Two-dimensional parts arrangement method based on dynamic iteration and extended area overlap rate optimization

Abstract

The present invention discloses a two-dimensional parts arrangement method based on dynamic iteration and optimization of the overlap rate of the extended area, and belongs to the field of computer science and engineering. The method mainly solves the problem of how to optimize the arrangement efficiency while improving the utilization rate of the plate when arranging irregular two-dimensional parts on the plate. By acquiring a binary image from the CAD model of the part, sorting the part image by size, and performing a series of extension processing on the part image based on the clearance requirements, the present invention uses an improved genetic algorithm to optimize the position of the part image on the plate. The optimization goal is to maximize the overlap rate of the second extended area of ​​the part image with the extended area of ​​the plate image and the arranged image. The method significantly improves the utilization efficiency of the plate through dynamic iterative selection and arrangement optimization, and simplifies the arrangement process, which is of great significance to improving production efficiency and reducing material costs.

Description

Technical Field

[0001] The present invention relates to a two-dimensional parts arrangement method based on dynamic iteration and optimization of extended area overlap rate, which belongs to the field of computer science and engineering. Background Art

[0002] In the fields of manufacturing and engineering design, especially in industries such as metalworking, apparel manufacturing, and aerospace parts production, the efficient arrangement of two-dimensional parts is crucial for improving material utilization, reducing costs, and increasing production efficiency. Traditionally, methods for arranging two-dimensional parts have relied on simple geometric alignment and manual optimization, which may be effective for processing simple shapes and small numbers of parts. However, with the increasing complexity of part shapes and the demand for mass production, the limitations of these traditional methods have gradually become apparent.

[0003] Existing 2D part placement technologies face multiple challenges. First, traditional methods struggle to fully account for complex interactions between parts, such as the minimum clearances that must be maintained, the impact of possible part rotation on material utilization, and how to handle irregularities in part shape. Second, existing technologies lack effective dynamic iteration mechanisms to adaptively adjust strategies during the placement process to find the optimal solution. Furthermore, many existing algorithms are inefficient when processing large numbers of parts, making them difficult to meet the demands of rapid production.

[0004] To address these challenges, this paper proposes a two-dimensional parts placement method based on dynamic iteration and extended region overlap optimization. This method aims to further improve placement efficiency and computational speed while ensuring high material utilization, particularly for placing large numbers of complex parts. Summary of the Invention

[0005] The present invention relates to a two-dimensional parts arrangement method based on dynamic iteration and extended area overlap rate optimization, which optimizes the arrangement of irregular two-dimensional parts on a plate, thereby further improving the arrangement efficiency while increasing the utilization rate of the plate.

[0006] The present invention adopts the following technical solution: a two-dimensional parts arrangement method based on dynamic iteration and optimization of the extended area overlap rate, wherein the arrangement process mainly includes the following steps:

[0007] A two-dimensional parts arrangement method based on dynamic iteration and extended area overlap rate optimization is characterized in that the arrangement process mainly includes the following steps:

[0008] S1. Obtain a part image based on the CAD model of the part and perform binarization on it, where black is the background and white is the part area;

[0009] S2. Get the horizontal dimension L of the part hand vertical dimension L v , and take the corresponding size as a reference to obtain the corresponding standard rectangular image;

[0010] S3, based on the lateral dimension L of the part h and vertical dimension L v The sum L t Sort the part images from large to small as a basis;

[0011] S4. Expand the part image for the first time according to the clearance requirements of the part;

[0012] S5. Calculate the width w2 of the corresponding expansion area according to the size of the part, and perform a second expansion on the part graphic;

[0013] S6. Based on a predetermined value N max And the corresponding size principle, select the first N part images as a group of images;

[0014] S7, the plate image is expanded to be consistent with the width w2 of the second expansion area of ​​the current image;

[0015] S8. Based on the improved genetic algorithm, the position of each image in the set of part images on the plate image is encoded, mutated, and decoded, and the overlap rate of the second expanded area of ​​the part image, the expanded area of ​​the plate image, and the arranged images is optimized as the optimization target;

[0016] S9, selecting the image with the highest overlap rate and arranging it on the plate image according to the specified optimized position, and removing the image from the previous image sequence;

[0017] S10. Repeat the above steps S6-9 until all parts are arranged.

