Optimization method of digital precision assembly of disk and blades

Abstract

The present invention discloses a method for optimizing the digital precision assembly of a wheel disc and blades, which realizes efficient and accurate modeling of the physical contour points of the wheel disc and each blade, determines the optimal assembly posture of each blade by accurately weighing and calculating the center of mass of the actual wheel disc and blades, and calculating the contact stress at N nodes on both sides of the wheel disc slot and the blade root tooth, calculates the total unbalance amount of the wheel disc and all blades, optimizes the arrangement of the serial numbers of each blade during assembly, obtains the total unbalance amount prediction of the optimized wheel disc and all blades, and thus realizes the sequential assembly of the blades; determines the grinding area, solves the optimal grinding amount of the blade position in the grinding area, grinds all the actual position points at the tail end of all blades in the grinding area, and performs dynamic simulation analysis on the wheel disc blade assembly simulation model after grinding, and obtains the dynamic balance performance of the wheel disc blade assembly simulation model after grinding. The present invention realizes digital precision verification of the wheel disc blade assembly performance, and finally realizes digital precision assembly of the wheel disc and blades.

Description

Technical Field

[0001] The invention relates to the technical field of digital precision assembly of turbine machinery rotors, and in particular to a modeling and optimization method for digital precision assembly of a wheel disc and blades. Background Art

[0002] Turbine machinery is a kind of dynamic fluid machinery with blades. The characteristic of turbine machinery is that the rotor equipped with blades rotates at high speed. When the fluid (gas or liquid) flows through the channels between the blades, the blades and the fluid interact with each other to achieve energy conversion.

[0003] As the core component of turbine machinery, the assembly performance of the rotor system directly affects the stability, reliability and life of the whole machine. Among them, the assembly of the disc and blades is crucial. However, during the grinding and assembly process, the quality changes, contour size errors, and assembly surface morphology deviations of the parts make it difficult to accurately predict the assembly performance. Therefore, it is urgent to design a precise modeling and optimization method for the digital precision assembly process of discs and blades to achieve precise modeling that integrates the calculation, modeling, measurement, verification, correction and dynamic balancing performance prediction of the key characteristic parameters of the parts. Summary of the invention

[0004] In view of the above-mentioned defects of the prior art, the object of the present invention is to provide a precise modeling and optimization method for the digital precision assembly process of impellers and blades, which can realize the precise modeling that integrates the calculation, modeling, measurement, verification, correction and dynamic balancing performance prediction of key characteristic parameters of components.

[0005] The specific technical solution adopted by the present invention to solve the above technical problems is:

[0006] The method for optimizing the digital precision assembly of a wheel disc and blades of the present invention comprises the following steps:

[0007] S1 performs 3D contour scanning on the wheel disc and blades to be assembled to obtain a preliminary solid scan model, compares the preliminary solid scan model with the design model, and obtains the features and corresponding feature points in the preliminary solid scan model that do not meet the size and tolerance requirements.

[0008] S2 Use a three-dimensional coordinate measuring machine to measure and correct the key feature points of the wheel disc and each blade to obtain the solid contour model of the wheel disc and each blade. The key feature points include the surface point clouds of the feature points obtained in step S1 and the wheel disc groove part of the wheel disc and the root tooth part of each blade.

[0009] S3 calculates the density and center of mass of the disk and each blade.

[0010] S4 Based on the solid contour models of the impeller and each blade, the final assembly posture of each blade is optimized, and the contact stress of N nodes on the contact surface of each impeller slot and blade root tooth on both sides and the gap between each impeller slot and blade root tooth under the final assembly posture of each blade are predicted.

[0011] S5 calculates the total imbalance between the wheel disc and all blades to optimize the arrangement of the blade numbers during assembly, and obtains the total imbalance prediction between the wheel disc and all blades after optimization, thereby realizing the sequential assembly of the blades. Then, the grinding area is determined, the optimal grinding amount of the blade position in the grinding area is solved, and the actual position points of the tail ends of all blades in the grinding area are ground.

