Optimization method of wafer transmission mechanical finger structure and wafer transmission mechanical finger structure

Abstract

The invention discloses a wafer transmission mechanical finger structure optimization method and a wafer transmission mechanical finger structure. The wafer transmission mechanical finger structure optimization method comprises the following steps: establishing a mathematical model for a wafer transmission mechanical finger structure; s2, setting constraint conditions and value ranges of structural parameters; s3, constructing a target function; and S4, obtaining an optimal solution of the target function through an optimization method so as to obtain an optimal structure parameter. According to the method, the mathematical model of the wafer transmission mechanical finger structure is established, and the volume function is defined, so that the volume of the structure can be accurately calculated and predicted, the volume is optimized in the design stage, unnecessary material use is reduced, and the weight of the structure is reduced; due to the fact that the weight of the optimized mechanical finger structure is reduced, inertia in the moving process can be reduced, and therefore transmission efficiency and response speed are improved, energy consumption is reduced, inertia shaking is reduced, and transmission precision and stability are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of wafer transfer machinery, and in particular to an optimization method for a wafer transfer machinery finger structure and a wafer transfer machinery finger structure. Background Art

[0002] With the rapid development of the semiconductor industry, the performance of wafer transfer mechanical finger structures, as key components in the wafer handling process, directly affects production efficiency and wafer safety. The wafer transfer mechanical finger structure usually adopts a cantilever beam design, and its motion state will change according to the acceleration. The volume and weight of the traditional wafer transfer mechanical finger structure are not optimized, making the traditional wafer transfer mechanical finger structure heavier, affecting the transmission efficiency and response speed, and accompanied by inertial shaking, reducing the transmission accuracy and stability. Summary of the invention

[0003] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes a method for optimizing the structure of a wafer transfer mechanical finger.

[0004] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0005] An optimization method for a wafer transfer robot finger structure comprises: S1, establishing a mathematical model for the wafer transfer robot finger structure: V=f(x), V is the volume of the wafer transfer robot finger structure, x is a set of structural parameters of the wafer transfer robot finger structure; f(x) is a volume function of the wafer transfer robot finger structure; S2, setting constraints and value ranges of structural parameters; S3, constructing an objective function M=min(V), min is a minimum value function; S4, obtaining an optimal solution of the objective function through an optimization method to obtain optimal structural parameters.

[0006] Further,

[0007] x=[l1,l2,l3,l4,R1,R3,H,h1,h2,d2,d3,b1,b2,b3,w,θ]; l1 is the distance from the two symmetrical circular holes to the bottom surface, l2 is the distance from the first step to the bottom surface, l3 is the distance from the second step to the bottom surface, l4 is the distance from the intersection line of the arc and the rectangle to the bottom surface, R1 is the radius of the hollow arc, R3 is the radius of the large arc, H is the thickness of the robot finger, h1 is the depth of the four equal circular holes, h2 is the height of the ventilation channel, d2 is the outer diameter of the ventilation hole, d3 is the inner diameter of the ventilation hole, b1 is the distance between the centers of the two circular holes, b2 is the distance from the upper narrow step, b3 is the distance from the lower wide step, w is the width of the cross section of the ventilation channel, and θ is the angle between the centers of the two upper ventilation holes and the center of the arc.

[0008] Furthermore, the constraint conditions are: -R3+R1+d2≤0; l3-l4≤0; l2-l3≤0; l1-l2≤0; 1.2w-b1≤0; h1-H≤0; h2-h1≤0; -0.8R3+0.8R1+d2≤0; -0.8d2+d3≤0; b2-b3≤0.

