Method and system for optimizing machining allowance of thin-walled parts

Abstract

The present application relates to a kind of thin-walled part machining allowance optimization method and system, comprising: obtaining first point cloud and second point cloud, first point cloud, second point cloud are respectively the point cloud of thin-walled workpiece to be processed and target workpiece, by rough registration mode, obtain the initial Euclidean transformation parameter between first point cloud coordinate system and second point cloud coordinate system;Thin-walled piece machining allowance optimization model about the Euclidean transformation parameter of second point cloud coordinate system relative to first point cloud coordinate system is constructed, its constraint condition is that the machining allowance of the machining surface of thin-walled workpiece to be processed is not negative, and the optimization goal is that the machining allowance distribution of the machining surface of thin-walled workpiece to be processed is uniform;A group of transformation bases is calculated using covariance, and the first point cloud and the second point cloud are dimensionally reduced using the transformation base, the thin-walled piece machining allowance optimization model is solved after dimension reduction, the Euclidean transformation parameter is calculated, and the thin-walled workpiece machining positioning is realized.The present application is more suitable for the processing of thin-walled parts, and the positioning result of higher precision workpiece can be obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of machining positioning and precision detection, in particular to a thin-walled part machining allowance optimization method and system. BACKGROUND

[0002] In the field of aerospace, the integral structure is often used, which is light in weight, small in assembly workload and high in reliability. Such integral structure not only has small machining allowance and complex process requirements, but also has problems such as large size or local deformation at the beginning of machining, which cannot obtain a reference, so that subsequent numerical control machining positioning is difficult and must be adjusted repeatedly by manual operation, which is not only low in efficiency, but also high in scrap rate.

[0003] With the development of CAD / CAM technology, the computer-aided positioning method combining three-dimensional measurement technology and surface matching technology has gradually become a research hotspot. The purpose of machining positioning and precision detection is to obtain a reasonable deviation distribution of the workpiece by using the optimal transformation state of the measured model relative to the theoretical model, so as to verify whether the deviation meets the engineering requirements. Non-contact measurement equipment is widely used in the acquisition of measured models due to its fast measurement speed, no special requirements for workpiece structure, and insensitivity to workpiece material. This measurement method can quickly obtain a complete measurement model of the part, which is described in the form of point cloud. For a blank part with a free-form surface, all the surfaces on the blank can be divided into machining surfaces, tolerance surfaces and non-machining surfaces. In the machining process, the allowance of the machined surface is usually small, and the deformation amount varies with the position. The tolerance surface also has contour and position requirements, which is a mixed containment / positioning problem.

[0004] The point cloud of the measured model is large in scale, usually containing hundreds of thousands to millions of discrete points, resulting in long positioning calculation time and low calculation efficiency. Therefore, it is of great significance to quickly and accurately complete the positioning of the prefabricated part according to the point cloud of the measured model.

[0005] The commonly used machining positioning methods can be roughly divided into five categories: feature-based correspondence method, iterative registration method, fuzzy correspondence method, probability distribution method and image matching method. The above methods take too long to process the large number of point clouds in the registration problem, resulting in long positioning time and low efficiency. In addition, the above methods will also reduce the registration accuracy due to the influence of the internal points of the workpiece. SUMMARY

[0006] In view of the deficiencies of the prior art, the present application provides a thin-walled part machining allowance optimization method and system, which aims to ensure that the machining allowance is uniform and that the machining allowance shortage does not occur during the machining process, and to complete the workpiece positioning more efficiently and realize the rapid machining of the thin-walled part.