[0012] Preferably, the specific implementation process of step S1 is as follows:

[0013] S11 Scan the three-dimensional profiles of the wheel disc and each blade in turn with an optical scanner to obtain surface point cloud data of the wheel disc and each blade. Before scanning, sort the blades and mark their serial numbers. The wheel disc and each blade establish a coordinate system with the rotation center as the origin.

[0014] S12 pre-processes and reverse models the surface point cloud data of the disk and each blade to obtain a preliminary solid scanning model of the disk and each blade.

[0015] S13 compares the design models of the wheel disc and each blade with the preliminary solid scan model, and obtains various features and corresponding feature points in the preliminary solid scan model that do not meet the size and tolerance requirements according to the size and tolerance requirements of the processing drawings.

[0016] More preferably, the preprocessing is to perform multi-view data stitching, denoising, and point cloud data triangular meshing on the surface point cloud data; the reverse modeling is to re-scan the missing data and then perform surface construction to obtain a preliminary solid scanning model of the wheel disc and each blade.

[0017] Preferably, the specific implementation process of step S2 is as follows:

[0018] S21 uses a three-dimensional coordinate measuring machine to measure the key feature points on the wheel disc and each blade, and obtains discrete point cloud data of the key feature points on the wheel disc and each blade. During measurement, the wheel disc and each blade establish a coordinate system with the rotation center position as the origin.

[0019] S22 uses Gaussian filtering to filter the discrete point cloud data of key feature points on the wheel and each blade.

[0020] S23: Using the filtered discrete point cloud data, the preliminary solid scan models of the wheel disc and blades obtained in step S1 are corrected to obtain solid contour models of the wheel disc and each blade.

[0021] Preferably, the specific implementation process of step S3 is as follows:

[0022] S31 weighs the wheel disc and each blade to be assembled in turn, and calculates the volume of the solid contour model of the wheel disc and each blade in the CAD software. Then, the density of the wheel disc and the blade is calculated based on the volume and weight information and assigned to the wheel disc and the blade.

[0023] S32 Calculate the center of mass of the disk and each blade and define it in the CAD software.

[0024] Preferably, the specific implementation process of step S4 is as follows:

[0025] S41 imports the solid contour models of the disk and each blade into the finite element simulation software, assigns material properties to the disk and each blade, preliminarily assembles each blade to the disk in turn, creates analysis steps, defines boundary conditions, and defines loads. Then, meshes are divided for the disk and each blade. Finally, N nodes are set on the contact surfaces on both sides of the disk slot and the blade root tooth. When the simulation is completed, the contact stress at the node is derived, and the standard deviations S1 and S2 of the contact stress at the N nodes on both sides are calculated respectively.

[0026] S42 sets the number of particles and the maximum number of iterations. When the blade is initially assembled on the wheel, the angle between the line connecting the wheel and the blade center of mass and the x-axis is , and set the angle between the line connecting the center of mass of the wheel and the blade and the x-axis to -5°~ +5°, and the fitness value is the difference between the standard deviations S1 and S2. Then, the angle between the line connecting the center of mass of the wheel and the blade and the x-axis is calculated. -5°~ The fitness values ​​of various positions vary between +5°, and the optimal position with the smallest fitness value is found.

[0027] S43 According to the optimal position of each blade obtained in step S42, the final assembly posture of each blade is determined, and the contact stress of N nodes on the contact surfaces on both sides of the corresponding wheel groove and the blade root tooth is output according to the simulation result of the final assembly posture of each blade. The simulation result model is cut in post-processing to obtain the spacing between the feature points of each corresponding position on the wheel groove and the blade root tooth.

[0028] Preferably, the specific implementation process of step S5 is as follows:

[0029] S51 calculates the unbalance of the wheel disc and the total unbalance of all blades.

[0030] S52 Calculates the total unbalance of the disk and all blades and the total unbalance Angle with x-axis .