[0009] Further, step S4 specifically includes: S41, establishing an initial population P0 = [X 1,0 ,X 2,0 ,…,X Np,0 ], where N P is the population size, X i,G is the vector of the i-th individual of the G-th generation, i = 1,…,Np; X i,G It has D dimensions, each of which corresponds to a type of structural parameter, X i，G =[x 1i,G ,x 2i,G ,…,x Di,G ], D is the number of types of structural parameters; G is the number of iterations, the initial value of G is 0, and the number of iterations of the initial population P0 is 0; the vector is composed of all structural parameters of the structural parameter set, and the structural parameters satisfy the constraint conditions and the value range; S42, select the best individual X from the latest generation population t,G ; All individuals X of the latest generation population i,G According to the learning rule, they all learn from the optimal individual to obtain the learned individual V i,G ; Compare V i,G and X i,G The fitness value of the new generation population P is selected as the population with the lowest fitness value. G+1 Individual X in i,G+1 ; S43, all individuals in the new generation population learn from each other according to the learning rule, and obtain the learned individual U i,G+1 ; Compare U i,G+1 and X i,G+1 The fitness value of the new generation population P is selected as the individuals with low fitness value. G+2 Individual X i,G+2 ; S44, randomly optimize the proportion of individuals in the new generation population to obtain new individuals W i,G+2 , compared with W i,G+2 and X i,G+2 The fitness value of the new generation population P is selected as the individuals with low fitness value. G+3 Individual X i,G+3 ; Update the G value, G = G + 3, and determine whether the G value is less than the iteration threshold G Max: If so, return to step S42, if not, stop the iteration, select the individual with the highest fitness value in the latest population as the optimal individual, and the structural parameters corresponding to the optimal individual are the optimal structural parameters.

[0010] Furthermore, before returning to step S42, the method further includes performing an adaptive boundary update: setting the upper boundary vector and and Update according to the following formula:

[0011] Furthermore, the new individuals W are obtained by randomly optimizing the proportion of individuals in the new generation population. i,G+2 , specifically including: obtaining W according to the following formula i,G+2 :W i,H+2 =X i,G+2 +rE G+1 +(1-r)(E G+1 -E G ); X t,H+1 and P G+1 The best individual and the worst individual, E G+1 P G+1 The maximum difference between two individuals in X t,G and P G The best individual and the worst individual, E G P G The maximum difference between two individuals in ; r is the random operator vector.

[0012] Furthermore, r=[r1,r2,…,r D ]; r k =rand(0,1)θ;k=1,…,D; θ is the iteration angle and m is the iteration index.

[0013] Furthermore, the learning rule of step S42 is: i,G =X i,G +r i (X t,G -T F M G );M G is the average vector of all individuals in the population, r i ∈(0,1), is a random coefficient; T F is the learning weight value.

[0014] Furthermore, the adaptation value is a function value obtained by substituting the structural parameters corresponding to the individual into the volume function f(x).

[0015] The present invention also provides a wafer transfer robot finger structure, wherein the structural parameters of the wafer transfer robot finger structure are optimal structural parameters obtained by the optimization method of the wafer transfer robot finger structure.

[0016] The present invention has the following beneficial effects:

[0017] By establishing a mathematical model of the wafer transfer mechanical finger structure and defining the volume function, the volume of the structure can be accurately calculated and predicted, so that the volume can be optimized in the design stage, unnecessary material use can be reduced, and the weight of the structure can be reduced; the optimized mechanical finger structure can reduce inertia during movement due to the reduced weight, thereby improving transmission efficiency and response speed and reducing energy consumption; by setting the constraints and value range of the structural parameters, it can ensure that the rigidity and strength of the structure are not sacrificed while reducing weight, thereby enhancing the stability during the transmission process and reducing wafer damage caused by inertial shaking. The objective function in the optimization method takes minimizing the volume as the optimization goal, while taking into account the stability and accuracy requirements of the structure, and by solving the optimal structural parameters, the structural optimization of the finger structure is achieved, the mass and motion inertia of the wafer transfer mechanical finger structure are reduced, and the transmission efficiency and response speed are effectively improved, and the inertial shaking is reduced, and the transmission accuracy and stability are improved.

[0018] In addition to the above-described purposes, features and advantages, the present invention has other purposes, features and advantages. The present invention will be further described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings constituting a part of this application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0020] Figure 1 It is a schematic diagram of the overall process of the present invention;

[0021] Figure 2 It is a partial flow diagram of the present invention;

[0022] Figure 3 It is the trend graph of iteration angle;

[0023] Figure 4 It is a schematic diagram of the structure of the wafer transfer mechanical finger structure;

[0024] Figure 5 This is a perspective view of the internal structure of the wafer transfer mechanical finger structure;

[0025] Figure 6 It is a schematic diagram of the size parameters of the wafer transfer mechanical finger structure;

[0026] Figure 7 It is a convergence trend diagram of the fitness value of an embodiment of the present invention. DETAILED DESCRIPTION

[0027] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] It should be noted that all directional indications in the embodiments of the present invention (such as up, down, left, right, front, back, etc.) are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0030] In addition, the descriptions of "first", "second", etc. in the present invention are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0031] Please refer to Figure 1 A method for optimizing a wafer transfer mechanical finger structure in a preferred embodiment of the present invention includes steps S1, S2, S3 and S4.