[0007] The technical scheme adopted by the present application is as follows:

[0008] A thin-walled part machining allowance optimization method, comprising:

[0009] Obtain a first point cloud and a second point cloud, the first point cloud and the second point cloud are point clouds of a thin-walled workpiece to be machined and a target workpiece respectively, and initial Euclidean transformation parameters between the first point cloud coordinate system and the second point cloud coordinate system are obtained through coarse registration;

[0010] A thin-walled part machining allowance optimization model about the Euclidean transformation parameters of the second point cloud coordinate system relative to the first point cloud coordinate system is constructed, which is used to describe that when the non-machining surfaces of the thin-walled workpiece to be machined and the target workpiece coincide, the machining allowance of the machining surface of the thin-walled workpiece to be machined is not negative under the constraint of the uniform distribution of the machining allowance of the machining surface of the thin-walled workpiece to be machined as the optimization goal;

[0011] A set of transformation bases is calculated using covariance, the first point cloud and the second point cloud are dimensionally reduced using the transformation bases, and the Euclidean transformation parameters are calculated by solving the thin-walled part machining allowance optimization model after dimension reduction, thereby realizing thin-walled workpiece machining positioning.

[0012] The objective function of the thin-walled part machining allowance optimization model with the optimization goal of the uniform distribution of the machining allowance of the machining surface of the thin-walled workpiece to be machined is:

[0013]

[0014] The constraint condition of the thin-walled part machining allowance optimization model that the machining allowance of the machining surface of the thin-walled workpiece to be machined is not negative is:

[0015] d i m (R,T)≥0,(i=1,2,...,n)

[0016] In the formula, d n (R,T) represents the directed distance of any non-boundary point between the first point cloud and the second point cloud, d i n (R,T) represents the directed distance of the non-boundary point between the first point cloud and the second point cloud, (R,T) represents the average value of the directed distance of the non-boundary point between the first point cloud and the second point cloud, (R,T) is the Euclidean transformation parameter of the second point cloud coordinate system relative to the first point cloud coordinate system, R represents the rotation matrix about α, β, γ, α, β, γ are the rotation angles about the x, y, z axes respectively, T represents the translation vector, the subscript i represents the coordinate point number of the first point cloud, and n is the number of coordinate points of the first point cloud; d i m (R,T) represents the directed distance of the boundary point between the first point cloud and the second point cloud.

[0017] The expression of the directed distance of any one non-boundary point on the first point cloud and the second point cloud is d n (R, T) = (gp - q) · n q , p is the coordinate of a non-boundary point in the first point cloud coordinate system, gp is the corresponding coordinate of p in the second point cloud coordinate system, g = (R, T), q is the point corresponding to p in the second point cloud coordinate system, n q is the unit outer normal vector of the curved surface at point q.

[0018] The expression of the directed distance of any one boundary point on the first point cloud and the second point cloud is p * is the coordinate of a boundary point in the first point cloud coordinate system, gp * is the corresponding coordinate of p * in the second point cloud coordinate system, q * is the point corresponding to p * in the second point cloud coordinate system, is the unit normal vector of the boundary curve at point q * , is the unit tangent vector of the boundary curve at point q * .

[0019] The covariance is used to calculate a set of transformation bases, comprising:

[0020] Taking initial base transformation parameters α0, β0, γ0 as input and final base transformation parameters α, β, γ as output, wherein α, β, γ are rotation angles about x, y, z axes respectively, and subscript 0 represents initial value;

[0021] Calculate a set of covariance matrices corresponding to the initial base transformation parameters, wherein the set of covariance matrices comprises covariance matrices C1, C2, C3 about xoy plane, xoz plane and yoz plane;

[0022] Establish an optimization function of eigenvalues of the set of covariance matrices:

[0023] min f(λ1, λ2, λ3, λ4, λ5, λ6) =

[0024] log 10 (log2λ1-log2λ2)+log 10 (log2λ3-log2λ4)+log 10 (log2λ5-log2λ6)

[0025] Wherein, λ1, λ2, λ3, λ4, λ5, λ6 are eigenvalues of the covariance matrices C1, C2, C3 respectively;

[0026] Solving the optimization function by using gradient descent method, output the final base transformation parameters.

[0027] Solving the thin-walled part machining allowance optimization model, calculating the Euclidean transformation parameters, including:

[0028] Taking the initial Euclidean transformation parameters as input, taking the final Euclidean transformation parameters α, β, γ, t x , t y , t z as output, where α, β, γ are the rotation angles about x, y, z axes respectively, t x , t y , t z are the translation distances along x, y, z axes respectively, and subscript 0 represents the initial value.

