Intelligent pattern layout optimization method considering quality zoning constraints

Abstract

A smart pattern layout optimization method considering quality zoning constraints includes the following steps: S1, data preprocessing; S2, vector grouping, where the layout is performed on a two-dimensional plane, and the vector grouping consists of two binary basis vectors; the core idea of ​​solving these two binary basis vectors is to find the minimum displacement vector that ensures two pattern pieces do not overlap at their junctions; S3, initial solution generation, solving for the pattern piece displacement vectors based on the vector grouping and generating a temporary layout; calculating the convex hull region based on the temporary layout, and identifying potential insertion spaces through the convex hull region; S4, remaining pattern piece layout, after the initial solution is generated, if there are still unplaced pattern pieces, a secondary optimization layout is needed based on the current initial solution; this method can effectively solve the problem of layout on irregular leather, thereby significantly improving material utilization and production efficiency.

Description

Technical Field

[0001] This invention relates to the field of two-dimensional material nesting technology, and in particular to an intelligent cut piece nesting optimization method that considers quality zoning constraints. Background Technology

[0002] The intelligent leather pattern layout optimization method considering quality zoning constraints has significant engineering value in modern manufacturing. For example, it is needed in high-end leather goods customization, automotive seat upholstery, and luxury furniture manufacturing to optimize layout and improve material utilization. The natural properties of raw materials and the functional differences of product components create a dual constraint: on the one hand, exposed decorative parts must be strictly matched with flawless, highly ductile areas to maintain aesthetic quality, while load-bearing structural parts require corresponding high tear-resistant areas to ensure physical performance; on the other hand, the growing trend of personalized customization forces companies to dynamically adapt to the production needs of multiple varieties and small batches on limited raw material slabs. Therefore, many industries urgently need to solve the leather pattern layout problem considering quality zoning constraints.

[0003] In today's manufacturing industry, low material utilization and inefficient pattern making remain prevalent problems. For example, in the garment manufacturing industry, due to the irregularity and complexity of leather, traditional pattern making methods often fail to fully utilize each piece of leather, leading to significant material waste. This not only increases raw material costs but also imposes an unnecessary burden on the environment. In the furniture manufacturing industry, leather is widely used in the production of sofas, chairs, beds, and other furniture. Different products have varying quality requirements for leather, and the quality of different areas on leather sheets varies significantly. High-quality areas need to be used to produce high-value products to maximize profits. Therefore, such pattern making problems are widespread across various industries and have become critical issues that urgently need to be addressed. Despite the different application scenarios, the core issue remains the same: how to efficiently arrange the required pieces of leather according to rules on limited materials, maximizing the utilization of each piece of leather, reducing waste, and improving material utilization to achieve resource conservation and optimization. Summary of the Invention

[0004] To address the aforementioned shortcomings, the present invention aims to propose a smart leather cutting pattern optimization method that considers quality zoning constraints. This method can effectively solve the problem of patterning on irregular leather, thereby significantly improving material utilization and production efficiency.

[0005] To achieve this objective, the present invention adopts the following technical solution:

[0006] The intelligent pattern layout optimization method considering quality zoning constraints includes the following steps:

[0007] S1. Data preprocessing, including obtaining the overall size information of the leather, the size and location distribution of each grade area, and the size, quantity and grade information of irregular pieces, and sorting all irregular pieces according to their reference size and grade information;

[0008] S2, Grouping Vector: The layout is carried out on a two-dimensional plane. The grouping vector consists of two binary basis vectors. The core idea of ​​solving these two binary basis vectors is to find the minimum displacement vector that ensures that the two pieces do not overlap at the junction.

[0009] S3. Initial solution generation: Solve for the piece displacement vector based on the grouping vector and generate a temporary layout; calculate the convex hull region based on the temporary layout and identify potential insertion spaces through the convex hull region; the initial remaining pieces are combined with the fitness function to find the optimal placement point in the convex hull region for insertion. If there are still unplaced initial remaining pieces after the insertion attempt, jump to step one to continue execution; the initial remaining pieces are inserted in this order. When all initial remaining pieces have completed the placement attempt or there is no placement area, the initial solution generation process ends.