[0032] S1. Establish a mathematical model for the wafer transfer robot finger structure: V = f(x), V is the volume of the wafer transfer robot finger structure, x is the set of structural parameters of the wafer transfer robot finger structure; f(x) is the volume function of the wafer transfer robot finger structure.

[0033] S2, set the constraints and value range of structural parameters.

[0034] S3, construct the objective function M = min (V), min is the minimum value function, that is, the optimization goal is to make the value of V as small as possible. When determining the value, it is necessary to take the trend of smaller values ​​to set the goal for subsequent optimization direction and judgment. That is, when there are three values, substitute the three values ​​into f (x), and the smallest value is the current optimal choice.

[0035] S4, the optimal solution of the objective function is obtained by optimization method to obtain the optimal structural parameters.

[0036] The present invention provides an optimization method for a wafer transmission mechanical finger structure. By establishing a mathematical model of the wafer transmission mechanical finger structure and defining a volume function, the volume of the structure can be accurately calculated and predicted, thereby optimizing the volume in the design stage, reducing unnecessary material use, and reducing the weight of the structure; the optimized mechanical finger structure can reduce inertia during movement due to the reduced weight, thereby improving transmission efficiency and response speed, and reducing energy consumption; by setting the constraint conditions and value ranges of the structural parameters, it can ensure that the rigidity and strength of the structure are not sacrificed while reducing the weight, thereby enhancing the stability during the transmission process and reducing wafer damage caused by inertial shaking. The objective function in the optimization method takes minimizing the volume as the optimization goal, while taking into account the stability and accuracy requirements of the structure, and by solving the optimal structural parameters, the structural optimization of the finger structure is achieved, the mass and motion inertia of the wafer transmission mechanical finger structure are reduced, and the transmission efficiency and response speed are effectively improved, and the inertial shaking is reduced, and the transmission accuracy and stability are improved.

[0037] Reference Figure 4 and Figure 5 The wafer transfer mechanical finger structure includes a rectangular section 100 at the bottom, the upper end of the rectangular section 100 is connected to a trapezoidal section 200 that narrows upward, the upper end of the trapezoidal section 200 is connected to a strip section 300, and the upper ends of the strip section 300 are connected to arc fingers 400 that are symmetrically arranged on both sides, and one end of the arc fingers 400 on both sides is connected, and the other end is disconnected to form a gap 410. The center of the circle and the inner and outer diameters of the arc fingers 400 on both sides are the same. Two circular ventilation holes 500 formed by depressions facing downward are respectively provided on the arc fingers 400 on both sides. The ventilation holes 500 on both sides are symmetrically arranged. A circular hole 110 is opened on one side of the rectangular section 100, and a ventilation channel 600 is provided in the wafer transfer mechanical finger structure, one end of the ventilation channel 600 is connected to the circular hole 110, and extends to the trapezoidal section 200, the strip section 300 and the arc fingers 400 in sequence, and finally connected to the ventilation circular hole 500. One end of the arc finger 400 away from the long strip section 300 is a semicircular arc, and the center of the semicircular arc is concentric with the center of the adjacent ventilation hole 500.

[0038] In a specific embodiment of the present invention,

[0039] f(x) is derived from the following formula:

[0040]

[0041] V3=wh2[(2l4+R3-R1-2l1-b1)+π(R1+R3)];

[0042] f(x)=V=V1-V2-V3.

[0043] V1 represents the approximate total volume of the wafer transfer robot finger, V2 represents the volume of four equal circular ventilation holes 500 , and V3 represents the approximate volume of the ventilation channel 600 .