[0029] Calculating the gradient of the objective function f(R, T) of the thin-walled part machining allowance optimization model with respect to the Euclidean transformation parameters α, β, γ, t x , t y , t z , and finally obtaining the Euclidean transformation parameters (R, T) by iteration.

[0030] Further comprising: evaluating the thin-walled workpiece machining positioning result by using two evaluation indexes:

[0031]

[0032] Wherein, e1 is used to describe the error fluctuation of each point on the non-machining surface of the thin-walled workpiece to be machined, and e2 is used to describe the overall allowance distribution of the thin-walled workpiece to be machined.

[0033] A thin-walled part machining allowance optimization system, comprising:

[0034] An acquisition module acquires a first point cloud and a second point cloud, the first point cloud and the second point cloud are respectively the point clouds of a thin-walled workpiece to be machined and a target workpiece, and the initial Euclidean transformation parameters between the first point cloud coordinate system and the second point cloud coordinate system are acquired by coarse registration;

[0035] A modeling module constructs a thin-walled part machining allowance optimization model about the Euclidean transformation parameters of the second point cloud coordinate system relative to the first point cloud coordinate system, which is used to describe that when the non-machining surfaces of the thin-walled workpiece to be machined and the target workpiece coincide, the machining allowance of the machining surface of the thin-walled workpiece to be machined is not negative under the constraint of uniform distribution of the machining allowance of the machining surface of the thin-walled workpiece to be machined.

[0036] The computing module calculates a set of transformation bases by covariance, reduces dimensions of the first point cloud and the second point cloud by the transformation bases, solves the machining allowance optimization model of the thin-walled workpiece after the reduction of dimensions, calculates the Euclidean transformation parameters, and realizes the machining positioning of the thin-walled workpiece.

[0037] The evaluation module is further included to evaluate the machining positioning result of the thin-walled workpiece by the evaluation index.

[0038] The present application has the following advantages:

[0039] The present application describes the machining allowance optimization and positioning of the to-be-machined thin-walled workpiece as the coincidence of the non-machining surfaces of the to-be-machined thin-walled workpiece and the target workpiece, and the uniform distribution of the machining allowance under the premise of ensuring sufficient allowance. The present application calculates the Euclidean transformation parameters of the first point cloud coordinate system and the second point cloud coordinate system, and finally calculates the optimized machining allowance, so as to ensure the uniformity of the machining allowance and the absence of shortage of the machining allowance during the machining of the workpiece. Compared with the existing machining allowance optimization method, the present application is more suitable for the machining of thin-walled parts, and can obtain a workpiece positioning result with higher precision.

[0040] The present application uses the eigenvalues of the covariance matrix to represent the features retained after the reduction of dimensions of the point cloud. The data after the reduction of dimensions is reduced compared with the untreated data, but the features of the workpiece point cloud are retained. The use of the data after the reduction of dimensions for machining positioning can improve the positioning speed and meet the actual machining efficiency requirements.

[0041] Other features and advantages of the present application will be described in the following description, and some will become apparent from the description, or will be understood by those skilled in the art from the description and practice of the present application. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 FIG. 1 is a schematic diagram of the first point cloud of the to-be-machined thin-walled workpiece and the second point cloud of the target workpiece measured in the embodiment of the present application.

[0043] Figure 2 FIG. 2 is a schematic diagram of the meaning of d in the embodiment of the present application. n

[0044] Figure 3 FIG. 3 is a schematic diagram of the meaning of d in the embodiment of the present application. m

[0045] Figure 4 FIG. 4 is a projection diagram of the point cloud after the reduction of dimensions by using the base transformation in the embodiment of the present application.

