Ω', it is determined to be an obstruction. When the prediction determines that an obstruction has occurred, i.e. the light cannot reach the inner wall point (A(p)=0), the preparation for the femtosecond laser subtractive process is completed simultaneously. After completing the current additive part, it is immediately stopped and the femtosecond laser subtractive process is entered; after the subtraction is completed and the geometry is updated, the additive process is resumed and the next cycle begins.

[0032] The subtractive material triggering module is used to simultaneously complete the preparation of the femtosecond laser subtractive material process when it is predicted that light occlusion will occur during continued additive material production (i.e., when A(p)=0 is satisfied for the predicted forming geometry Ω'). After completing the current additive part, the additive material production is stopped immediately and the femtosecond laser subtractive material is triggered. The subtraction module is used in the femtosecond laser subtraction process to change the incident point in real time along the circumference or axis. Within the effective incident angle range of the corresponding curvature interval, the incident attitude is dynamically adjusted. Under the constraints of the incident angle and working distance, subtraction is performed on the inner wall region that is currently reachable (A(p)=1) and will be blocked in the next stage, removing the reserved allowance, eliminating steps, and improving the surface morphology of the inner wall. During the femtosecond laser subtraction process, the attitude is adjusted to ensure that the angle between the laser beam and the local normal of the inner wall of the flow channel is not less than a set lower limit of 20°, preferably 20° to 40°.

[0033] A further implementation involves a femtosecond laser subtraction module comprising one or more combinations of femtosecond laser ablation, femtosecond laser micro-removal, or femtosecond laser polishing. Femtosecond laser ablation is suitable for coarse removal of large allowance areas, femtosecond laser micro-removal is suitable for fine finishing of areas with dimensional deviations, and femtosecond laser polishing is suitable for improving the surface morphology of the inner wall.

[0034] A further implementation involves detecting the surface morphology or local residual material of the inner wall after femtosecond laser subtraction. If the set conditions are not met, the incident point trajectory, incident attitude, femtosecond laser parameters, and / or the execution frequency of the next occlusion determination are corrected.

[0035] Specifically, when the switch is triggered, femtosecond laser subtraction is performed, prioritizing the subtraction of the currently accessible inner wall area that is predicted to be blocked in the next stage. This removes the reserved allowance, eliminates the additive step effect, and improves the surface morphology of the inner wall. The reserved allowance removal rate must be ≥95%.

[0036] During the material reduction process, both the working distance constraint W and the allowable incident angle range must be satisfied simultaneously. The specific steps are as follows: (1) The incident point is changed in real time along the preset trajectory in the circumferential / axial direction to avoid dead angles on the inner wall caused by a fixed incident point; (2) The incident direction is in Internal adjustments are made to ensure that the angle between the laser beam and the local normal of the inner wall is not less than the set lower limit (e.g., 20°–40°) to guarantee sufficient energy and processing stability. (3) The femtosecond laser parameters can be set according to the material and target effect (e.g., pulse width 20–80 fs, power 8–25 W, scanning speed 80–300 mm / s, scanning spacing 0.01–0.04 mm) to achieve micro-removal / polishing / ablation subtraction.

[0037] The repeat operation module is used to repeatedly trigger the optical path reachability prediction module and the subtractive material module after the optical path reachable geometry is formed, until the entire curved flow channel is manufactured.

[0038] Example 2 This embodiment provides two specific application processes of the system described in Embodiment 1: The following example takes e max =0.003 mm, and the curvature k(m) -1 Converted to radius of curvature ρ (mm): .

[0039] Example 1: High curvature S-shaped flow channel (Inconel 718, flow channel diameter 5 mm).

[0040] Curvature range: The middle curved section is the high curvature region, k=22~30m -1 Both ends are low curvature regions, k=5~8m -1 When the inner wall is discrete, the sampling step size is determined according to equation (1): (1) High curvature region: =33.33~45.45mm, substituting into equation (1) yields ΔS≈0.89~1.04mm, corresponding to an effective incident angle range of ±25°.

[0041] (2) Low curvature region: =125~200 mm, substituting into equation (1) yields ΔS≈1.73~2.19 mm, corresponding to an effective incident angle range of ±60°.

[0042] Therefore, automatic encryption sampling is performed in the high curvature region, 2 layers are continuously added in a single round in the high curvature region, and 8 layers are continuously added in a single round in the low curvature region. During the single round of addition process, the optical path reachability prediction is completed simultaneously. If any inner wall point A(p)=0 is detected according to formula (2), the preparation of the femtosecond laser subtraction process is completed simultaneously. After the current addition part is completed, it is immediately stopped and switched to femtosecond subtraction.

[0043] Additive and subtractive parameters: Additive 190 W, 1000 mm / s, scan interval 0.09 mm; Femtosecond subtractive 50 fs, 12 W, 150 mm / s, scan interval 0.02 mm; circumferential interval of incident points 0.2 mm, incident angle adaptively adjusted within the corresponding range, axial line-by-line scanning.

[0044] Example 2: Medium curvature U-shaped flow channel (Ti6Al4V, flow channel diameter 8mm).

[0045] Curvature range: The bottom is the medium curvature range, k=12~16m -1 The two sides are low curvature regions, k=6~10m -1 When the inner wall is discrete, the sampling step size is determined according to equation (1): (1) Medium curvature region: =62.5~83.33 mm, substituting into equation (1) yields ΔS≈1.22~1.41 mm, corresponding to an effective incident angle range of ±40°.

[0046] (2) Low curvature region: =100~166.67 mm, substituting into equation (1) yields ΔS≈1.55~2.00 mm, corresponding to an effective incident angle range of ±55°.

