A milling and grinding integrated turbine blade machining method and device

Abstract

This invention discloses a method and apparatus for machining turbine blades based on integrated milling and grinding. The method constructs a parametric geometric model including the blade, fixture, cutting tool, and grinding wheel. It then partitions the parametric geometric model into feature regions and obtains the optimal combination of milling parameters for each region based on the partitioning results. The grinding parameter range is obtained based on the milled surface morphology corresponding to the optimal milling parameter combination. A comprehensive objective function for grinding is constructed, and the optimal grinding parameter combination is selected from the grinding parameter range based on this objective function. The turbine blade is then machined based on the optimal milling parameter combination and the optimal grinding parameter combination. Seamless integration is achieved by establishing a mapping relationship between key process parameters of milling and grinding. After the milling operation is completed and the tool is automatically switched to a grinding wheel, the optimized grinding parameters are directly called for machining, avoiding abrupt changes in surface quality due to mismatched process parameters and achieving a smooth transition from milling to grinding.

Description

Technical Field

[0001] This invention belongs to the field of precision machining technology, specifically relating to a method and apparatus for machining turbine blades based on integrated milling and grinding. Background Technology

[0002] High-efficiency gas turbines are key equipment for clean energy power generation, and gas turbine blades, as one of the important components of gas turbines, significantly affect the efficiency of the turbines due to their quality. As commonly used blades, the complex curved thin-walled structure and harsh hot corrosion environment of turbine blades place higher demands on the blade manufacturing process. Currently, gas turbine blades are typically manufactured using a "milling + manual polishing" method, which suffers from low processing efficiency, poor processing consistency, and poor surface integrity after processing. This results in high blade production costs and makes it difficult to improve product quality and production efficiency.

[0003] Currently, there are some relatively mature methods in the field of gas turbine blade machining, but none of them have fully solved the above-mentioned problems. Taking the invention patent with patent application number CN105057992B as an example, it introduces a low-rigidity turbine blade milling fixture and its usage method. It improves machining rigidity by casting and fixing the blade with phase change material. However, this method still requires two castings, two clampings, and multiple disassemblies of the cover plate. Each process requires manual intervention, which cannot achieve continuous automated machining. It not only fails to solve the time-consuming problems of multiple clampings, tool changes, and manual intervention in traditional processes. Furthermore, the utility model patent with authorization announcement number CN203266190U introduces a gas turbine blade cooling hole machining fixture. It achieves angle adjustment through a hinge and gauge block structure. However, this fixture relies on manual positioning and has no active compensation mechanism. After multiple angle switching, the hinge gap and gauge block wear will lead to a decrease in positioning accuracy, and it cannot suppress the problem of geometric error accumulation caused by process switching. Furthermore, the invention patent with publication number CN106736700A introduces a fixture device for milling the bottom surface of turbine blade root. It uses replaceable positioning templates to adapt to different blade profiles, but the fixture itself lacks adaptive pressure or vibration suppression functions. Cutting parameters still need to be uniform during processing. For blades with uneven materials or complex structures, rigid clamping can easily lead to local stress overload. This not only fails to solve the problem of machining defects caused by the "one-cut" parameters in traditional processes, but may also cause new surface defects due to vibration. Finally, the invention patent with publication number CN110605543A introduces a new machining process for turbine blades, which uses a vertical fixed single-end clamping for roughing and finishing. However, when roughing and finishing are carried out continuously under the same clamping, there is no process parameter transition mechanism. When switching to finishing, the cutting force and thermal deformation cause abrupt changes in the allowance, resulting in uneven surface texture and even tool marks. It lacks online detection and compensation methods, and completely fails to solve the problem of abrupt changes in surface quality caused by process switching. Summary of the Invention

[0004] This invention provides a method and apparatus for machining turbine blades based on integrated milling and grinding.