[0031] S53 indicates the distribution of each blade on the rotor by arranging the serial numbers of each blade, sets the initial population, the number of iterations, and sets the fitness value to The termination condition is set to the total imbalance between the disk and all blades The actual production requirement is less than 1 / 8 of the limit value. The particle swarm algorithm is used to optimize the arrangement of the blade numbers during assembly. During the optimization process, each blade is reassembled to the disk in turn for each arrangement of the blade numbers during assembly, so that the disk and the solid contour model of each blade are updated, and then steps S4, S51 and S52 are executed to obtain the total imbalance between the disk and all blades. The total imbalance between the optimized disk and all blades is recorded as , total unbalance The angle with the x-axis is .

[0032] S54: The total unbalance value after optimization obtained in step S53 The vector direction is taken as the reference, and the blade tail position within 30° clockwise or counterclockwise is set as the grinding area.

[0033] S55 realizes multi-objective optimization through the interaction of genetic algorithm and grinding simulation, solves the optimal grinding amount of blade position in the grinding area, and grinds all actual position points of the tail end of all blades in the grinding area.

[0034] More preferably, the specific implementation process of step S51 is as follows: according to the wheel disc and the center of mass of each blade when each blade is finally assembled, the distance L and the angle between the wheel disc and the center of mass of each blade deviating from the center of the wheel disc are calculated. , then the unbalance M0 of the wheel and each blade is determined by the vector from the wheel center to the wheel center of mass of each blade. The components of the unbalance M0 in the x and y directions are , Calculated according to the following formula:

[0035]

[0036]

[0037] Where M is the mass of the disk or blade.

[0038] The components of the wheel's unbalance in the x and y directions are recorded as , The components of the total unbalance of all blades in the x and y directions are recorded as , The components of the total unbalance of the blade in the x and y directions , It is the cumulative sum of the unbalanced components of each blade in the x and y directions.

[0039] More preferably, the specific implementation process of step S55 is as follows: first, a total of K actual position points are taken at the tail ends of all blades in the grinding area, and the grinding amount of each actual position point is assumed to be equal. ( ), the grinding direction is along the blade tail end at point The normal of the corresponding grinding position point is set to ( ), then the grinding amount is:

[0040]

[0041] Then, the centroid of the blade after grinding according to the grinding amount is recalculated, and then the total imbalance of the wheel disc and all blades after grinding is calculated according to step S51 and step S52, which is recorded as .

[0042] The optimization target of the genetic algorithm is the difference between the standard deviation S1 and the standard deviation S2 and the total imbalance Minimum, set the initial population and the number of iterations, the optimization variable is the grinding amount of all actual position points of the tail ends of all blades in the grinding area, and after optimization, the optimal grinding amount of all actual position points of the tail ends of all blades in the grinding area is obtained, and then all actual position points of the tail ends of all blades in the grinding area are ground according to the optimal grinding amount.

[0043] Preferably, the step S5 is followed by the following step: performing dynamic simulation analysis on the wheel disc blade assembly simulation model after grinding to obtain the dynamic balance performance of the wheel disc blade assembly simulation model after grinding.

[0044] The present invention has the following beneficial effects:

[0045] 1. The present invention realizes precise modeling of the entity contours of the wheel disc and each blade through rapid scanning by a three-dimensional optical scanner combined with efficient and accurate measurement of key feature points by a three-dimensional coordinate measuring machine; then, the optimal assembly posture of each blade is determined by precise weighing of the actual wheel disc and blades and calculation of the center of mass of the wheel disc and each blade in the entity contour model, as well as calculation of contact stress at N nodes on both sides of the wheel disc groove and the blade root tooth, and then the total unbalance amount of the wheel disc and all blades is precisely calculated to optimize the arrangement of the serial numbers of each blade during assembly, and the total unbalance amount of the optimized wheel disc and all blades is predicted, thereby realizing sequential assembly of the blades, and then determining the grinding area, solving the optimal grinding amount of the blade position in the grinding area, grinding all the actual position points of the tail ends of all blades in the grinding area, and finally performing dynamic simulation analysis on the wheel disc blade assembly simulation model after grinding, obtaining the dynamic balancing performance of the wheel disc blade assembly simulation model after grinding, realizing digital precision verification of the wheel disc blade assembly performance, and finally realizing digital precision assembly of the wheel disc and blades.