[0018] In step S5, the corresponding size principle is that the size L of the last part, that is, the smallest part, in each group of part images selected is t It cannot be smaller than α times the size of the first part, which is the largest part.

[0019] In step S5, after each image is arranged, a new group of images needs to be selected and the overlap rate of the extended area after the position of each image in the group of images is optimized needs to be recalculated.

[0020] The beneficial effects of the present invention are as follows: it changes the traditional idea of ​​sequential optimization based solely on the size of the part images. Although the general idea is still based on the order of size from large to small, the specific arrangement order of each part is further guided by the overlap rate of the expanded area.

[0021] The core of this invention lies in the use of a dynamic iterative strategy and an extended area overlap optimization mechanism. By accurately analyzing the CAD model of the part, this method can effectively handle irregularities in part shape and rotation requirements. First, the parts are binarized based on their actual size and shape characteristics to more accurately calculate the spatial relationship between parts. Then, by dynamically adjusting the arrangement order and direction of the parts and continuously optimizing the extended area overlap between parts during the iterative process, this method can ensure a compact and efficient part arrangement while reducing material waste.

[0022] To achieve this goal, the present invention employs an improved genetic algorithm to optimize parts placement. By encoding, mutating, and decoding the positions of part images and optimizing the overlap ratio of the extended regions, this method can find a near-optimal placement solution in a relatively short time. Furthermore, by dynamically adjusting the search range and parameter settings, the present invention can adapt to placement tasks of varying complexity, providing a flexible and effective solution.

[0023] Another innovation of this invention is that by precisely controlling the gaps and expansion areas between parts, it effectively reduces material waste during the cutting process, further improving material utilization. Furthermore, this method considers practical issues that may be encountered during production, such as the impact of tool size and cutting paths, ensuring that the proposed layout is not only efficient in theory but also feasible in actual production.

[0024] Overall, the proposed two-dimensional parts layout method based on dynamic iteration and extended region overlap optimization is a significant improvement over existing technologies. It not only effectively addresses the challenges of arranging complex parts, but also significantly improves material utilization and production efficiency, possessing broad application prospects and practical value. Through this invention, companies in the manufacturing and engineering design fields will be able to reduce waste, lower costs, and enhance product competitiveness. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 Schematic diagram of the image pre-adjustment process; 1(a) shows the binarization of the original image; 1(b) shows an example of image cropping; 1(c) shows the cropped image;

[0026] Figure 2 Schematic diagram of the image expansion process; 2(a) is the identification of the expanded area; 2(b) is the visualization of the expanded area;

[0027] Figure 3 Schematic diagram of image expansion results; 3(a) is the initial expansion result; 3(b) is the subsequent expansion result; 3(c) is the subsequent expansion area;

[0028] Figure 4Schematic diagram of the comparison of plate images before and after expansion; 4(a) is the plate image before expansion; 4(b) is the plate image after expansion;

[0029] Figure 5 is a schematic diagram of pattern size;

[0030] Figure 6 Schematic diagram of binary coding in genetic algorithm;

[0031] Figure 7 is a schematic diagram of pattern size and position;

[0032] Figure 8 Flowchart of theoretical framework for pattern arrangement;

[0033] Figure 9 Schematic diagrams of arrangements with and without overlapping patterns; 9(a) is pattern 1; 9(b) is pattern 2; 9(c) is an arrangement with two overlapping patterns; and 9(d) is an arrangement with two non-overlapping patterns.