[0005] In a first aspect, the present invention provides a method for machining turbine blades based on integrated milling and grinding, the method comprising:

[0006] Construct a parametric geometric model including blades, fixtures, cutting tools, and grinding wheels; perform feature partitioning on the parametric geometric model to obtain a complex surface geometric model; and plan the turbine blade machining path based on the complex surface geometric model.

[0007] Objective functions for different processing regions are constructed respectively; the processing process is simulated for the randomly generated initial milling parameter combinations for each processing region, and the optimal milling parameter combinations for each processing region are selected based on the corresponding objective functions.

[0008] The grinding parameter range is set according to the predicted milled surface morphology corresponding to the optimal combination of milling parameters; a comprehensive objective function for grinding is constructed, and the optimal combination of grinding parameters is selected from the grinding parameter range based on the comprehensive objective function;

[0009] Repeat the above process until the termination condition is met. Use the optimal combination of milling parameters and the optimal combination of grinding parameters obtained in each iteration to actually process the turbine blade through the planned path.

[0010] Preferably, the feature partitioning includes curvature feature partitioning and structural feature partitioning; if the curvature feature partitioning results and the structural feature partitioning results overlap, the optimal combination of milling parameters is selected using the objective function of the structural feature partitioning.

[0011] Preferably, the curvature feature partitioning method is as follows: obtaining the Gaussian curvature at different vertices in the parametric geometric model; and dividing the parametric geometric model into high curvature region, medium curvature region and low curvature region based on the magnitude of the Gaussian curvature.

[0012] Preferably, the objective functions for the high curvature region and the medium curvature region are constructed based on surface roughness, residual stress, and processing time; the objective function for the low curvature region is constructed based on material removal rate and surface roughness.

[0013] As a preferred embodiment, the method for partitioning the structural features is as follows: thin-walled regions are selected based on blade thickness; corner regions are selected based on fillet radius; and transition regions are selected based on curvature changes at different locations.

[0014] Preferably, the objective function of the thin-walled region is constructed based on the workpiece deformation; the objective function of the corner region is constructed based on the tool engagement angle and feed rate; and the objective function of the transition region is constructed based on surface roughness, residual stress, machining time, and curvature gradient modulus.

[0015] Preferably, the grinding parameter range is set based on residual stress, heat-affected zone identification, maximum residual height, and surface roughness.

[0016] Preferably, the integrated objective function is constructed based on the efficiency objective function and the surface integrity objective function.

[0017] Preferably, the milling parameters include cutting speed, feed rate, depth of cut, and tool inclination angle; the grinding parameters include grinding pressure, grinding wheel speed, feed rate, depth of cut, and coolant flow rate.

[0018] Secondly, the present invention provides a turbine blade processing device based on integrated milling and grinding, which is used to perform the above-mentioned turbine blade processing method; the turbine blade processing device includes a model building module, a feature partitioning module, a path building module, a parameter optimization module, and a processing module; the model building module is used to build a geometric model for simulating turbine blade processing; the feature partitioning module is used to partition the built geometric model into features; the path building module is used to build the processing path of the turbine blade; the parameter optimization module is used to obtain the optimal parameters in the turbine blade processing process; the processing module is used to perform milling and grinding processing on the turbine blade to be processed using the optimal parameters.

[0019] The beneficial effects of this invention are:

[0020] 1. This invention integrates milling and grinding processes through a collaborative design. After the workpiece is fixed in the fixture, the machining center automatically completes the rough milling and semi-finish milling operations. Subsequently, the machine tool's automatic tool changer replaces the milling cutter with a grinding wheel, directly performing the finish grinding operation without stopping the machine or moving the workpiece. The entire process is based on a unified positioning reference, and the CNC program automatically switches tools and machining modes, avoiding error accumulation and time waste caused by multiple clamping operations.

[0021] 2. This invention achieves seamless connection by establishing a mapping relationship between key process parameters of milling and grinding. When the milling process is completed and the tool is automatically changed to a grinding wheel, the optimized grinding parameters are directly called for processing, avoiding sudden changes in surface quality caused by mismatch of process parameters and ensuring a smooth transition from milling to grinding.