[0010] S4. Remaining piece layout: After the initial solution is generated, if there are still unplaced remaining pieces, it is necessary to select individual remaining pieces in the final set of remaining pieces in order based on the layout of the current initial solution and select their placement positions in combination with the fitness function; perform secondary optimization layout of the remaining pieces.

[0011] Furthermore, step S2, in the solution process, includes the following steps:

[0012] S21, First, fix the first piece of fabric in sequence, and use the NFP geometric tool to calculate the possible placement position of the second piece of fabric;

[0013] S22, then find the minimum distance to move the second piece of fabric horizontally to the right from the position of the first piece of fabric to a non-overlapping position, and you can get the vector v1 = (x1, 0);

[0014] S23. Finally, the set of possible placement points for the third piece of fabric is calculated using NFP. The point that does not overlap with the first and second pieces of fabric and has the smallest displacement is selected from the set. This gives us the second vector v2 = (x2, y), which is the grouped vector v = (v1, v2).

[0015] Furthermore, step S3 specifically includes the following steps:

[0016] S31. Select the corresponding cut piece according to the pattern arrangement order, and calculate the non-overlapping constraint space between the cut piece and the already arranged area through NFP to generate a set of geometrically feasible candidate placement points. In this process, the lower left placement strategy is adopted: prioritize the selection of the point with the smallest horizontal coordinate, and if there are multiple points, further select the point with the smallest vertical coordinate.

[0017] S32, combined with the grouping vector, calculate the subsequent placement position of the cut piece using formula (1) until the remaining number of cut pieces is zero or there is no place to place them.

[0018] v = i*v1 + j*v2 Equation (1);

[0019] Where v is the displacement vector of the cut piece, and i and j are both integers;

[0020] S33. The temporary layout generated based on the grouping vector may have underutilized space. Therefore, it is necessary to calculate the convex hull region formed by all placed pieces to reflect the space occupancy of the current layout and provide a basis for optimizing space utilization. The construction of the convex hull region first selects the point with the smallest x-coordinate from all placed pieces as the starting point. If there are multiple points with the same x-coordinate, the point with the smallest y-coordinate is selected from the point set. Then, starting from the current point, all vertices are traversed to find the point with the smallest polar angle connecting it to the current point as the next convex hull point. This process is iterated until the starting point is returned, and the convex hull is finally constructed. Through the convex hull region, potential insertion space can be identified, laying a geometric foundation for the subsequent placement and layout optimization of pieces, thereby improving space utilization and layout efficiency.

[0021] S34. Select the initial remaining pieces in the order of insertion and find the optimal placement point in the convex hull region using the fitness function. The introduction of the fitness function ensures that the placement of the initial remaining pieces can make the most of the remaining space while avoiding overlap with already placed pieces. This process continues until all remaining pieces have completed the insertion attempt.

[0022] S35, if there are still unplaced initial remaining pieces after completing the above steps, jump to the first step to continue execution; when all initial remaining pieces have completed placement attempts or there is no placeable area, the initial solution generation process ends.

[0023] Preferably, step S4 specifically includes the following steps:

[0024] S41. According to the arrangement order of the remaining pieces, select the remaining pieces from the last set of remaining pieces, calculate the candidate placement point set of the remaining pieces in its placeable area, and select the placement point with the highest fitness according to formula (2) for placement; when placing, the remaining pieces may overlap with the area to be arranged.

[0025]

[0026] Where A ij k represents the overlapping area of ​​the minimum envelope rectangles of pattern pieces i and j; i ,k j k represents the area of ​​the smallest envelope rectangle of the cut piece divided by the actual area of ​​the cut piece; ij This represents the distance between the centers of gravity of cut piece i and cut piece j divided by the distance between their center of reference; A im γ represents the area of ​​the smallest envelope rectangle of the outer contour of cut piece i outside the fabric; γ is a constant coefficient (0–2.0). Adjusting the coefficient γ can produce different nesting results. γ represents the weight of the cut piece's fit with the fabric boundary. The larger the value of γ, the more likely the position closer to the fabric boundary will be selected.