[0044] The structural parameter set is x = [l1, l2, l3, l4, R1, R3, H, h1, h2, d2, d3, b1, b2, b3, w, θ]; refer to Figure 6 , l1 is the distance from the two symmetrical circular holes to the bottom surface, that is, the distance from the circular hole 110 to the bottom of the rectangular segment 100. l2 is the distance from the first step to the bottom surface, that is, the distance from the junction of the rectangular segment 100 and the trapezoidal segment 200 to the bottom of the rectangular segment 100. l3 is the distance from the second step to the bottom surface, that is, the distance from the junction of the trapezoidal segment 200 and the long strip segment 300 to the bottom of the rectangular segment 100. l4 is the distance from the intersection line of the arc and the rectangle to the bottom surface, that is, the distance from the junction of the upper end of the long strip segment 300 and the circular arc finger 400 to the bottom of the rectangular segment 100. l5 is the distance from the center of the arc to the bottom surface, that is, the distance from the center of the circular arc finger 400 to the bottom of the rectangular segment 100. R1 is the radius of the hollow arc, that is, the radius of the inner circle of the circular arc finger 400. R2 is the radius of the distribution circle where the centers of the four ventilation circular holes 500 are located, and R3 is the radius of the large arc, that is, the radius of the outer circle of the circular arc finger 400. H is the thickness of the robot finger, h1 is the depth of the four equal ventilation holes 500, h2 is the height (thickness) of the ventilation channel 600, d2 is the outer diameter of the ventilation hole 500, d3 is the inner diameter of the ventilation hole 500, b1 is the distance between the centers of the two holes 110, b2 is the distance of the upper narrow step (the width of the long strip segment 300), b3 is the distance of the lower wide step (the width of the rectangular segment 100), w is the cross-sectional width of the ventilation channel 600, and θ is the angle between the center of the two upper ventilation holes 110 and the center of the arc, as shown in FIG. Figure 7 As shown, the angle between the center of the two lower vent holes 110 and the center of the arc is also θ.

[0045] In a specific embodiment of the present invention, the constraint condition is:

[0046] -R3+R1+d2≤0; l3-l4≤0; l2-l3≤0; l1-l2≤0; 1.2w-b1≤0; h1-H≤0; h2-h1≤0; -0.8R3+0.8R1+d2≤0; -0.8d2+d3≤0; b2-b3≤0.

[0047] The value range is: 5≤l1≤25; 25≤l2≤40; 40≤l3≤80; 80≤l4≤160; 20≤R1≤36; 40≤R3≤60; 4≤H≤6; 1≤h1≤3; 0.8≤h2≤1.2; 10≤d2≤20; 10≤d3≤20; 6≤b1≤15; 20≤b2≤40; 30≤b3≤50; 2≤w≤6;

[0048] Reference Figure 2 In a specific embodiment of the present invention, step S4 specifically includes steps S41, S42, S43, and S44.

[0049] S41, establish an initial population P0 = [X 1,0 ,X 2,0 ,…,X Np,0 ], where N P is the population size, X i,G is the i-th individual vector of the G-th generation, i = 1,…,Np; X i,G It has D dimensions, each of which corresponds to a type of structural parameter, X i，G =[x 1i,G ,x 2i,G ,…,x Di,G ], D is the number of structural parameters, x ki,G is the vector X i，G The components of the kth dimension of , k = 1, ..., D. G is the number of iterations, the initial value of G is 0, and the number of iterations of the initial population P0 is 0; the components of all dimensions of the vector correspond to all structural parameters of the structural parameter set, and the structural parameters satisfy the constraints and the value range.

[0050] In the embodiment of the present invention, the structural parameter set is x=[l1,l2,l3,l4,R1,R3,H,h1,h2,d2,d3,b1,b2,b3,w,θ], which has 16 variable parameters, so the value of D is 16. 1i,G , x 2i,G , …, x Di,G They correspond to a selection of values ​​for the 16 parameters l1, l2, l3, l4, R1, R3, H, h1, h2, d2, d3, b1, b2, b3, w, and θ respectively.

[0051] S42, from the latest generation population P G Select the best individual X t,G ; The optimal individual is the one whose corresponding structural parameter is substituted into the volume function f(x) in the population and the function value obtained is the smallest. i,G According to the learning rule, they all learn from the optimal individual to obtain the learned individual V i,G ; Compare V i,G and X i,G The fitness value of the new generation population P is selected as the population with the lowest fitness value. G+1 Individual X in i,G+1 When executed for the first time, the latest generation population is P0. In a specific embodiment of the present invention, the fitness value is the function value obtained by substituting the structural parameters corresponding to the individual vector into the volume function f(x). The update logic of step S42 can be performed by the following formula:

[0052] That is, when f(X i,G )≤f(V i,G ), X i,G+1 =X i,G , otherwise X i,G+1 =V i,G . f(X i,G ) is the individual vector X i,G The function value of the volume function corresponding to the corresponding structural parameters.