[0022] 3. This invention employs a complex surface local optimization strategy, implementing differentiated processing for the multi-regional geometric features of turbine blades (such as the blade root, blade crown, and blade body). Based on curvature and structural feature partitioning, high-efficiency processing is achieved using larger feed rates and greater depths of cut for low-curvature regions such as the blade body; for high-curvature regions such as the blade root and blade crown, as well as thin-walled areas, the process automatically switches to small step distances and low feed rates for finishing, while applying strict deformation and temperature constraints. This approach balances efficiency and accuracy, resolving the processing defects caused by the "one-size-fits-all" parameters in traditional processes. Attached Figure Description

[0023] Figure 1 This is the overall flowchart of the present invention.

[0024] Figure 2 This is a flowchart for obtaining the optimal combination of milling parameters in this invention.

[0025] Figure 3 This is a flowchart for obtaining the optimal combination of grinding parameters in this invention. Detailed Implementation

[0026] The present invention will be further described below with reference to the accompanying drawings.

[0027] A method for machining turbine blades based on integrated milling and grinding is disclosed. The turbine blade machining device includes a model building module, a feature partitioning module, a path building module, a parameter optimization module, and a machining module. The model building module constructs the geometric model required for simulating turbine blade machining. The feature partitioning module partitions the constructed geometric model into feature areas. The path building module constructs the machining path for the turbine blade. The parameter optimization module obtains the optimal parameters during the turbine blade machining process. The machining module performs milling and grinding on the turbine blade using the optimal parameters.

[0028] like Figure 1 As shown, the turbine blade processing method includes the following steps:

[0029] Step 1: Determine material properties

[0030] Based on the principles of metal cutting and engineering experience, the initial processing parameter ranges for different materials are constructed, as shown in Table 1.

[0031] Table 1 Recommended Parameters for Common Metallic Materials

[0032]

[0033] Orthogonal experimental design was used to design milling parameter combinations. Within the range of Table 1, the influence of different machining parameters (cutting speed, feed rate, depth of cut) on the machining results (roughness, residual stress) was obtained through experiments. The average results of each factor at different levels were compared, and the machining parameters with the optimal machining results were selected. The influence of each locally optimal machining parameter on the machining results was quantified to assist in the construction of structural feature regions.

[0034] Step 2: Construct a parametric geometric model

[0035] Parametric geometric models of the blade, fixture, cutting tool, and grinding wheel were constructed using CATIA. The complex curvature of the blade was fitted using spline surfaces. Advanced modeling tools, such as spline surfaces, were used in modeling software to directly design and construct a digital model of the blade containing complex curvature. Feature modeling of the complex surface of the blade was performed, transforming it into a mesh composed of countless tiny triangles to approximate the minute surface. Curvature feature partitioning and structural feature partitioning were performed on the parametric geometric model of the blade to obtain the complex surface geometric model. The process of curvature feature partitioning and structural feature partitioning of the parametric geometric model of the blade is as follows:

[0036] 2-1. Curvature Feature Partitioning

[0037] Gaussian curvature at different vertices The expression is:

[0038]

[0039] in, The area of ​​the neighborhood of vertex i is the area of ​​the surface region that "influences" it, which is the sum of 1 / 3 of the area of ​​the triangle containing vertex i. Let be the set of neighboring vertices, representing the set of all vertices directly connected to vertex i; It is the angle between adjacent surfaces at vertex i, that is, the interior angle formed by the triangle containing vertex i at vertex i.