[0027] S42. Calculation of the overlap value between the remaining piece and the placement position. The calculation of the overlap value is achieved through the following steps: Traverse all edges of the two overlapping polygons of the placement position and the remaining piece polygon, calculate the intersection points of the edges and store them in a temporary array, identify the points in the two polygons to form a new intersecting polygon, and finally use the shoelace theorem formula to calculate the area of ​​the intersecting polygon. This area is the overlap value between the polygon of the placement position and the remaining piece polygon. By traversing all the polygons of the placement position and the remaining piece polygon and repeating the above calculation, the overlap value of the current layout can be obtained.

[0028] S43. Exchange and Move Operations: To reduce the overlap value to 0 and obtain the optimal layout, exchange and move operations are used for optimization. The exchange operation randomly selects two pieces of the same grade to swap positions. In the initial solution, pieces that overlap with the remaining pieces are swapped. In the current layout, other pieces of the same grade as the overlapping pieces are selected and swapped to reduce or eliminate the overlap value. If the selected remaining pieces still overlap after the exchange, the overlapping pieces are moved to a placeable area in the current layout other than their original positions, thus minimizing the overlap value. The exchange and move operations are repeated until the number of operations exceeds the limit or the overlap value of the current layout is 0.

[0029] S44. If all remaining pieces have been placed, the algorithm ends and outputs the final solution; otherwise, jump to S41 and continue execution until all remaining pieces have been placed.

[0030] Preferably, in step S1, in terms of size information processing, it is first necessary to accurately record the input leather size information, the size and position information of each grade area on the leather, and the size, quantity and grade information of irregular pieces. This size information is usually represented in the form of a set of points stored in counterclockwise order; the area of ​​these polygonal areas can be calculated using the shoelace theorem formula.

[0031] Based on the correspondence between the cut piece grade and the leather area grade, the placement area for each cut piece can be further calculated.

[0032] The shoelace theorem formula is:

[0033] In terms of grouping and sorting, all pieces are first sorted from highest to lowest grade; if two pieces are of the same grade, their areas are further compared, and the piece with the larger area is prioritized.

[0034] Ideally, the combination of exchange and move operations constitutes a variable neighborhood search algorithm.

[0035] Preferably, the step also includes generating the cutting path:

[0036] S5: Extract the independent contour of each part, including the outer contour and internal holes, calculate the geometric features of each contour to prepare for subsequent path planning, and initialize the list of unprocessed contours.

[0037] S6, take the contour closest to the origin as the starting cutting contour, take the lower left corner of the piece as the cutting point, and remove the contour from the list of unprocessed contours.

[0038] S7: Select the piece of fabric closest to the cutting point from the list of unprocessed outlines as the pre-cut piece of fabric, and remove the outline from the list of unprocessed outlines.

[0039] S8. If the unprocessed outline list is not empty, continue to step 3; if it is empty, the algorithm ends.

[0040] One of the above technical solutions includes the following beneficial effects: This solution solves the problem of irregular leather layout in multiple regions. By optimizing the division of multiple regions and the layout strategy, it improves the utilization rate of raw materials and makes the layout scheme more in line with actual production needs. In addition, this solution fully considers the requirements of the cut pieces for leather materials during the layout process, improving the leather utilization rate while ensuring product quality. The cutting layout optimization method proposed in this solution reduces the reliance on manual layout. Through intelligent algorithms and automation technology, it reduces manual intervention, improves layout efficiency and raw material utilization, and promotes the automation process of enterprise production. Attached Figure Description

[0041] Figure 1This is a schematic diagram of the intelligent leather cutting pattern optimization method considering quality zoning constraints of the present invention;

[0042] Figure 2 This is a flowchart of the intelligent leather cutting pattern optimization method considering quality zoning constraints of the present invention;

[0043] Figure 3 This is a flowchart of the initial solution generation process of the present invention;

[0044] Figure 4 This is a flowchart of the remaining cut pieces layout process of the present invention;

[0045] Figure 5 This is a layout diagram of Case 1;

[0046] Figure 6 This is a layout diagram for Case 2. Detailed Implementation

[0047] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0048] like Figure 1-2 As shown, the intelligent pattern layout optimization method considering quality zoning constraints includes the following steps:

[0049] S1. Data preprocessing, including obtaining the overall size information of the leather, the size and location distribution of each grade area, and the size, quantity and grade information of irregular pieces, and sorting all irregular pieces according to their reference size and grade information;

[0050] S2, Grouping Vector: The layout is carried out on a two-dimensional plane. The grouping vector consists of two binary basis vectors. The core idea of ​​solving these two binary basis vectors is to find the minimum displacement vector that ensures that the two pieces do not overlap at the junction.