[0053] In a specific embodiment of the present invention, the learning rule of step S42 is:

[0054] V i,G =X i,G +r i (X t,G -T F M G );M G is the population P G The average vector of all individuals in, r i ∈(0,1), is a random coefficient; T F is the learning weight value. F =round[1+rand(0,1)], rand(0,1) means taking any value between 0 and 1, and round() means rounding the value to the nearest integer. It can be understood that the value of any dimension in the vector needs to be within the range. If it exceeds the range, the boundary value beyond the corresponding boundary is taken. For example, 5≤l1≤25, then V 1i,G The value range is 5-25. If V 1i,G If the calculation exceeds 25, then V 1i,G The value is 25.

[0055] S43, all individuals in the new generation population learn from each other according to the learning rule, and obtain the learned individual U i,G+1 ; Compare U i,G+1 and X i,G+1 The fitness value of the new generation population P is selected as the individuals with low fitness value. G+2 Individual X i,G+2 Steps S41, S42 and S43 are improved steps based on the algorithm TLBO.

[0056] Specifically, the learning rule of step S43 is:

[0057]

[0058] r m ∈(0,1), is a random value. m,G+1 and X n,G+1 are two from the current population P G+1 Two individual vectors randomly selected from , and m≠n. That is, when f(X m,G+1 )≤f(X n,G+1 ),U i,G+1 =X m,G+1 +r m (X m,G+1 -X n,G+1 ), otherwise U i,G+1 =X m,G+1 +r m (X m,G+1 +X n,G+1 ).

[0059] In step S43, individual updating can be performed according to the following formula:

[0060] That is, when f(X i,G+1 )≤f(U i,G+1 ), X i,G+2 =X i,G+1 , otherwise X i,G+2 =U i,G+1 .

[0061] S44, randomly optimize the proportion of individuals in the new generation population to obtain new individuals W i,G+2 , compared with W i,G+2 and X i,G+2 The fitness value of the new generation population P is selected as the individuals with low fitness value. G+3 Individual X i,G+3 ; Update G value, G = G + 3, here update G value, that is, the three iterations of steps S42, S43 and S44 are counted into the number of iterations, the number of iterations is updated to obtain the latest number of iterations, and it is determined whether G value is less than the iteration threshold G Max: If so, return to step S42, indicating that the number of iterations is not enough and further iterative updates are needed.

[0062] If not, the iteration is stopped, indicating that the number of iterations has reached the standard, and the individual with the lowest fitness value (that is, the lowest volume function value) in the latest population can be selected as the optimal individual, and the structural parameters corresponding to the optimal individual are the optimal structural parameters.

[0063] In a specific embodiment of the present invention, in step S44, before returning to step S42, it also includes performing adaptive boundary update: setting the upper boundary vector and the lower boundary vector is the value of the first dimension of the upper boundary vector, is the value of the i-th dimension of the upper boundary vector, i=1,…,D. is the value of the first dimension of the lower boundary vector, is the value of the i-th dimension of the lower boundary vector, i = 1,…,d. The values ​​of each dimension in the population individual vector need to be within the value range of the corresponding dimension of the upper boundary vector and the lower boundary vector, such as If the range is exceeded in the subsequent calculation and update process, the boundary value of the exceeded boundary can be taken. Each component of all individual vectors satisfies the following formula: x ji,G Represents x i,G The component of the j-th dimension of , j = 1,…,D.

[0064] The initial value ranges of the parameters of the finger structure are as follows: 5≤l1≤25; 25≤l2≤40; 40≤l3≤80; 80≤l4≤160; 20≤R1≤36; 40≤R3≤60; 4≤H≤6; 1≤h1≤3; 0.8≤h2≤1.2; 10≤d2≤20; 10≤d3≤20; 6≤b1≤15; 20≤b2≤40; 30≤b3≤50; 2≤w≤6; That is, it can be understood that The initial value of is 5. The initial value is 25.