[0040] Based on the Gaussian curvature of each vertex Perform curvature feature partitioning; if the absolute value of the Gaussian curvature at the vertex is not less than 0.05 mm... -1 If the vertex corresponds to a high curvature region, it indicates that the surface morphology of the triangular mesh is complex and the processing precision requirements are stringent; if the absolute value of the Gaussian curvature of the vertex is greater than 0.01mm... -1 And less than 0.05mm -1 If the vertex corresponds to a triangle in the medium curvature region, it indicates that the surface transition of the triangle is relatively smooth, requiring a balance between efficiency and quality; if the absolute value of the Gaussian curvature of the vertex is no greater than 0.01 mm... -1If the triangle mesh corresponding to the vertex belongs to the low curvature region, it means that the triangle mesh belongs to the near-planar region, and efficiency improvement can be emphasized.

[0041] 2-2. Structural Feature Partitioning

[0042] blade thickness Regions less than or equal to 3mm are classified as thin-walled regions (weak stiffness, prone to cutting vibration and deformation); regions with a maximum principal curvature greater than 0.5 are classified as corner regions (toolpath planning needs to avoid overcutting, and residual material needs to be carefully handled); transition regions with continuously changing curvature are defined as transition zones, quantified by a curvature gradient modulus greater than or equal to 0.01; the above-mentioned regions with different structural characteristics are as follows:

[0043]

[0044]

[0045]

[0046] in, These are respectively the thin-walled region, the corner region, and the transition region; For the point set of the surface; For the thickness of the blade, ; Let be the set of points on the opposite surface; a is a point on the surface Γ; b is a point on the OppositeFace surface. The maximum principal curvature; Let be the curvature gradient at point a. .

[0047] In this embodiment, the maximum principal curvature The method for obtaining the equation is as follows: fit a quadratic surface equation onto the vertex and its neighborhood point set. Construct a 2x2 matrix from the coefficients of the quadratic term in the equation, and calculate the eigenvalues ​​of this matrix. Select the eigenvalue with the largest absolute value as the maximum principal curvature at that vertex.

[0048] Step 3: Interference Detection

[0049] The processing path is planned using the equal residual height method to obtain the initial processing path. Then, the octree algorithm is used to process the blade model and the fixture model. The three-dimensional space is recursively divided into eight sub-cubes, which become eight child nodes. The above division is repeated for each child node until the stopping condition is met, and a hierarchical data structure is constructed to efficiently locate potential interference areas.

[0050] The system quickly determines whether the tool and grinding wheel models intersect with each sub-cube, directly filtering out non-interference spatial regions and performing precise minimum distance calculations only on the geometric objects in the intersecting regions. This rapid filtering of non-interference regions improves computational efficiency, while precise minimum distance calculations are performed on triangular facets in potential interference regions. This ensures that the safe distance between the tool and grinding wheel and the fixture is no less than 3mm, and the safe distance from the blades is no less than 1mm (including a 0.5mm safety margin).

[0051] Step 4: Path Optimization

[0052] The collision risk areas between the cutting tool and the fixture and blades are divided into high-risk and low-risk areas. High-risk areas refer to regions where the distance between the cutting tool and obstacles is less than the limit threshold or is in a geometric dead angle, making collisions highly likely. For high-risk areas, the most conservative, high-standard avoidance strategy will be adopted, such as raising the tool to a safe plane of Z=200 mm for overtaking. Simultaneously, the machining area is divided into multiple sub-regions, and a dedicated safe plane is set for each sub-region. When the cutting tool moves between sub-regions, it is preferentially raised to the corresponding safe height Z=200 mm before quickly overtaking, significantly reducing the number of unnecessary tool raisings.

[0053] Low-risk areas refer to areas where the distance between the tool and an obstacle is greater than the limit threshold but less than the safety threshold. A local path fine-tuning strategy is adopted for low-risk areas. An arc detour path is automatically inserted at the boundary of the interference area, and the tool axis vector is adjusted to avoid the protruding parts of the fixture. While ensuring absolute safety, the machining efficiency is maintained to the maximum extent. At the same time, an inclined detour trajectory with a 5° rotation of the machine tool's Z-axis is automatically generated, and the feed rate of the detour segment is reduced to ensure a smooth transition.