[0034] Figure 10 Schematic diagram of the occupied area of ​​the pattern extension area and the final optimization result of the pattern position; 10(a) is the random position of the pattern on the plate image; 10(b) is the corresponding extension area position; 10(c) is the overlapping area in the extension area; 10(d) is the optimization result of the pattern extension area position; 10(e) is the corresponding optimized pattern position;

[0035] Figure 11 Schematic diagram of the search and optimization process for the first set of patterns; 11(a) is the optimized position of the first pattern in the first set; 11(b) is the optimized position of the second pattern in the first set; 11(c) is the optimized position of the third pattern in the first set; 11(d) is the optimized position of the fourth pattern in the first set; 11(e) is the finalized first pattern placed on the board;

[0036] Figure 12 Schematic diagram of the search and optimization process for the second set of patterns; 12(a) shows the optimized position of the first pattern in the second set; 12(b) shows the optimized position of the second pattern in the second set; 12(c) shows the optimized position of the third pattern in the second set; 12(d) shows the first two patterns finally determined and placed on the board;

[0037] Figure 13 Schematic diagram of the search and optimization process for the third set of patterns; 13(a) shows the optimized position of the first pattern in the third set; 13(b) shows the optimized position of the second pattern in the third set; 13(c) shows the first three patterns finally determined and placed on the board;

[0038] Figure 14Schematic diagram of placing a third pattern in the occupied area; 14(a) is the width occupied by the first two patterns on the plate diagram; 14(b) is the permitted range for placing the third pattern in the occupied area;

[0039] Figure 15 Schematic diagram for expanding the width of the search area;

[0040] Figure 16 Schematic diagram comparing the results of iterative and non-iterative overlapping optimization arrangements; 16(a) is the iterative overlapping optimization arrangement result of the first group of patterns; 16(b) is the non-iterative overlapping optimization arrangement result of the first group of patterns; 16(c) is the iterative overlapping optimization arrangement result of the second group of patterns; 16(d) is the non-iterative overlapping optimization arrangement result of the second group of patterns. DETAILED DESCRIPTION

[0041] The following is combined with Figure 1-16 The present invention is further described.

[0042] The present invention adopts the following technical solution: a two-dimensional parts arrangement method based on dynamic iteration and optimization of the extended area overlap rate, wherein the arrangement process mainly includes the following steps:

[0043] A two-dimensional parts arrangement method based on dynamic iteration and extended area overlap rate optimization is characterized in that the arrangement process mainly includes the following steps:

[0044] S1. Obtain a part image based on the CAD model of the part and perform binarization on it, where black is the background and white is the part area;

[0045] S2. Get the horizontal dimension L of the part h and vertical dimension L v , and take the corresponding size as a reference to obtain the corresponding standard rectangular image;

[0046] S3, based on the lateral dimension L of the part h and vertical dimension L v The sum L t Sort the part images from large to small as a basis;

[0047] S4. Expand the part image for the first time according to the clearance requirements of the part;

[0048] S5. Calculate the width w2 of the corresponding expansion area according to the size of the part, and perform a second expansion on the part graphic;

[0049] S6. Based on a predetermined value N max And the corresponding size principle, select the first N part images as a group of images;

[0050] S7, the plate image is expanded to be consistent with the width w2 of the second expansion area of ​​the current image;

[0051] S8. Based on the improved genetic algorithm, the position of each image in the set of part images on the plate image is encoded, mutated, and decoded, and the overlap rate of the second expanded area of ​​the part image, the expanded area of ​​the plate image, and the arranged images is optimized as the optimization target;

[0052] S9, selecting the image with the highest overlap rate and arranging it on the plate image according to the specified optimized position, and removing the image from the previous image sequence;

[0053] S10. Repeat the above steps S6-9 until all parts are arranged.

[0054] In step S5, the corresponding size principle is that the size L of the last part, that is, the smallest part, in each group of part images selected is t It cannot be smaller than α times the size of the first part, which is the largest part.