[0047] Similarly, in the medium curvature region, 4 layers are continuously added in a single cycle, and in the low curvature region, 6 layers are continuously added in a single cycle. During the single-cycle additive process, the optical path reachability prediction is completed simultaneously. If any inner wall point A(p)=0 occurs in the occlusion determination, the femtosecond laser subtraction process preparation is completed simultaneously. After the current additive part is completed, it is immediately stopped and switched to femtosecond subtraction. After the subtraction is completed and the formed geometric model is updated, the next cycle of additive-determination-subtraction continues.

[0048] Additive and subtractive parameters: Additive 185 W, 950 mm / s, scan interval 0.085 mm; Femtosecond subtractive 40 fs, 18 W, 200 mm / s, scan interval 0.03 mm; circumferential interval of incident points 0.25 mm, incident angle adaptively adjusted within the corresponding range, axial line-by-line scanning.

[0049] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A hybrid manufacturing system for curved flow channels based on curvature adaptation and occlusion determination, characterized in that, include: The curvature calculation module is used to obtain a three-dimensional model of the target curved flow channel, discretize the inner wall surface to obtain the inner wall point set and calculate the curvature distribution of the inner wall point set, divide the curvature interval, and synchronously match the effective incident angle range of the femtosecond laser corresponding to each curvature interval. The additive layer thickness setting module is used to set an adaptive additive layer thickness based on the curvature range and the corresponding effective incident angle range of the femtosecond laser. The optical path reachability prediction module is used to perform optical path reachability prediction on the expected generated geometry by using powder bed additive manufacturing technology to add materials layer by layer according to the adaptive additive layer thickness and simultaneously performing optical path reachability prediction during the additive process. Based on the currently formed entity, the module simulates and superimposes the preset additive layer thickness to predict the optical path reachability of the expected generated geometry. The subtractive material triggering module is used to simultaneously complete the preparation of the femtosecond laser subtractive material process when it is predicted that light occlusion will occur during the continuation of additive material. After the current additive part is completed, it immediately stops and triggers the femtosecond laser subtractive material process. The subtraction module is used to change the incident point in real time along the circumference or axis during the femtosecond laser subtraction process, dynamically adjust the incident attitude within the effective incident angle range of the corresponding curvature interval, and perform subtraction on the inner wall area that is currently accessible and will be blocked in the next stage under the constraints of incident angle and working distance, removing the reserved allowance. The repeat operation module is used to repeatedly trigger the optical path reachability prediction module and the subtractive material module after the optical path reachable geometry is formed, until the entire curved flow channel is manufactured.

2. The system according to claim 1, characterized in that, In the curvature calculation module, the process of dividing the curvature interval and matching the effective incident angle range includes: Based on the three-dimensional model of the target curved flow channel, the target curved surface of the inner wall of the flow channel is extracted and discretized based on the NURBS surface discretization algorithm to form a set of inner wall points for occlusion determination; The local curvature k of the inner wall is calculated based on the discrete surface, and the inner wall is divided into at least two curvature intervals according to the curvature magnitude, including a high curvature interval, a medium curvature interval, and a low curvature interval. Among them, the high curvature interval k≥18m -1 Medium curvature range 8m -1 <k<18m -1 Low curvature interval k≤8m -1 .

3. The system according to claim 2, characterized in that, In the additive layer thickness setting module, the layer thickness parameters used during additive manufacturing in each curvature range are related as follows: high curvature layer thickness < medium curvature layer thickness < low curvature layer thickness; the layer thickness in each range is linked and matched with the corresponding effective incident angle range.

4. The system according to claim 2, characterized in that, The process of performing optical path reachability prediction in the optical path reachability prediction module includes: Based on the existing solid, the predicted geometry is generated by simulating the superposition of preset additive layers. A set of pre-defined femtosecond laser incident point locations is used, along with a defined effective incident angle range, working distance / attitude reachability constraint, and spot radius for the corresponding curvature interval. Optical path reachability prediction is performed on the set of inner wall points. When the optical path intersects with the predicted geometry, it is determined to be an obstruction. The optical path reachability prediction judgment also includes ensuring that the spot radius is not truncated, the incident attitude meets the allowable incident angle range of the corresponding curvature interval, and the working distance is within the constraint, and searching for the incident direction within the allowable incident angle range of the corresponding curvature interval.

5. The system according to claim 4, characterized in that, The optical path reachability prediction module is performed synchronously during the additive manufacturing process, and adaptively sets the number of continuous additive layers per round according to the inner wall curvature and the corresponding effective incident angle range; 1 to 3 layers are continuously added in the high curvature range, and 3 to 10 layers are continuously added in the low curvature range; the optical path reachability prediction is completed synchronously during the single round of additive manufacturing.

6. The system according to claim 4, characterized in that, The set of femtosecond laser incident points are distributed in the accessible space above and adjacent to the additive direction, the process window reserved during the additive process, and the preset incident trajectory; The real-time change of the femtosecond laser incident point position includes discrete switching or continuous movement, and the incident attitude is dynamically adjusted within the effective incident angle range of the corresponding curvature interval, subject to the constraints of equipment mechanics and working distance.

7. The system according to claim 4, characterized in that, In the subtractive material module, the femtosecond laser subtractive material includes one or more combinations of femtosecond laser ablation, femtosecond laser micro-removal, or femtosecond laser polishing.

8. The system according to claim 4, characterized in that, After femtosecond laser subtraction, the surface morphology or local residual material of the inner wall is detected. When the set conditions are not met, the incident point trajectory, incident attitude, femtosecond laser parameters and the execution frequency of the next occlusion judgment are corrected.

Images