[0054] Throughout the machining process, force sensors monitor sudden changes in cutting force in real time. Once a sudden interference is detected that causes the cutting force to exceed a preset threshold (such as 150% of the normal value), an emergency stop protocol is immediately triggered and the tool is controlled to return to the previous safe position. At the same time, the safe zone division and avoidance parameters are dynamically updated based on historical interference data, forming a closed-loop control of "detection-avoidance-verification". Ultimately, a collision-free optimized milling path is generated that maximizes machining efficiency while maintaining continuous cutting, and ensures that the tool and fixture maintain a safe distance of ≥3mm and the tool and workpiece maintain a safe distance of ≥1mm.

[0055] Step 5: Construct a surface morphology prediction model

[0056] like Figure 2As shown, since theoretical geometric models cannot fully reflect the texture caused by dynamic factors in actual machining, numerical simulations of the milling process are performed in the time or frequency domains to predict the vibration trajectory generated during cutting. This vibration trajectory is then superimposed as a dynamic offset onto the complex surface geometric model, resulting in an optimized complex surface geometric model. To further ensure machining quality and efficiency, the finite element analysis method is introduced to simulate and evaluate the machining process for randomly generated initial milling parameter combinations in each region. The specific process is as follows:

[0057] Based on the optimized complex surface geometric model, its local geometric features are calculated using differential geometry, including the Gaussian curvature K of the surface (in mm). -1 The parameters include the surface normal vector n (in units of dimensionless direction components i, j, k) and the surface slope angle θ (in degrees). Based on this, milling path parameters are further integrated, including the tool center point coordinates CL(X, Y, Z) (in mm), the tool axis vector A(i, j, k), and the feed rate V. f (Unit: mm / min), Spindle speed N (unit: rpm), Feed per tooth f z (Unit: mm / z), line spacing / step spacing L c (Unit: mm), Cutting depth a p (Unit: mm) and radial depth of cut a e (Unit: mm). Simultaneously, tool information is integrated, including tool diameter D (unit: mm), number of teeth z, ball end mill radius R (unit: mm), tool runout TIR (unit: μm), and tool flank wear VB (unit: mm). Regarding the materials and physical models, cutting force coefficients are introduced or predicted cutting forces Kc and torque Ke are directly used, and cooling and lubrication conditions are considered as process variables.

[0058] For each machining area (high / medium / low curvature zone, thin-walled zone, corner zone, transition zone), multiple sets of initial milling parameter combinations are randomly generated, including the cutting speed V. c (m / min), feed per tooth f z Cutting depth a p The tool tilt angle θ0 (°) is used. Finite element simulations are performed on each set of initial milling parameter combinations based on the optimized complex surface geometry model to predict surface morphology features and surface physical states. Surface morphology features include surface roughness R. a Maximum height roughness R zThe residual height distribution δ(x,y) and surface physical state include residual stress distribution σ(x,y), microhardness variation ΔHV(x,y), and heat-affected zone (HAZ) markers. The simulation results of each initial milling parameter combination are comprehensively evaluated based on the preset machining objectives. The evaluation methods for different machining areas are as follows:

[0059] (1) Constructing the comprehensive optimization objective function for the high curvature region Its expression is:

[0060]

[0061] in, Maximum surface roughness; The measured stress value; The target value of stress; This is the limit for stress deviation; Total processing time; This represents the maximum total processing time.

[0062] Comprehensive optimization objective function The key constraints are as follows:

[0063]

[0064] in, This refers to the wear amount on the flank face of the cutting tool. This represents the amount of workpiece deformation.

[0065] In the high curvature region, the cutting depth a p Not greater than 0.3mm; feed per tooth f z No greater than 0.05 mm / z.

[0066] (2) Constructing the comprehensive optimization objective function for the low curvature region Its expression is:

[0067]

[0068] in, Material removal rate; This represents the maximum material removal rate.