[0055] In step S5, after each image is arranged, a new group of images needs to be selected and the overlap rate of the extended area after the position of each image in the group of images is optimized needs to be recalculated.

[0056] Image preprocessing

[0057] (1) Image cropping and resizing

[0058] The corresponding two-dimensional part image is extracted from the DWG file generated by AutoCAD software based on the C++ programming language. Then, based on the color difference between the target object and the background in the image, the image is binarized as follows: Figure 1 As shown in (a).

[0059] Usually, the corresponding part image often does not accurately reflect the real size of the object. Therefore, it is necessary to resize the image proportionally. The process begins by trimming the edges of the image, such as Figure 1 (b) and 1(c). The image is then resized to accurately match the actual size of the workpiece.

[0060] After resizing the image to accurately match the actual size of the part, it is then pasted onto another image with a black background.

[0061] (2) Image expansion

[0062] In the process of arranging parts, the gap between them is crucial because the tool has a certain size during the cutting stage, so some material will be removed. In order to solve this problem, the part image is expanded accordingly based on formula (1).

[0063] Here, "g" represents the gap distance between different parts.

[0064]

[0065] Here, the layout accuracy is set to 0.5mm, so each pixel will correspond to a size of 0.5mm. In addition, the part gap is set to a gap of 4mm. The image expansion process is as follows Figure 2 shown.

[0066] exist Figure 2 In (a), white pixels represent the target, gray pixels represent the pixels within the region, and black pixels represent the background. When the position of any pixel satisfies equation (1), the pixel will be converted to white. The expanded image is shown in Figure 2 (b) in.

[0067] The expanded image is Figure 3 As shown in (a), the expansion area is indicated in red. The basic principle of pattern layout optimization is mainly based on the overlap ratio of its later expansion area, which includes the overlap with the expansion areas of other patterns and the plate image. This concept will be discussed more fully later. Therefore, it is necessary to further expand the part image based on the first expansion. Equation (2) is the definition of the expansion area width for subsequent expansion:

[0068] w2=round[(L length +L width ) / 13] (2)

[0069] Among them, L length and L width Represent the length and width of the part image, respectively. As the pattern size decreases, the width w2 of the extended area decreases accordingly. The minimum value of w2 is set to 15, which can be expressed as follows:

[0070] w2=max(w2,15) (3)

[0071] This expression ensures that the part image retains enough area for layout optimization calculations.

[0072] The first stage of image expansion, such as Figure 3 (a) shows the expansion area highlighted in red. The next stage of expansion is as follows Figure 3 (b), where the expanded area is shown in green. In the second stage of expansion, Figure 3 As shown in (c), it is specifically used for the calculation and optimization process of the later arrangement of images.

[0073] like Figure 4As shown in (a), the plate image also needs to be expanded, and the width of its expanded area is the same as that of the corresponding part image, as shown in Figure 4 (b) shows the black area representing the sheet material, and the green area represents the expansion area. Although this expansion area does not physically exist, it is used specifically for optimizing the pattern placement process. Because the width of the expansion area varies between different patterns, expansion of the sheet material image should be performed before optimizing the placement of each part image.

[0074] The arrangement of part images basically follows the order of size from large to small. Here, L h and L v Represents the horizontal and vertical sizes respectively, such as Figure 5 shown.

[0075] All part images specified for arrangement are arranged according to their size L t To sort, the definition is as follows:

[0076] L t =L h +L v (4)

[0077] Improved genetic algorithm

[0078] (1) Coding

[0079] A genetic algorithm (GA) is a search technique that simulates natural selection and heredity to find solutions to optimization challenges. It begins with a set of random solutions and systematically uses selection, crossover, and mutation principles to evolve new sets of solutions, aiming to identify the optimal solution over multiple iterations. Here, a modified genetic algorithm was used to determine the optimal spatial arrangement of parts. The positional characteristics of these patterns in the image are defined by three parameters: x-coordinate, y-coordinate, and rotation angle. This data was initially encoded in a binary system.