[0046] 2. The present invention realizes accurate prediction of blade assembly posture, contact stress and clearance by means of interaction between particle swarm algorithm and assembly simulation.

[0047] 3. The present invention optimizes the assembly sequence of blades through particle swarm algorithm, thereby realizing sequential assembly of blades; and realizes accurate prediction of localized and quantitative grinding parameters of blades through the interaction of genetic algorithm and grinding simulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a flow chart of the present invention.

[0049] Figure 2 It is a schematic diagram of the wheel of a turbine machine.

[0050] Figure 3 It is a schematic diagram of the blades of a turbine machine.

[0051] Figure 4 It is a schematic diagram of the present invention after optimizing the arrangement of serial numbers of blades during assembly through the particle swarm algorithm. DETAILED DESCRIPTION

[0052] The present invention will be further described below in conjunction with the accompanying drawings.

[0053] like Figure 1 As shown in the figure, the optimization method of digital precision assembly of the wheel disc and the blade is as follows:

[0054] S1 Perform 3D contour scanning on the wheel disc and blades to be assembled, compare the preliminary entity scan model with the design model, and obtain the features and corresponding feature points in the preliminary entity scan model that do not meet the size and tolerance requirements. The specific implementation process is as follows:

[0055] S11 uses a high-precision optical scanner to scan Figure 2 The roulette wheel shown and Figure 3 The 3D profile of each blade is scanned to quickly obtain the surface point cloud data of the wheel disc and each blade. Before scanning, the blades need to be sorted and marked with serial numbers. The wheel disc and each blade establish a coordinate system with the rotation center as the origin. For the darker parts on the wheel disc and each blade, a matte white developer can be used to cover the surface before scanning to improve the scanning effect.

[0056] S12 pre-processes and reverse models the surface point cloud data of the disk and each blade to quickly obtain the preliminary solid scanning models of the disk and each blade.

[0057] Specifically, the preprocessing is to perform multi-view data stitching, denoising, and triangular meshing of the surface point cloud data; the reverse modeling is to re-scan the missing data and then construct the surface to obtain the preliminary solid scanning models of the wheel disc and each blade.

[0058] S13: Compare the design models of the wheel disc and each blade with the preliminary solid scan model, and obtain the features and corresponding feature points in the preliminary solid scan model that do not meet the size and tolerance requirements according to the size and tolerance requirements of the processing drawings. This step is performed in the CAD software.

[0059] S2 Use a three-dimensional coordinate measuring machine to precisely measure and correct the key feature points of the wheel disc and each blade to obtain the solid contour model of the wheel disc and each blade. Specifically, the key feature points include the various feature points obtained in step S13 and the surface point clouds of the wheel disc grooves and blade root teeth on the wheel disc and each blade that affect the assembly accuracy. The specific implementation process is as follows:

[0060] S21 uses a three-dimensional coordinate measuring machine to measure the key feature points on the wheel disc and each blade, and obtains high-precision discrete point cloud data of the key feature points on the wheel disc and each blade. During measurement, the wheel disc and each blade also establish a coordinate system with the rotation center position as the origin.

[0061] S22 uses Gaussian filtering to filter the discrete point cloud data of key feature points on the wheel and each blade.

[0062] S23 uses the filtered high-precision discrete point cloud data to correct the preliminary solid scan models of the wheel disc and blades obtained in S1 to obtain solid contour models of the wheel disc and each blade.

[0063] S3 calculates the density and center of mass of the disk and each blade. The specific implementation process is as follows:

[0064] S31 weighs the wheel disc and each blade to be assembled in turn, and calculates the volume of the solid contour model of the wheel disc and each blade in the CAD software. Then, the density of the wheel disc and the blade is calculated based on the volume and weight information and assigned to the wheel disc and the blade. Specifically, a high-precision digital weighing scale is used when weighing the wheel disc and each blade in turn.