[0069] Comprehensive optimization objective function The key constraints are as follows:

[0070]

[0071] In the low curvature region, the cutting depth a p Not greater than 2mm; feed per tooth f z No greater than 0.2 mm / z.

[0072] (3) Construct the comprehensive optimization objective function for the medium curvature region Its expression is:

[0073]

[0074]

[0075]

[0076]

[0077] in, This represents the dynamic weighting coefficient for the surface quality term; For the dynamic weighting coefficients of the stress control term; This represents the dynamic weighting coefficient for the processing efficiency term; The curvature at the current point; This is the lower limit of the curvature in the intermediate curvature region; This represents the upper limit of curvature in the mid-curvature region.

[0078] Comprehensive optimization objective function The key constraints are as follows:

[0079]

[0080] In the medium curvature region, the cutting depth a p Not greater than 1mm; feed per tooth f z No greater than 0.1 mm / z.

[0081] (4) Construct the comprehensive optimization objective function for the thin-walled region Its expression is:

[0082]

[0083] in, The objective function value related to curvature; This represents the maximum deformation.

[0084] Comprehensive optimization objective function The key constraints are as follows:

[0085]

[0086] in, This refers to the actual cutting force. This represents the maximum cutting force.

[0087] In the thin-walled region, the cutting depth a p Not greater than 0.5mm; feed per tooth f z Not greater than 0.08 mm / z.

[0088] (5) Construct the comprehensive optimization objective function for the corner area Its expression is:

[0089]

[0090] in, It is the tool engagement angle; This is the standard feed rate.

[0091] Comprehensive optimization objective function The key constraints are as follows:

[0092]

[0093] in, Radial cutting depth; The diameter is the cutting tool.

[0094] In the corner area, the cutting depth a p No more than 0.4mm; in the finishing process, the tool's feed step distance is halved and the number of feeds is doubled.

[0095] (6) Construct the comprehensive optimization objective function for the transition zone Its expression is:

[0096]

[0097] in, For curvature gradient magnitude, For the maximum curvature gradient magnitude, These are the weighting coefficients.

[0098] The milling parameter combinations that satisfy the constraints and have the best optimization objectives are selected as the input benchmark for subsequent grinding processes. Since the structural partitioning results cannot completely cover the model, when regional features overlap, the most stringent constraint combination is adopted according to the priority order of thin-walled area / corner area / transition area > high curvature area > medium curvature area > low curvature area. The parameter selection is appropriately modified based on the objective of the highest priority area, and finally, the optimized process parameter set with differentiated configuration for each region is output, providing an accurate parameter benchmark for actual machining execution.

[0099] Step Six, as Figure 3 As shown, the predicted milled surface morphology (especially the residual height distribution) is a key input to the grinding process. The inputs are categorized into the following four types:

[0100] If the residual stress σ > 0 (tensile stress) or the heat-affected zone is marked as TRUE, then an integrity repair decision is executed, and the values ​​of each parameter are set as shown in Table 2.

[0101] Table 2 Integrity Repair Decision Parameter Library

[0102]

[0103] Otherwise, a combined strategy of margin and roughness is implemented, the specific process of which is as follows:

[0104] If the maximum residual height R t > 15μm and surface roughness R a If the material size is > 3.2 μm, an efficient removal decision is made, which can quickly remove the material and ensure efficiency, as shown in Table 3.

[0105] Table 3 Efficient Removal of Decision Parameter Library

[0106] Grinding pressure P (kPa) 200~350 Higher pressure improves material removal rate Grinding wheel speed N (rpm) 5000~8000 Standard speed, balancing efficiency and mass feed rate Vf (mm / min) 1000~2000 Higher feed rate improves machining efficiency Grinding depth ap (mm) 0.02~0.05 Larger cutting depth, efficient material removal Coolant flow rate Q (L / min) 6~10 Standard cooling, controlled thermal impact

[0107] If the maximum residual height R t < 5μm and surface roughness R a If the thickness is less than 1.6 μm, a stable finishing decision is made to efficiently achieve the final surface finish while retaining compressive stress, as shown in Table 4.