[0080] The binary code consists of 27 bits in total. The first 12 bits are used to represent the x-coordinate of the pattern, followed by bits 13 to 15 to specify the rotation angle of the pattern. Subsequently, bits 16 to 27 are used to represent the y-coordinate of the component. This encoding process is Figure 6 There are pictures in it.

[0081] (2) Variation

[0082] In mutation, the value of a specific binary bit is switched from 0 to 1, or vice versa. max The encoding of a given sample is labeled SC, where |SC|1 represents the number of "1"s in the encoding SC and |SC|0 represents the number of "0". Figure 6, we can draw the following insights:

[0083] |SC|1+|SC|0=27 (5)

[0084] Nr max The value of is randomly determined to be |SC|1 or |SC|0, as shown in the following formula:

[0085] Nr max ∈{|SC|1,|SC|0} (6)

[0086] The number of bits that need to be mutated is represented by Nr, and its maximum allowed value is Nr max The exact value of Nr is chosen to be between 0 and Nr max In the context of , this relationship can be described as follows:

[0087] Nr∈{0,1,2,...,Nr max} (7)

[0088] In addition, a random variable r is used to determine the exact bit to be modified during the mutation phase. The value of r is set to fall within the interval 1 to 27 and can be expressed as follows:

[0089] r∈{1,2,3,...,27} (8)

[0090] (3) Decoding

[0091] The x coordinate of the part (x i ) and y coordinate (y i ) must be within the specified range to ensure that the part is positioned on the plate, which can be expressed as:

[0092]

[0093] Among them, L x and L y represent the size of the search area along the X-axis and Y-axis respectively. x and l y Respectively represent the part size along the X-axis and Y-axis, such as Figure 7 shown.

[0094] Considering that a 12-bit binary number can represent 2^12 = 4096 different information combinations, and a 3-bit binary number has 2^3 = 8 different information combinations, the spatial properties of the pattern are defined by a 27-bit binary sequence as follows:

[0095] x i =BIN(1:12)·(L x -l x ) / 4096 (10)

[0096] y i =BIN(16:27)·(L y -l y ) / 4096 (11)

[0097] θ=BIN(13:15)·45 (12)

[0098] Here, "BIN" refers to assigning a 27-bit binary code to each sample, which concisely captures the digital attributes associated with each sample.

[0099] Iterative Overlap Optimization Placement (IOOP)

[0100] Figure 8 The detailed flow chart of the corresponding parts arrangement principle details the key steps of the search method used in the present invention. Figure 7 The patterns are sorted by the pattern size shown in . Then, the patterns are sorted by the set value N max Using the specific size criteria in Equation (18), an initial set of N patterns is selected. The placement of each pattern is optimized using the modified genetic algorithm outlined in this paper. These N patterns are then evaluated and compared, and the pattern with the largest overlap ratio is selected for placement. The remainder of this paper will explore each step presented in the flowchart in detail, fully exploring the algorithmic strategy implemented for optimization.

[0101] (1) Overlap detection

[0102] During the placement process, it is crucial to avoid overlaps between part patterns. In this regard, we propose a simple yet effective technique that utilizes the total number of white pixels in the image as a mechanism to identify potential overlaps.

[0103] Formula (13) establishes the conditions that need to be met:

[0104] N white_total =N white1 +N white2 (13)

[0105] Among them, N white_total is the cumulative count of white pixels in the merged image IMG, and N white1 and N white2 Reflects the number of white pixels in images IMG1 and IMG2 respectively. The calculation of these values ​​is described as follows:

[0106]

[0107]

[0108]

[0109] By comparing the white pixel counts, we can identify whether there is overlap between patterns. If Equation (13) is not satisfied, it means that there is overlap between patterns and they should be excluded from consideration.