[0065] S32 Calculate the center of mass of the disk and each blade in the solid contour model of the disk and each blade and define it in the CAD software. Center of mass coordinates The calculation formula is as follows:

[0066]

[0067]

[0068]

[0069] Where N is the number of feature points of the disk or blade surface point cloud data, , , are the x, y, and z coordinates of the i-th feature point.

[0070] S4 optimizes the final assembly posture of each blade based on the solid contour model of the disk and each blade, and predicts the contact stress of N nodes on the contact surface of each disk slot and blade root tooth on both sides of each blade in the final assembly posture, as well as the gap between each disk slot and blade root tooth. The specific implementation process is as follows:

[0071] S41 Based on the solid contour models of the disk and each blade, a disk-blade assembly simulation model is established to calculate the standard deviations S1 and S2 of the contact stress at N nodes on both sides of the disk slot and the blade root tooth: First, the solid contour models of the disk and each blade are imported into the abaqus finite element simulation software, and the corresponding material properties (including the density of the disk and the blade) are assigned to the disk and each blade, where the material behavior is set to elastic. Then, each blade is preliminarily assembled to the disk in turn, and the analysis step, boundary conditions, and loads are defined. When defining the boundary conditions, the disk is set to be fixed. When defining the load, the load is set to gravity, and the gravity acceleration is -9800mm / s². Then, the disk and each blade are meshed separately. In order to improve the accuracy of the simulation results, when meshing the disk or blade, it is mainly divided into two areas, namely the assembly part of the blade root tooth and the disk slot of the disk, and the remaining part. Finally, N nodes are set on the contact surfaces on both sides of the wheel groove and the blade root tooth. When the simulation is completed, the contact stress at the nodes is derived, and the standard deviations S1 and S2 of the contact stress at the N nodes on both sides are calculated respectively.

[0072] S42 uses particle swarm algorithm to interact with assembly simulation to optimize the assembly posture of each blade. The particle swarm algorithm has a strong global search capability and supports parallel computing. First, determine the algorithm parameters: the number of particles is set to 100, the maximum number of iterations is set to 100, and the angle between the line connecting the wheel and the blade center of mass and the x-axis when the blade is initially assembled on the wheel is set to , and set the angle between the line connecting the center of mass of the wheel and the blade and the x-axis to -5°~ +5°, and the fitness value is the difference between the standard deviations S1 and S2. Then, calculate the angle between the line connecting the center of mass of the wheel and the blade and the x-axis. -5°~ +5°) and find the optimal position with the smallest fitness value.

[0073] S43 According to the optimal position of each blade obtained in step S42, the final assembly posture of each blade can be determined and adjusted. According to the simulation results of the final assembly posture of each blade, the contact stress of N nodes on the contact surfaces on both sides of the corresponding wheel disc groove and the blade root tooth can be output. The simulation result model is cut in post-processing to obtain the gap between the corresponding wheel disc groove and the blade root tooth (the distance between the feature points at the corresponding positions on the wheel disc groove and the blade root tooth).

[0074] S5 precisely calculates the total imbalance between the wheel disc and all blades, optimizes the arrangement of the serial numbers of each blade during assembly, and predicts the total imbalance between the wheel disc and all blades after optimization, thereby realizing the sequential assembly of the blades. Then, the grinding area is determined, the optimal grinding amount of the blade position in the grinding area is solved, and the actual position points of the tail ends of all blades in the grinding area are ground. The specific implementation process is as follows:

[0075] S51 calculates the unbalance of the wheel disc and the total unbalance of all blades. According to the wheel disc and the center of mass of each blade when the blade is finally assembled, the distance L and the angle between the wheel disc and the center of mass of each blade deviating from the center of the wheel disc can be calculated. , then the unbalance M0 of the wheel and each blade can be determined by the vector from the wheel center to the wheel center of mass of each blade. The components of the unbalance M0 in the x and y directions are , Calculated according to the following formula:

[0076]

[0077]

[0078] Where M is the mass of the disk or blade.

[0079] The components of the wheel's unbalance in the x and y directions are recorded as , The components of the total unbalance of all blades in the x and y directions are recorded as , The components of the total unbalance of the blade in the x and y directions , That is, it is the cumulative sum of the unbalanced components of each blade in the x and y directions.