[0051] S3. Initial solution generation: Solve for the piece displacement vector based on the grouping vector and generate a temporary layout; calculate the convex hull region based on the temporary layout and identify potential insertion spaces through the convex hull region; the initial remaining pieces are combined with the fitness function to find the optimal placement point in the convex hull region for insertion. If there are still unplaced initial remaining pieces after the insertion attempt, jump to step one to continue execution; the initial remaining pieces are inserted in this order. When all initial remaining pieces have completed the placement attempt or there is no placement area, the initial solution generation process ends.

[0052] S4. Remaining piece layout: After the initial solution is generated, if there are still unplaced remaining pieces, it is necessary to select individual remaining pieces in the final set of remaining pieces in order based on the layout of the current initial solution and select their placement positions in combination with the fitness function; perform secondary optimization layout of the remaining pieces.

[0053] Beneficial effects: This solution solves the problem of irregular leather layout in multiple regions. By optimizing the division of multiple regions and the layout strategy, it improves the utilization rate of raw materials and makes the layout plan more in line with actual production needs. In addition, this solution fully considers the requirements of the cut pieces for leather material during the layout process, improving the utilization rate of leather while ensuring product quality. The cutting layout optimization method proposed in this solution reduces the reliance on manual layout. Through intelligent algorithms and automation technology, it reduces human intervention, improves layout efficiency and raw material utilization, and promotes the automation process of enterprise production.

[0054] like Figure 2 As shown, step S2 in the solution process includes the following steps:

[0055] S21, First, fix the first piece of fabric in sequence, and use the NFP geometric tool to calculate the possible placement position of the second piece of fabric;

[0056] S22, then find the minimum distance to move the second piece of fabric horizontally to the right from the position of the first piece of fabric to a non-overlapping position, and you can get the vector v1 = (x1, 0);

[0057] S23. Finally, the set of possible placement points for the third piece of fabric is calculated using NFP. The point that does not overlap with the first and second pieces of fabric and has the smallest displacement is selected from the set. This gives us the second vector v2 = (x2, y), which is the grouped vector v = (v1, v2).

[0058] Beneficial Effects: This process not only ensures the geometric feasibility of the pattern piece arrangement but also lays a solid foundation for the efficient optimization of subsequent layout. The pattern piece arrangement is a key factor in achieving neat pattern piece arrangement. By calculating the minimum non-overlapping displacement between pattern pieces, the pattern piece arrangement can effectively guide the arrangement, thereby ensuring space utilization while avoiding overlap between pattern pieces, providing important support for layout optimization. This is achieved through the No-Fit Polygon (NFP).

[0059] like Figure 3 As shown, step S3 specifically includes the following steps:

[0060] S31. Select the corresponding cut piece according to the pattern arrangement order, and calculate the non-overlapping constraint space between the cut piece and the already arranged area through NFP to generate a set of geometrically feasible candidate placement points. In this process, the lower left placement strategy is adopted: prioritize the selection of the point with the smallest horizontal coordinate, and if there are multiple points, further select the point with the smallest vertical coordinate.

[0061] S32, combined with the grouping vector, calculate the subsequent placement position of the cut piece using formula (1) until the remaining number of cut pieces is zero or there is no place to place them.

[0062] v = i*v1 + j*v2 Equation (1);

[0063] Where v is the displacement vector of the cut piece, and i and j are both integers;

[0064] S33. The temporary layout generated based on the grouping vector may have underutilized space. Therefore, it is necessary to calculate the convex hull region formed by all placed pieces to reflect the space occupancy of the current layout and provide a basis for optimizing space utilization. The construction of the convex hull region first selects the point with the smallest x-coordinate from all placed pieces as the starting point. If there are multiple points with the same x-coordinate, the point with the smallest y-coordinate is selected from the point set. Then, starting from the current point, all vertices are traversed to find the point with the smallest polar angle connecting it to the current point as the next convex hull point. This process is iterated until the starting point is returned, and the convex hull is finally constructed. Through the convex hull region, potential insertion space can be identified, laying a geometric foundation for the subsequent placement and layout optimization of pieces, thereby improving space utilization and layout efficiency.