[0108] Table 4. Stable Refining Decision Parameter Library

[0109]

[0110] If the maximum residual height and surface roughness are otherwise specified, an adaptive balancing decision is executed. This adaptive balancing decision seeks the optimal balance between efficiency and quality. The system generates adaptive balancing decision parameters by linearly interpolating between efficient removal decision parameters and stable finishing decision parameters based on the specific values ​​of the maximum residual height and surface roughness. Its expression is:

[0111]

[0112] in, and These are respectively: efficient removal of decision parameters and stable smoothing of decision parameters; These are the weighting coefficients for adaptive equilibrium decision-making.

[0113] Step 7: Obtain the core output parameters, including surface roughness R, through finite element analysis based on the global multi-objective parameter cooperative function. a The parameters include (μm), residual stress distribution σ(x,y), microhardness variation ΔHV (HV), and thermodynamic response including temperature field distribution T(x,y), cutting force Fc, and torque Tc. Process parameters include material removal rate (MRR) and machining time T. total Construct the comprehensive objective function F. totalIt is represented as:

[0114]

[0115] Among them, F total The value of the comprehensive objective function; w1 and w2 represent weight coefficients, w1 + w2 = 1; Let the efficiency objective function be... The objective function is surface integrity. Indicates the efficiency benchmark value; This represents the baseline value for surface integrity.

[0116] In this embodiment, the typical values ​​of the weighting coefficients w1 and w2 are: w1 = 0.7 and w2 = 0.3 for roughing, w1 = 0.5 and w2 = 0.5 for semi-finishing, and w1 = 0.3 and w2 = 0.7 for finishing.

[0117] (1) Efficiency objective function:

[0118]

[0119] in, This is a global processing efficiency indicator; The total material removal rate, ; The material removal rate during milling; The material removal rate is the grinding material removal rate.

[0120] (2) Surface integrity objective function:

[0121]

[0122] Among them, S global Represents the global surface integrity index; λ i F represents the weight coefficient for region i; region,i Represents the local objective function value of region i; α represents the penalty coefficient; Penalty constraints The expression for the constraint violation penalty term is:

[0123]

[0124] Where, β j To constrain the penalty weight of j; Represents the actual value of constraint j; g j , max This represents the maximum allowed value for constraint j.

[0125] In this embodiment, the regional weight coefficient λ iTypical values ​​are: 0.25-0.35 for high curvature region, 0.20-0.30 for thin-walled region, 0.15-0.25 for corner region, 0.10-0.20 for medium curvature region, 0.05-0.15 for low curvature region, and 0.15-0.25 for transition region.

[0126] Step 8: Construct the joint matrix of grinding parameters Optimization Iteration: Search the parameter matrix X for solutions that simultaneously satisfy three constraints (physical constraints: machining temperature, tool wear, and workpiece deformation must not exceed preset safety values; equipment constraints: spindle speed and polishing pressure must not exceed machine tool rated values; geometric constraints: no collisions and tool reachable positions). Multiple solutions satisfying the conditions and achieving the optimization objective are identified. Inappropriate solutions are eliminated. Finally, from all parameter combinations satisfying the constraints, the one that maximizes the comprehensive objective function value F is selected. total The optimal parameter set is used to retain the solution that better balances efficiency and surface quality, and to expand the solution set to generate a final solution set.

[0127] Step 9: Repeat steps 2 through 8 until the global processing efficiency index output by the multi-objective parameter cooperative function is obtained. and global surface integrity index At the same time, it reaches the finishing stage, that is Then, the blade is machined based on the optimal combination of milling and grinding parameters obtained in each iteration; where MRR max This represents the theoretical maximum material removal rate based on equipment capacity and material properties.

[0128] In this embodiment, the semi-finishing stage is as follows: The roughing stage is as follows: .