[0046] Figure 1 In the figure: 1, to-be-machined thin-walled workpiece; 2, target workpiece. DETAILED DESCRIPTION

[0047] The specific embodiments of the present application will be described below with reference to the accompanying drawings.​​

[0048] Embodiments of the present application provide a thin-walled part machining allowance optimization method, comprising:

[0049] Obtain a first point cloud and a second point cloud, the first point cloud and the second point cloud are point clouds of a thin-walled workpiece to be machined and a target workpiece (i.e. a target workpiece to be machined from the thin-walled workpiece to be machined), and initial Euclidean transformation parameters between the first point cloud coordinate system and the second point cloud coordinate system are obtained through coarse registration;

[0050] A thin-walled part machining allowance optimization model for the Euclidean transformation parameters of the second point cloud coordinate system relative to the first point cloud coordinate system is constructed, which is used to describe that when the non-machining surfaces of the thin-walled workpiece to be machined and the target workpiece coincide, the machining allowance of the machining surface of the thin-walled workpiece to be machined is not negative under the constraint of the uniform distribution of the machining allowance of the machining surface of the thin-walled workpiece to be machined as the optimization target;

[0051] A set of transformation bases is calculated using covariance, and the first point cloud and the second point cloud are reduced in dimension using the transformation bases, and the Euclidean transformation parameters are calculated by solving the thin-walled part machining allowance optimization model after dimension reduction, and thin-walled workpiece machining positioning is realized.

[0052] Referring to Figure 1 The first power supply of the thin-walled workpiece to be machined 1 and the second point cloud of the target workpiece 2 to which the embodiments of the present application are directed are shown in the schematic diagram. The first point cloud can be obtained by using a Leica scanner, and the second point cloud can be obtained from a CAD model.

[0053] The objective function of the thin-walled part machining allowance optimization model with the optimization target of uniform distribution of the machining allowance of the machining surface of the thin-walled workpiece to be machined is:

[0054]

[0055] The constraint condition of the thin-walled part machining allowance optimization model that the machining allowance of the machining surface of the thin-walled workpiece to be machined is not negative is:

[0056] d i m (R,T)≥0,(i=1,2,...,n)

[0057] Wherein, d n (R,T) represents the directed distance of any one non-boundary point on the first point cloud and the second point cloud, d i n (R,T) represents the directed distance of any one non-boundary point on the first point cloud and the second point cloud, The average value of the directed distance between the first point cloud and the second point cloud, (R, T) is the Euclidean transformation parameter of the second point cloud coordinate system relative to the first point cloud coordinate system, R represents the rotation matrix about α, β, γ, α, β, γ are the rotation angles about the x, y, z axes respectively, T represents the translation vector, the subscript i represents the coordinate point number of the first point cloud, and n is the number of coordinate points of the first point cloud;

[0058] The target function represents the minimum variance of the machining allowance of each point of the thin-walled workpiece to be processed, so as to meet the uniform machining allowance standard. If the machining allowance is uniformly distributed, the cutting force and the elastic deformation of the process system during machining will be more uniform. Thus, the blank is protected from damage caused by excessive vibration. Ideally, the variance of the machining allowance of each point is 0.

[0059] Wherein, d i m (R, T) represents the directed distance between the first point cloud and the second point cloud, the constraint condition d i m (R, T) ≥ 0, (i = 1, 2,..., n) represents that the machining allowance of the thin-walled workpiece to be processed should be sufficient.

[0060] Wherein, (R, T) ∈ SE(3), SE(3) is a special Euclidean group, R ∈ SO(3), SO(3) is a characteristic orthogonal group, specifically: R = R(α)R(β)R(γ),

[0061]

[0062]

[0063]

[0064] Referring to Figure 2 , the expression of the directed distance of any one non-boundary point on the first point cloud (Measure frame) and the second point cloud (Model frame) is:

[0065] d n (R, T) = (gp-q) · n q

[0066] In the formula, p is the coordinate of a non-boundary point in the first point cloud coordinate system, gp is the corresponding coordinate of p in the second point cloud coordinate system, g = (R, T), q is the point corresponding to p in the second point cloud coordinate system, n q is the unit outer normal vector of the curved surface S (i.e. the surface of the target workpiece) at the point q.