[0065] S34. Select the initial remaining pieces in the order of insertion and find the optimal placement point in the convex hull region using the fitness function. The introduction of the fitness function ensures that the placement of the initial remaining pieces can make the most of the remaining space while avoiding overlap with already placed pieces. This process continues until all remaining pieces have completed the insertion attempt.

[0066] S35, if there are still unplaced initial remaining pieces after completing the above steps, jump to the first step to continue execution; when all initial remaining pieces have completed placement attempts or there is no placeable area, the initial solution generation process ends.

[0067] Beneficial effects: By employing the lower-left placement strategy, the cut pieces can be precisely positioned in the local optimum of the layout area, ensuring maximum space utilization. Simultaneously, placing the cut pieces based on the grouping vector results in a more compact layout. For the gaps in the temporary layout obtained based on the grouping vector, a convex hull insertion strategy is used to fill the gaps, further improving the overall layout. This insertion strategy is strictly limited to the feasible region defined by the convex hull boundary, ensuring it does not interfere with subsequent grouping cut piece placement. The initial solution obtained through S3 is very compact overall and satisfies the leather quality constraint, providing a guarantee for subsequent iterative optimization.

[0068] like Figure 4 As shown, step S4 specifically includes the following steps:

[0069] S41. According to the arrangement order of the remaining pieces, select the remaining pieces from the last set of remaining pieces, calculate the candidate placement point set of the remaining pieces in its placeable area, and select the placement point with the highest fitness according to formula (2) for placement; when placing, the remaining pieces may overlap with the area to be arranged.

[0070]

[0071] Where A ij k represents the overlapping area of ​​the minimum envelope rectangles of pattern pieces i and j; i ,k j k represents the area of ​​the smallest envelope rectangle of the cut piece divided by the actual area of ​​the cut piece; ij This represents the distance between the centers of gravity of cut piece i and cut piece j divided by the distance between their center of reference; A im γ represents the area of ​​the smallest envelope rectangle of the outer contour of cut piece i outside the fabric; γ is a constant coefficient (0–2.0). Adjusting the coefficient γ can produce different nesting results. γ represents the weight of the cut piece's fit with the fabric boundary. The larger the value of γ, the more likely the position closer to the fabric boundary will be selected.

[0072] S42. Calculation of the overlap value between the remaining piece and the placement position. The calculation of the overlap value is achieved through the following steps: Traverse all edges of the two overlapping polygons of the placement position and the remaining piece polygon, calculate the intersection points of the edges and store them in a temporary array, identify the points in the two polygons to form a new intersecting polygon, and finally use the shoelace theorem formula to calculate the area of ​​the intersecting polygon. This area is the overlap value between the polygon of the placement position and the remaining piece polygon. By traversing all the polygons of the placement position and the remaining piece polygon and repeating the above calculation, the overlap value of the current layout can be obtained.

[0073] S43. Exchange and Move Operations: To reduce the overlap value to 0 and obtain the optimal layout, exchange and move operations are used for optimization. The exchange operation randomly selects two pieces of the same grade to swap positions. In the initial solution, pieces that overlap with the remaining pieces are swapped. In the current layout, other pieces of the same grade as the overlapping pieces are selected and swapped to reduce or eliminate the overlap value. If the selected remaining pieces still overlap after the exchange, the overlapping pieces are moved to a placeable area in the current layout other than their original positions, thus minimizing the overlap value. The exchange and move operations are repeated until the number of operations exceeds the limit or the overlap value of the current layout is 0.

[0074] S44. If all remaining pieces have been placed, the algorithm ends and outputs the final solution; otherwise, jump to S41 and continue execution until all remaining pieces have been placed.

[0075] By representing the overlap value using the overlapping area between fabric pieces, the overlap situation of the current layout can be displayed more intuitively. The coordinated application of exchange and move operations can effectively adjust the current layout, significantly reducing or eliminating the overlap value, transforming an invalid layout with overlap into a valid layout without overlap. Through the secondary nesting operation in S4, more fabric pieces can be placed on top of the initial solution, further improving the utilization rate of the nesting layout.