[0110] The target object of the pattern is represented by white pixels, and the background pixels are black. The combined image is represented as IMG, as shown in (17):

[0111] IMG=IMG1|IMG2 (17)

[0112] The whole process Figure 9 It is important to visualize the Figure 9 (c) shows the Figure 9 The resulting pattern after combining the patterns shown in (a) and 9(b).

[0113] Figure 9 (c) clearly shows the overlap between the two patterns, with the overlap marked in red. Figure 9 (d) shows two patterns without any overlap, which are clearly separated from each other and thus satisfy the condition of equation (13).

[0114] (2) Overlap optimization

[0115] For pattern placement, it is initially crucial to ensure that the entire pattern is contained within the board boundaries. This requirement is confirmed by equations (10)-(12), which guarantee that for the x-coordinate (x c ), y coordinate (y c ) and any value of the rotation angle that is obtained complies with this constraint. Second, as previously emphasized, it is crucial to avoid overlap between patterns. The theory guiding pattern position optimization focuses on the overlap ratio of the pattern extension area. This extension area is determined by Figure 3 The green area shown in (b) indicates.

[0116] Figure 10 (a) shows a random placement of a pattern on a plate image, which is consistent with Figure 10 The location of the extended area shown in (b) corresponds to the location of the extended area of ​​the pattern. For the overlapping area between the extended area of ​​the pattern and the extended area of ​​the plate (the green area at the edge of the plate image), as shown in Figure 10 (c) is highlighted in red. Clearly, this position is not the optimal result. After five consecutive attempts using the modified genetic algorithm described previously failed to find a better result, the position at this stage was considered the optimal placement. Figure 10(d) shows the final optimized position of the pattern’s extended region, which is consistent with Figure 10 The positions of the patterns shown in (e) correspond to each other, thereby maximizing the ratio of the overlapping area.

[0117] (3) Arrangement of the first pattern

[0118] The order of pattern arrangement depends on two key factors: the size of the pattern and the overlap ratio of the pattern extension area. Initially, the order of the patterns is determined based solely on the size of the pattern. However, the overlap ratio of the first pattern may not be the highest. To account for this discrepancy, the first Nmax (here 10) patterns in the sequence are selected, and any pattern that is smaller than α (here 0.75) times the size of the first pattern is excluded. The corresponding conditional expression is:

[0119]

[0120] According to formula (18), only the first four patterns are finally selected. The positions of these selected patterns are optimized using the improved genetic algorithm, as shown in Figure 11 (a)-(d) are shown. Among them, the extended area of ​​the third pattern is determined to have the highest overlapping area ratio, as shown in Figure 11 (c). This corresponds to Figure 11 The position of the pattern shown in (e) is determined to be the first pattern placed on the board.

[0121] (4) Arrangement of subsequent patterns

[0122] So far, only the first pattern has been placed. Patterns already placed on the board will be removed from the pattern sequence. Using the same selection criteria, another set of patterns will be selected; from this set, three patterns will be selected for further evaluation and placement.

[0123] Before placing the initial pattern, the plate image remains unoccupied and the extension area of ​​the part image can only overlap with the extension area of ​​the plate. However, for the patterns arranged subsequently, their extension areas can overlap not only with the extension area of ​​the plate, but also with the part patterns already arranged on the plate. The overlapping area is shown in red, as shown in the following example. Figure 12 As shown in (a)-(c).

[0124] final, Figure 12 The extended region of the pattern shown in (c) is determined to have the highest overlap ratio. This corresponds to Figure 12 The position of the pattern shown in (d) completes the arrangement of the second pattern.

[0125] Using the same method, Figure 13 The arrangement process of the third pattern is shown, and the final result of the arrangement is presented in Figure 13 In (c), this includes the arrangements of the first three patterns.