[0067] Referring to Figure 3The expression for the directed distance between any boundary point on the first point cloud (Measure frame) and the second point cloud (Model frame) is:

[0068]

[0069] In the formula, p * It is the coordinate of a boundary point in the first point cloud coordinate system, gp * For p * The corresponding coordinates q in the second point cloud coordinate system * It is in the second point cloud coordinate system with p * Corresponding points The boundary curve (i.e., the boundary curve of the target workpiece surface) at point q * The unit normal vector at that location, The boundary curve at point q ＊ The unit tangent vector.

[0070] The calculation of a set of transformation bases using covariance includes:

[0071] The initial basis transformation parameters α0, β0, γ0 are used as inputs, and the final basis transformation parameters α, β, γ are used as outputs, where α, β, γ are the rotation angles about the x, y, z axes, respectively, and the subscript 0 represents the initial value.

[0072] Calculate the set of covariance matrices corresponding to the initial basis transformation parameters, wherein the set of covariance matrices includes covariance matrices C1, C2, and C3 with respect to the xoy plane, xoz plane, and yoz plane:

[0073]

[0074]

[0075]

[0076] Establish an optimization function for the eigenvalues ​​of the covariance matrix group:

[0077] min f(λ1,λ2,λ3,λ4,λ5,λ6)=

[0078] log 10 (log₂λ₁-log₂λ₂)+log 10 (log₂λ₃-log₂λ₄)+log 10 (log2λ5-log2λ6)

[0079] Where λ1, λ2, λ3, λ4, λ5, λ6 are the eigenvalues ​​of the covariance matrices C1, C2, and C3, respectively;

[0080] Solve the optimization function by gradient descent method, output the final base transformation parameters:

[0081] Calculate the gradient of the optimization function:

[0082]

[0083] Update the base transformation parameters:

[0084]

[0085]

[0086]

[0087] The final base transformation parameters are as follows:

[0088]

[0089] After projecting the point cloud to the above base transformation parameters, the point cloud projection map is obtained. Figure 4 (a), (b), (c) respectively represent the results of projecting the first point cloud and the second point cloud to the xoy plane, the xoz plane and the yoz plane after base transformation.

[0090] Solve the thin-walled part machining allowance optimization model to calculate the Euclidean transformation parameters, including:

[0091] Take the initial Euclidean transformation parameters as input, and take the final Euclidean transformation parameters α, β, γ, t x ,t y ,t z as output, where α, β, γ are the rotation angles about x, y, z axes respectively, and t x ,t y ,t z are the translation distances along x, y, z axes respectively, and subscript 0 represents the initial value.

[0092] Calculate the gradient of the objective function f(R, T) of the thin-walled part machining allowance optimization model with respect to the Euclidean transformation parameters α, β, γ, t x ,t y ,t z :

[0093]

[0094] Iteratively calculate the Euclidean transformation parameters:

[0095]

[0096]

[0097]

[0098]

[0099]

[0100]

[0101] Finally, the European transformation parameters (R, T) are obtained:

[0102]

[0103] T(x, y, z) = [-16.5154 21.9271 -1.7368].

[0104] The thin-walled part machining allowance optimization method of the embodiment of the application further comprises: evaluating the thin-walled workpiece machining positioning result by using two evaluation indexes:

[0105]

[0106] Wherein, e1 is used to describe the error fluctuation of each point of the non-machining surface of the thin-walled workpiece to be machined, and the smaller the value is, the smaller the error of the non-machining surface is; e2 is used to describe the overall allowance distribution of the thin-walled workpiece to be machined, indicating that the workpiece has an average machining allowance.

[0107] For this embodiment, e1 = 0.167391 mm, e2 = 0.786690 mm, indicating that the average error on the non-machining surface is about 0.158 mm, and the workpiece satisfies the constraint condition of sufficient machining allowance.