[0129] After reaching the finishing stage, the final combination of process parameters can achieve the optimal balance between processing efficiency and surface quality, thereby ensuring that the turbine blade can complete the entire process from roughing to finishing in a single clamping and reach high quality standards.

Claims

1. A method for machining turbine blades based on integrated milling and grinding, characterized in that: The method includes: Construct a parametric geometric model including blades, fixtures, cutting tools, and grinding wheels; perform feature partitioning on the parametric geometric model to obtain a complex surface geometric model; and plan the turbine blade machining path based on the complex surface geometric model. Objective functions for different processing regions are constructed respectively; the processing process is simulated for the randomly generated initial milling parameter combinations for each processing region, and the optimal milling parameter combinations for each processing region are selected based on the corresponding objective functions. The grinding parameter range is set according to the predicted milled surface morphology corresponding to the optimal combination of milling parameters; a comprehensive objective function for grinding is constructed, and the optimal combination of grinding parameters is selected from the grinding parameter range based on the comprehensive objective function; Repeat the above process until the termination condition is met. Use the optimal combination of milling parameters and the optimal combination of grinding parameters obtained in each iteration to actually process the turbine blade through the planned path.

2. The turbine blade machining method based on integrated milling and grinding according to claim 1, characterized in that: The feature partitions include curvature feature partitions and structural feature partitions; If the curvature feature partitioning results overlap with the structural feature partitioning results, the optimal combination of milling parameters is selected using the objective function of the structural feature partitioning.

3. The turbine blade machining method based on integrated milling and grinding according to claim 2, characterized in that: The curvature feature partitioning method is as follows: obtain the Gaussian curvature at different vertices in the parametric geometric model; and divide the parametric geometric model into high curvature region, medium curvature region and low curvature region based on the magnitude of the Gaussian curvature.

4. The turbine blade machining method based on integrated milling and grinding according to claim 3, characterized in that: The objective functions for the high curvature and medium curvature regions are constructed based on surface roughness, residual stress, and processing time; the objective function for the low curvature region is constructed based on material removal rate and surface roughness.

5. The turbine blade machining method based on integrated milling and grinding according to claim 2, characterized in that: The method for partitioning the structural features is as follows: thin-walled regions are selected based on blade thickness; corner regions are selected based on fillet radius; and transition regions are selected based on curvature changes at different locations.

6. The turbine blade machining method based on integrated milling and grinding according to claim 5, characterized in that: The objective function for the thin-walled region is constructed based on the workpiece deformation; the objective function for the corner region is constructed based on the tool engagement angle and feed rate; and the objective function for the transition region is constructed based on surface roughness, residual stress, machining time, and curvature gradient modulus.

7. The turbine blade machining method based on integrated milling and grinding according to claim 1, characterized in that: The grinding parameter range is set based on residual stress, heat-affected zone identification, maximum residual height, and surface roughness.

8. The turbine blade machining method based on integrated milling and grinding according to claim 1, characterized in that: The integrated objective function is constructed based on the efficiency objective function and the surface integrity objective function.

9. A method for machining turbine blades based on integrated milling and grinding according to claim 1, characterized in that: The milling parameters include cutting speed, feed rate, depth of cut, and tool inclination angle; the grinding parameters include grinding pressure, grinding wheel speed, feed rate, depth of cut, and coolant flow rate.

10. A turbine blade processing device based on integrated milling and grinding, characterized in that: A turbine blade machining method based on milling and grinding integration as described in claim 1 is used to execute the turbine blade machining device, which includes a model building module, a feature partitioning module, a path building module, a parameter optimization module, and a machining module. The model building module is used to build a geometric model for simulating turbine blade machining. The feature partitioning module is used to partition the built geometric model into features. The path building module is used to build a machining path for the turbine blade. The parameter optimization module is used to obtain the optimal parameters during the turbine blade machining process. The machining module is used to perform milling and grinding on the turbine blade to be machined using the optimal parameters.

Images