[0080] S52 Calculates the total unbalance of the disk and all blades and the total unbalance Angle with x-axis , the formula is as follows:

[0081]

[0082]

[0083]

[0084]

[0085] in, , The total unbalance Components in the x and y directions.

[0086] S53 uses particle swarm algorithm to optimize the arrangement of the serial numbers of each blade during assembly, thereby realizing the sequential assembly of the blades. Specifically, the serial number arrangement of each blade represents the distribution of each blade on the rotor, and then sets the initial population to 100, the number of iterations to 1000, and the fitness value to The termination condition is set to the total imbalance between the disk and all blades The limit value is less than 1 / 8 of the actual production requirement. The particle swarm algorithm is used to optimize the arrangement of the blade numbers during assembly. During the optimization process, each blade is reassembled to the disk in turn for each arrangement of the blade numbers during assembly, so as to update the solid contour models of the disk and each blade. Figure 4 As shown, then step S4, step S51 and step S52 are performed to obtain the total imbalance between the wheel and all blades. The total imbalance between the optimized disk and all blades is recorded as , total unbalance The angle with the x-axis is .

[0087] S54 The total unbalance value after optimization obtained in step S53 Determine the grinding area. The blade position within 30° clockwise or counterclockwise is set as the grinding area based on the vector direction of the blade. In order to facilitate the determination of the grinding amount, the blade grinding area is set at the blade tail end.

[0088] S55 realizes multi-objective optimization through the interaction of genetic algorithm and grinding simulation, solves the optimal grinding amount of blade position in the grinding area, and grinds all actual position points of the tail end of all blades in the grinding area. Since the genetic algorithm converges quickly, the genetic algorithm is used to achieve this optimization. First, a total of K actual position points are taken at the tail end of all blades in the grinding area, and the grinding amount of each actual position point is set to be equal. The actual position point is set to ( ), the grinding direction is along the blade tail end at point The normal of the corresponding grinding position point is set to ( ), then the grinding amount is:

[0089]

[0090] Then, the centroid of the blade after grinding according to the grinding amount is recalculated in step S32, and the total imbalance of the wheel disc and all blades after grinding is calculated according to steps S51 and S52, which is recorded as .

[0091] The optimization target of the genetic algorithm is the difference between the standard deviation S1 and the standard deviation S2 and the total imbalance Minimum, so the fitness value is set to the difference between the standard deviation S1 and the standard deviation S2 and the total imbalance , the initial population is set to 100, the number of iterations is set to 500, the optimization variable is the grinding amount of all actual position points of the tail end of all blades in the grinding area, and the optimal grinding amount of all actual position points of the tail end of all blades in the grinding area is obtained after optimization, and then all actual position points of the tail end of all blades in the grinding area are ground according to the optimal grinding amount.

[0092] S6 performs dynamic simulation analysis on the wheel blade assembly simulation model after grinding, obtains the dynamic balance performance of the wheel blade assembly simulation model after grinding, and realizes digital precision verification of the wheel blade assembly performance.

[0093] After optimization by the present invention, the wheel disc and blades to be assembled are assembled according to the optimized blade serial number arrangement, and all actual position points of the blade tail ends in the grinding area are ground according to the optimal grinding amount, so that the dynamic balance performance of the wheel disc blade assembly after assembly is greatly improved.

Claims

1. A method for optimizing the digital precision assembly of a wheel disc and blades, characterized in that: The method comprises the following steps: S1. Perform 3D contour scanning on the wheel disc and blades to be assembled to obtain a preliminary solid scan model. Compare the preliminary solid scan model with the design model to obtain the features and corresponding feature points in the preliminary solid scan model that do not meet the size and tolerance requirements. S2 Using a three-dimensional coordinate measuring machine to measure and correct the key feature points of the wheel disc and each blade, and obtain a solid contour model of the wheel disc and each blade; the key feature points include the feature points obtained in step S1 and the surface point clouds of the wheel disc groove part of the wheel disc and the blade root tooth part of each blade; S3 calculates the density and center of mass of the disk and each blade; S4 Based on the solid contour models of the disk and each blade, the final assembly posture of each blade is optimized, and the contact stress of N nodes on the contact surface of each disk slot and blade root tooth on both sides and the gap between each disk slot and blade root tooth under the final assembly posture of each blade are predicted; S5 calculates the total imbalance between the wheel disc and all blades to optimize the arrangement of the blade numbers during assembly, and obtains the total imbalance prediction between the wheel disc and all blades after optimization, thereby realizing the sequential assembly of the blades. Then, the grinding area is determined, the optimal grinding amount of the blade position in the grinding area is solved, and the actual position points of the tail ends of all blades in the grinding area are ground.