[0108] The embodiment of the application also provides a thin-walled part machining allowance optimization system, comprising:

[0109] An acquisition module acquires a first point cloud and a second point cloud, the first point cloud and the second point cloud are point clouds of a thin-walled workpiece to be machined and a target workpiece respectively, and an initial European transformation parameter between a first point cloud coordinate system and a second point cloud coordinate system is acquired by a coarse registration method;

[0110] A modeling module constructs a thin-walled part machining allowance optimization model about the European transformation parameter of the second point cloud coordinate system relative to the first point cloud coordinate system, which is used to describe that when the non-machining surfaces of the thin-walled workpiece to be machined and the target workpiece coincide, the machining allowance of the machining surface of the thin-walled workpiece to be machined is not negative under the constraint that the machining allowance of the machining surface of the thin-walled workpiece to be machined is uniform, and the machining allowance distribution of the machining surface of the thin-walled workpiece to be machined is even as an optimization target;

[0111] The computing module calculates a set of transformation bases by covariance, reduces dimensions of the first point cloud and the second point cloud by the transformation bases, solves the machining allowance optimization model of the thin-walled workpiece after the reduction of dimensions, calculates the Euclidean transformation parameters, and realizes the machining positioning of the thin-walled workpiece.

[0112] The evaluation module is further included to evaluate the machining positioning result of the thin-walled workpiece by using evaluation indexes.

[0113] Those skilled in the art can understand that the above description is only preferred embodiments of the present application, and is not used to limit the present application, although the present application has been described in detail with reference to the foregoing embodiments, and the technical solutions recorded in the foregoing embodiments can still be modified or equivalent replaced by some technical features for those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method of thin-walled part machining allowance optimization, characterized in that, The method comprises the following steps: obtaining a first point cloud and a second point cloud, the first point cloud and the second point cloud being point clouds of a thin-walled workpiece to be processed and a target workpiece respectively, and initial Euclidean transformation parameters between a first point cloud coordinate system and a second point cloud coordinate system being obtained through coarse registration; constructing a thin-walled workpiece processing allowance optimization model of Euclidean transformation parameters of the second point cloud coordinate system relative to the first point cloud coordinate system, which is used to describe that, when non-processing surfaces of the thin-walled workpiece to be processed and the target workpiece are coincident, under the constraint that processing allowances of processing surfaces of the thin-walled workpiece to be processed are not negative, a uniform distribution of the processing allowances of the processing surfaces of the thin-walled workpiece to be processed is an optimization target; calculating a set of transformation bases by covariance, and performing dimension reduction on the first point cloud and the second point cloud by using the transformation bases, and then solving the thin-walled workpiece processing allowance optimization model to calculate the Euclidean transformation parameters, so as to realize thin-walled workpiece processing positioning; a target function of the thin-walled workpiece processing allowance optimization model, which takes the uniform distribution of the processing allowances of the processing surfaces of the thin-walled workpiece to be processed as the optimization target, is as follows: ， a constraint condition of the thin-walled workpiece processing allowance optimization model, which is that the processing allowances of the processing surfaces of the thin-walled workpiece to be processed are not negative, is as follows: ， wherein, represents the directed distance from the first point cloud to the second point cloud, represents the directed distance from the first point cloud to the second point cloud, represents the average of the directed distance from the first point cloud to the second point cloud, is the Euclidean transformation parameter of the second point cloud coordinate system relative to the first point cloud coordinate system, represents the rotation matrix about , represents the rotation angle about , represents the translation vector, and the subscript represents the coordinate point number of the first point cloud, is the number of coordinate points of the first point cloud; represents the directed distance from the first point cloud to the second point cloud.

2. The thin-walled part allowance optimization method of claim 1, wherein, The expression of the directed distance of any one non-boundary point on the first point cloud and the second point cloud is , is the coordinate of a non-boundary point in the first point cloud coordinate system, is the corresponding coordinate in the second point cloud coordinate system, , is the corresponding point in the second point cloud coordinate system, is the unit outer normal vector of the curved surface at the point is the unit outer normal vector of the curved surface at the point .