[0065] Of course, when it comes to step S44, three iterations have been performed. and It needs to be updated according to the following formula:

[0066]

[0067]

[0068] for The component of the j-th dimension of ; for That is, after the latest generation of population is obtained in step S44, the maximum and minimum values ​​of each dimension corresponding to the individuals of the new population are used as the value range of the dimension in the future. The values ​​of each dimension of the individual are the values ​​of the structural parameters of the corresponding species. j1,G That is, the individual x of the latest generation population 1,G The value of the component of the jth dimension in , and so on. Max is the function of taking the maximum value, and Min is the function of taking the minimum value, so that the value range can be gradually converged to improve the optimization efficiency.

[0069] In a specific embodiment of the present invention, the new individuals W are obtained by performing random ratio optimization on the individuals of the new generation population. i,G+2 , specifically including:

[0070] W is obtained according to the following formula i,G+2 :

[0071] W i,G+2 =X i,G+2 +rE G+1 +(1-r)(E G+1 -E G );

[0072]

[0073] X t,G+1 and P G+1 The best individual vector and the worst individual vector are represented by , and the best individual and the worst individual represent that the function value of the corresponding volume function f(x) is the smallest and largest in the population.

[0074] E G+1 P G+1 The maximum difference vector between two individuals in ;

[0075] X t,G and P G The best individual vector and the worst individual vector in E G P G The maximum difference vector between two individuals in . r is a random operator vector with the same number of dimensions as the individual vector, which is 16 dimensions.

[0076] In a specific embodiment of the present invention, r=[r1, r2, ..., r D ]; r k is the component of the kth dimension of r, k=1,…,D.

[0077] r k=rand(0,1)θ;k=1,…,D;

[0078] θ is the iteration angle and m is the iteration index.

[0079] The θ designed in the r operator decreases slowly as G increases at the beginning and end stages, which improves the algorithm's exploration ability. On the other hand, in the middle stage, θ decreases rapidly as G increases, which can improve the algorithm's development ability. Therefore, by selecting different m values, the algorithm can achieve a balance between exploration and development capabilities. Figure 3 As shown in Figure 1, the changing trend of θ when m = 8. Figure 7 As shown in the figure, the horizontal axis is the number of iterations, and the vertical axis is the log10 value of the volume function f(x) corresponding to the optimal individual, that is, log10(f(x)). After adding the r operator, as the number of iterations increases, the fitness value decreases rapidly at the beginning, which shows that this method can obtain a better value in a shorter time. At the same time, it can also obtain better values ​​with a relatively large number of iterations. Therefore, this method can achieve a convergence effect when solving, which reflects the effectiveness of this method in practical applications.

[0080] The present invention also provides a wafer transfer robot finger structure, wherein the structural parameters of the wafer transfer robot finger structure are optimal structural parameters obtained by an optimization method of the wafer transfer robot finger structure.

[0081] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for optimizing the structure of wafer transfer mechanical fingers, characterized in that: The steps include: S1, establish a mathematical model for the wafer transfer robot finger structure: V = f(x), V is the volume of the wafer transfer robot finger structure, x is the set of structural parameters of the wafer transfer robot finger structure; f(x) is the volume function of the wafer transfer robot finger structure; S2, setting constraints and value ranges of structural parameters; S3, construct the objective function M=min(V), where min is the minimum value function; S4, the optimal solution of the objective function is obtained by optimization method to obtain the optimal structural parameters.

2. The method for optimizing the wafer transfer mechanical finger structure according to claim 1, characterized in that: f x=[l1,l2,l3,l4,R1,R3,H,h1,h2,d2,d3,b1,b2,b3,w,θ]; l1 is the distance from the two symmetrical circular holes to the bottom surface, l2 is the distance from the first step to the bottom surface, l3 is the distance from the second step to the bottom surface, l4 is the distance from the intersection line of the arc and the rectangle to the bottom surface, R1 is the radius of the hollow arc, R3 is the radius of the large arc, H is the thickness of the robot finger, h1 is the depth of the four equal circular holes, h2 is the height of the ventilation channel, d2 is the outer diameter of the ventilation hole, d3 is the inner diameter of the ventilation hole, b1 is the distance between the centers of the two circular holes, b2 is the distance from the upper narrow step, b3 is the distance from the lower wide step, w is the width of the cross-section of the ventilation channel, and θ is the angle between the centers of the two upper ventilation holes and the center of the arc.