2. The method for optimizing the digital precision assembly of a wheel disc and blades according to claim 1 is characterized in that: The specific implementation process of step S1 is as follows: S11 Scan the three-dimensional profiles of the wheel disc and each blade in turn through an optical scanner to obtain surface point cloud data of the wheel disc and each blade; before scanning, sort the blades and mark the serial numbers, and establish a coordinate system for the wheel disc and each blade with the rotation center position as the origin; S12 pre-processes and reversely models the surface point cloud data of the disk and each blade to obtain a preliminary solid scanning model of the disk and each blade; S13 compares the design models of the wheel disc and each blade with the preliminary solid scan model, and obtains various features and corresponding feature points in the preliminary solid scan model that do not meet the size and tolerance requirements according to the size and tolerance requirements of the processing drawings.

3. The method for optimizing the digital precision assembly of a wheel disc and blades according to claim 2 is characterized in that: The preprocessing is to perform multi-view data splicing, denoising, and point cloud data triangular meshing on the surface point cloud data; the reverse modeling is to re-scan the missing data and then construct the surface to obtain a preliminary solid scanning model of the wheel disc and each blade.

4. The method for optimizing the digital precision assembly of a wheel disc and blades according to claim 1 is characterized in that: The specific implementation process of step S2 is as follows: S21 uses a three-dimensional coordinate measuring machine to measure the key feature points on the wheel disc and each blade, and obtains discrete point cloud data of the key feature points on the wheel disc and each blade. During the measurement, the wheel disc and each blade establish a coordinate system with the rotation center position as the origin; S22 uses Gaussian filtering to filter the discrete point cloud data of key feature points on the wheel and each blade; S23: Using the filtered discrete point cloud data, the preliminary solid scan models of the wheel disc and blades obtained in step S1 are corrected to obtain solid contour models of the wheel disc and each blade.

5. The method for optimizing the digital precision assembly of a wheel disc and blades according to claim 1 is characterized in that: The specific implementation process of step S3 is as follows: S31 weighs the wheel disc and each blade to be assembled in turn, and calculates the volume of the solid contour model of the wheel disc and each blade in the CAD software; then calculates the density of the wheel disc and the blade according to the volume and weight information and assigns them to the wheel disc and the blade; S32 Calculate the center of mass of the disk and each blade and define it in the CAD software.

6. The method for optimizing the digital precision assembly of a wheel disc and blades according to claim 1 is characterized in that: The specific implementation process of step S4 is as follows: S41 imports the solid contour models of the disk and each blade into the finite element simulation software, assigns material properties to the disk and each blade, preliminarily assembles each blade onto the disk in turn, creates analysis steps, defines boundary conditions, and defines loads; then, meshes are divided for the disk and each blade; finally, N nodes are set on the contact surfaces on both sides of the disk slot and the blade root tooth, and the contact stress at the node is derived when the simulation is completed, and the standard deviations S1 and S2 of the contact stress at the N nodes on both sides are calculated respectively; S42 sets the number of particles and the maximum number of iterations. When the blade is initially assembled on the wheel, the angle between the line connecting the wheel and the blade center of mass and the x-axis is , and set the angle between the line connecting the center of mass of the wheel and the blade and the x-axis to -5°~ +5°, the fitness value is the difference between the standard deviations S1 and S2; Then, calculate the angle between the line connecting the center of mass of the wheel and the blade and the x-axis. -5°~ The fitness values ​​of various positions vary between +5°, and the optimal position with the smallest fitness value is found; S43 determines the final assembly posture of each blade according to the optimal position of each blade obtained in step S42, outputs the contact stress of N nodes on the contact surfaces of the corresponding wheel disc slot and the blade root tooth on both sides according to the simulation results of the final assembly posture of each blade, and cuts the simulation result model in post-processing to obtain the spacing between the feature points of each corresponding position on the wheel disc slot and the blade root tooth.