3. The thin-walled part allowance optimization method of claim 1, wherein, The expression for the directed distance between any boundary point on the first point cloud and the second point cloud is: , These are the coordinates of a boundary point in the first point cloud coordinate system. for The corresponding coordinates in the second point cloud coordinate system It is in the second point cloud coordinate system and Corresponding points , Is the boundary curve at point The unit normal vector at that location, Is the boundary curve at point The unit tangent vector.

4. The thin-walled part allowance optimization method of claim 1, wherein, the step of calculating the set of transformation bases by covariance comprises the following steps: With initial basis transformation parameters As input, with the final basis transformation parameters For output, where They are respectively about The rotation angle of the axis, with the subscript 0 representing the initial value; computing a set of covariance matrices corresponding to the initial basis transform parameters, the set of covariance matrices comprising a covariance matrix for each of the plurality of planes planes, planes and planes ; establishing an optimization function of eigenvalues of the covariance matrix set: wherein , , are eigenvalues of the covariance matrix respectively. solving the optimization function by using a gradient descent method, and outputting final base transformation parameters.

5. The thin-walled part allowance optimization method of claim 1, wherein, the step of solving the thin-walled workpiece processing allowance optimization model to calculate the Euclidean transformation parameters comprises the following steps: with initial euclidean transformation parameters as input, with final euclidean transformation parameters as output, wherein are the rotation angles about the axes, respectively, are the translation distances along the axes directions, respectively, with index 0 representing the initial values; Objective function of computing thin-walled part machining allowance optimization model On the gradient of the equirectangular transform parameters Iterative calculation finally obtains the equirectangular transform parameters .

6. The thin-walled part allowance optimization method of claim 1, wherein, the method further comprises the following steps: evaluating the thin-walled workpiece processing positioning result by using two evaluation indexes; ， wherein, for describing the error fluctuation of each point of the non-machining surface of the thin-walled workpiece to be machined, for describing the overall allowance distribution of the thin-walled workpiece to be machined.

7. A thin-walled part machining allowance optimization system, characterized by, the method comprises the following steps: an obtaining module is configured to obtain a first point cloud and a second point cloud, the first point cloud and the second point cloud being point clouds of a thin-walled workpiece to be processed and a target workpiece respectively, and initial Euclidean transformation parameters between a first point cloud coordinate system and a second point cloud coordinate system being obtained through coarse registration; a modeling module is configured to construct a thin-walled workpiece processing allowance optimization model of Euclidean transformation parameters of the second point cloud coordinate system relative to the first point cloud coordinate system, which is used to describe that, when non-processing surfaces of the thin-walled workpiece to be processed and the target workpiece are coincident, under the constraint that processing allowances of processing surfaces of the thin-walled workpiece to be processed are not negative, a uniform distribution of the processing allowances of the processing surfaces of the thin-walled workpiece to be processed is an optimization target; a calculation module is configured to calculate a set of transformation bases by covariance, and perform dimension reduction on the first point cloud and the second point cloud by using the transformation bases, and then solve the thin-walled workpiece processing allowance optimization model to calculate the Euclidean transformation parameters, so as to realize thin-walled workpiece processing positioning; a target function of the thin-walled workpiece processing allowance optimization model, which takes the uniform distribution of the processing allowances of the processing surfaces of the thin-walled workpiece to be processed as the optimization target, is as follows: ， a constraint condition of the thin-walled workpiece processing allowance optimization model, which is that the processing allowances of the processing surfaces of the thin-walled workpiece to be processed are not negative, is as follows: ， wherein, represents the directed distance from the first point cloud to the second point cloud, represents the directed distance from the first point cloud to the second point cloud, represents the average of the directed distance from the first point cloud to the second point cloud, is the Euclidean transformation parameter of the second point cloud coordinate system relative to the first point cloud coordinate system, represents the rotation matrix about , represents the rotation angle about , represents the translation vector, and the subscript represents the coordinate point number of the first point cloud, is the number of coordinate points of the first point cloud; represents the directed distance from the first point cloud to the second point cloud.

8. The thin-walled part allowance optimization system of claim 7, wherein, the method further comprises an evaluation module configured to evaluate the thin-walled workpiece processing positioning result by using an evaluation index.

Images