[0126] (5) Layout principle from left to right

[0127] The process of pattern arrangement follows the principle of going from left to right, starting from the occupied area, such as Figure 14 (a). The currently occupied area is the portion to the left of the red dashed line. When placing the third pattern, it is important to place it within these predefined boundaries to avoid increasing the width already used. Figure 14 As shown in (b), the yellow pattern is the leftmost boundary that can be placed, and the aqua pattern on the right is the rightmost boundary that can be placed. This method ensures that the third pattern fits the specified L occupied within a limited area, thereby making efficient use of space without expanding the occupied area.

[0128] according to Figure 14 (b), the formula for calculating the width of the third pattern search area is as follows:

[0129] L range =L occupied -L width (19)

[0130] The x-coordinate for placing the third pattern is determined by the following expression:

[0131] x3=BIN(1:12)·(L occupied -L width ) / 4096 (20)

[0132] Sometimes you can occupied Find the appropriate position of the third pattern within the defined width, such as Figure 14 (a). However, sometimes it is difficult. Figure 15 As shown, within the improved genetic algorithm framework, L occupied The constraints may make it difficult to find an ideal location for the current pattern, but opportunities for appropriate placement may still exist.

[0133] To overcome this difficulty, the overlapping area of ​​all samples is determined, and the location with the least overlap is considered the most suitable location for placing the pattern. Therefore, the codes of all samples are updated to reflect the code of the sample with the least overlap. Subsequently, the mutation operation is performed on these codes. If a pattern position is found that does not overlap with any other pattern, then the mutation operation is initiated. Figure 8 However, in some cases, finding a suitable location for the pattern is still challenging. In this case, the width of the search area specified by the current pattern will be extended by L width, which corresponds to the width of the pattern, as Figure 15 shown.

[0134] As the occupied width expands, increasing the number of samples in the genetic algorithm is crucial to improving the chances of finding the best position of the pattern. The present invention calculates the required number of samples according to the given formula:

[0135]

[0136] Among them, the symbol Represents the mathematical ceiling function, which is used to round a number up to the nearest integer.

[0137] (6) Pseudocode of the drawing framework

[0138] The pseudo code of the corresponding arrangement algorithm is as follows:

[0139]

[0140]

[0141] Sorting results

[0142] In order to highlight the superiority of the iterative overlap optimization arrangement (IOOP) method proposed in this study, Figure 16 A comparative analysis between IOOP and non-iterative overlapping optimization placement methods is presented.

[0143] It can be seen that compared with the arrangement method of pattern order based solely on size, the dynamic iteration process of IOOP can further improve the utilization rate of the board.

[0144] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to the specific embodiments described. Obviously, many modifications and variations are possible based on the content of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A two-dimensional parts arrangement method based on dynamic iteration and optimization of the overlap rate of the extended area, characterized in that: The arrangement process mainly includes the following steps: S1. Obtain a part image according to the CAD model of the part and perform binarization on it, where the black is the background and the white is the part area; S2. Get the lateral dimension L of the part h And vertical dimension L v , and taking the corresponding size as a reference, obtain the corresponding standard rectangular image; S3, based on the lateral dimension L of the part h And vertical dimension L v The sum of L t As a basis, the part images are sorted from large to small; S4, expanding the part image for the first time according to the clearance requirements of the parts; S5, calculating the width w2 of the corresponding expansion area according to the size of the part, and performing a second expansion on the graphics of the part; S6. Based on a predetermined value N max and the corresponding size principle, select the first N part images as a group of images; S7, the expansion of the plate image is consistent with the width w2 of the second expansion area of ​​the current image; S8. Based on the improved genetic algorithm, the position of each image in the group of part images on the plate image is encoded, mutated, and decoded, and the optimization is performed with the overlap rate of the second extended area of ​​the part image and the extended area of ​​the plate image and the arranged images as the optimization target; S9, selecting an image with the highest overlap rate and arranging it on the plate image according to a specified optimized position, and removing the image from the previous image sequence; S10, repeat the above steps S6-9 until all parts are arranged.

Images