3. The method for optimizing the wafer transfer robot finger structure according to claim 2, characterized in that: The constraints are: -R3+R1+d2≤0; l3-l4≤0； l2-l3≤0； l1-l2≤0； 1.2w-b1≤0; h1-H≤0; h2-h1≤0; -0.8R3+0.8R1+d2≤0; -0.8d2+d3≤0; b2-b3≤0.

4. The method for optimizing the wafer transfer mechanical finger structure according to claim 1, characterized in that: Step S4 specifically includes: S41, establish an initial population P0 = [X 1,0 ,X 2,0 ,…,X Np,0 ], where N P is the population size, X i,G is the vector of the i-th individual of the G-th generation, i = 1,…,Np; X i,G It has D-dimensional components, each of which corresponds to a type of structural parameter, X i，G =[x 1i,G ,x 2i,G ,…,x Di,G ], D is the number of types of structural parameters; G is the number of iterations, the initial value of G is 0, and the number of iterations of the initial population P0 is 0; S42, select the best individual X from the latest generation population t,G ; All individuals X of the latest generation population i,G According to the learning rule, they all learn from the optimal individual to obtain the learned individual V i,G ; Compare V i,G and X i,G The fitness value of the new generation population P is selected as the population with the lowest fitness value. G+1 Individual X in i,G+1 ; S43, all individuals in the new generation population learn from each other according to the learning rule, and obtain the learned individual U i,G+1 ; Compare U i,G+1 and X i,G+1 The fitness value of the new generation population P is selected as the individuals with low fitness value. G+2 Individual X i,G+2 ; S44, randomly optimize the proportion of individuals in the new generation population to obtain new individuals W i,G+2 , compared with W i,G+2 and X i,G+2 The fitness value of the new generation population P is selected as the individuals with low fitness value. G+3 Individual X i,G+3 ; Update the G value, G = G + 3, and determine whether the G value is less than the iteration threshold G Max : If so, return to step S42, if not, stop the iteration, select the individual with the highest fitness value in the latest population as the optimal individual, and the structural parameters corresponding to the optimal individual are the optimal structural parameters.

5. The method for optimizing the wafer transfer robot finger structure according to claim 4, characterized in that: Before returning to step S42, the method further includes performing adaptive boundary update: Let the upper boundary vector and the lower boundary vector and Update according to the following formula:

6. The method for optimizing the wafer transfer robot finger structure according to claim 4, characterized in that: The new generation of individuals in the population is randomly optimized to obtain new individuals W i,G+2 , specifically including: W is obtained according to the following formula i,G+2 : W i,G+2 =X i,G+2 +rE G+1 +(1-r)(E G+1 -HAVE BEEN G ); E G+1 =X t,G+1 -X d1,G+1 ; E G =X t,G -X d2,G ; X t,G+1 and X d1,G+1 P G+1 The best individual and the worst individual, E G+1 P G+1 The maximum difference between two individuals in ; X t,G and X d1,G P G The best individual and the worst individual, E G P G The maximum difference between two individuals in ; r is the random operator vector.

7. The method for optimizing the wafer transfer robot finger structure according to claim 6, characterized in that: r＝[r1,r2,…,r D ]； r k rand(0,1)θ; k 1,…,D; θ is the iteration angle and m is the iteration index.

8. The method for optimizing the wafer transfer robot finger structure according to claim 4, characterized in that: The learning rule of step S42 is: V i,G =X i,G +r i (X t,G -T F M G ); M G is the average vector of all individuals in the population, r i ∈(0,1), is a random coefficient; T F is the learning weight value.

9. The method for optimizing the wafer transfer robot finger structure according to claim 4, characterized in that: The fitness value is the function value obtained by substituting the structural parameters corresponding to the individual into the volume function f(x).

10. A wafer transfer mechanical finger structure, characterized in that: The structural parameters of the wafer transfer robot finger structure are the optimal structural parameters obtained by the optimization method of the wafer transfer robot finger structure described in any one of claims 1 to 9.

Images