7. The method for optimizing the digital precision assembly of a wheel disc and blades according to claim 1 is characterized in that: The specific implementation process of step S5 is as follows: S51 calculates the unbalance of the wheel disc and the total unbalance of all blades; S52 Calculates the total unbalance of the disk and all blades and the total unbalance Angle with x-axis ; S53 indicates the distribution of each blade on the rotor by arranging the serial numbers of each blade, sets the initial population, the number of iterations, and sets the fitness value to The termination condition is set to the total imbalance between the disk and all blades The actual production requirement is less than 1 / 8 of the limit value. The particle swarm algorithm is used to optimize the arrangement of the blade numbers during assembly. During the optimization process, each blade is reassembled to the disk in turn for each arrangement of the blade numbers during assembly, so that the disk and the solid contour model of each blade are updated, and then steps S4, S51 and S52 are executed to obtain the total imbalance between the disk and all blades. ; The total imbalance between the optimized disk and all blades is recorded as , total unbalance The angle with the x-axis is ; S54: The total unbalance value after optimization obtained in step S53 The vector direction is taken as the reference, and the blade tail position within 30° clockwise or counterclockwise is set as the grinding area; S55 realizes multi-objective optimization through the interaction of genetic algorithm and grinding simulation, solves the optimal grinding amount of blade position in the grinding area, and grinds all actual position points of the tail end of all blades in the grinding area.

8. The method for optimizing the digital precision assembly of a wheel disc and blades according to claim 7 is characterized in that: The specific implementation process of step S51 is as follows: according to the wheel disc and the center of mass of each blade when the blade is finally assembled, the distance L and the angle between the wheel disc and the center of mass of each blade deviating from the center of the wheel disc are calculated. , then the unbalance M0 of the wheel and each blade is determined by the vector from the wheel center to the wheel center of mass of each blade. The components of the unbalance M0 in the x and y directions are , Calculated according to the following formula: Where, M is the mass of the disk or blade; The components of the wheel's unbalance in the x and y directions are recorded as , The components of the total unbalance of all blades in the x and y directions are recorded as , ; Components of the total unbalance of the blade in the x and y directions , It is the cumulative sum of the unbalanced components of each blade in the x and y directions.

9. The method for optimizing the digital precision assembly of a wheel disc and blades according to claim 7, characterized in that: The specific implementation process of step S55 is as follows: first, a total of K actual position points are taken at the tail ends of all blades in the grinding area, and the grinding amount of each actual position point is assumed to be equal. ( ), the grinding direction is along the blade tail end at point The normal of the corresponding grinding position point is set to ( ), then the grinding amount is: Then, the centroid of the blade after grinding according to the grinding amount is recalculated, and then the total imbalance of the wheel disc and all blades after grinding is calculated according to step S51 and step S52, which is recorded as ; The optimization target of the genetic algorithm is the difference between the standard deviation S1 and the standard deviation S2 and the total imbalance Minimum, set the initial population and the number of iterations, the optimization variable is the grinding amount of all actual position points of the tail ends of all blades in the grinding area, and after optimization, the optimal grinding amount of all actual position points of the tail ends of all blades in the grinding area is obtained, and then all actual position points of the tail ends of all blades in the grinding area are ground according to the optimal grinding amount.

10. The method for optimizing the digital precision assembly of a wheel disc and blades according to claim 1, characterized in that: The step S5 is followed by the following step: performing dynamic simulation analysis on the wheel disc blade assembly simulation model after grinding to obtain the dynamic balance performance of the wheel disc blade assembly simulation model after grinding.

Images