[0076] As a supplement to the above technology, in step S1, in terms of size information processing, it is first necessary to accurately record the input leather size information, the size and position information of each grade area on the leather, and the size, quantity and grade information of irregular cut pieces. This size information is usually represented in the form of a set of points stored in counterclockwise order; the area of ​​these polygonal areas can be calculated using the shoelace theorem formula.

[0077] Based on the correspondence between the cut piece grade and the leather area grade, the placement area for each cut piece can be further calculated.

[0078] The shoelace theorem formula is:

[0079] In terms of grouping and sorting, all pieces are first sorted from highest to lowest grade; if two pieces are of the same grade, their areas are further compared, and the piece with the larger area is prioritized.

[0080] The core idea of ​​the shoelace theorem is: regardless of how the origin of the coordinate system is chosen, as long as the coordinates A of each vertex of the polygon are input in a clockwise or counterclockwise direction, lThe area of ​​the polygon is half the sum of the absolute values ​​of the cross products of adjacent vertices. Based on this formula, the total area of ​​the leather, the area of ​​each grade area, and the total area of ​​the irregular pieces can be calculated.

[0081] Furthermore, based on the correspondence between the cut piece grade and the leather area grade, the placement area for each cut piece can be further calculated. Specifically, cut pieces can only be placed in areas where their grade is no higher than the leather area grade; that is, the leather grade in the area must be higher than or equal to the cut piece grade. This constraint ensures that high-grade cut pieces preferentially consume high-grade leather areas, thereby maximizing the utilization rate of high-grade leather and providing clear area restrictions for optimizing the layout.

[0082] This sorting strategy ensures that high-grade pieces are prioritized for use in high-grade leather areas, maximizing the utilization of high-grade leather. During the initial solution generation process, a similar grouping and sorting method is used for the insertion order: first, the remaining pieces are grouped, and then the pieces within each group are sorted by area from smallest to largest. This sorting method helps to prioritize the placement of smaller pieces during insertion, thus utilizing the remaining space more efficiently and improving the overall quality and space utilization of the layout.

[0083] Furthermore, in step S43, the combination of exchange and move operations constitutes the variable neighborhood search algorithm.

[0084] It can effectively optimize two-dimensional irregular layout, improve space utilization and layout effect.

[0085] In addition, it also includes the step of generating the cutting path:

[0086] S5: Extract the independent contour of each part, including the outer contour and internal holes, calculate the geometric features of each contour to prepare for subsequent path planning, and initialize the list of unprocessed contours.

[0087] S6, take the contour closest to the origin as the starting cutting contour, take the lower left corner of the piece as the cutting point, and remove the contour from the list of unprocessed contours.

[0088] S7: Select the piece of fabric closest to the cutting point from the list of unprocessed outlines as the pre-cut piece of fabric, and remove the outline from the list of unprocessed outlines.

[0089] S8. If the unprocessed outline list is not empty, continue to step 3; if it is empty, the algorithm ends.

[0090] Using the above method, the cutting path of the patterned pieces can also be planned, and the leather cutting head can efficiently complete the cutting work through the planned shortest cutting path.

[0091] Figure 5 and Figure 6 Examples 1 and 2 are from the following implementation case results.

[0092]

[0093] The technical principles of the present invention have been described above with reference to specific embodiments. These descriptions are merely for explaining the principles of the invention and should not be construed as limiting the scope of protection of the invention in any way. Based on this explanation, those skilled in the art can readily conceive of other specific embodiments of the invention without inventive effort, and these embodiments will all fall within the scope of protection of the present invention.

Claims

1. A smart pattern layout optimization method considering quality zoning constraints, characterized in that, Includes the following steps: S1. Data preprocessing, including obtaining the overall size information of the leather, the size and location distribution of each grade area, and the size, quantity and grade information of irregular pieces, and sorting all irregular pieces according to their reference size and grade information; S2, Grouping Vector: The layout is carried out on a two-dimensional plane. The grouping vector consists of two binary basis vectors. The core idea of ​​solving these two binary basis vectors is to find the minimum displacement vector that ensures that the two pieces do not overlap at the junction. S3. Initial solution generation: Solve for the piece displacement vector based on the grouping vector and generate a temporary layout; calculate the convex hull region based on the temporary layout and identify potential insertion spaces through the convex hull region; the initial remaining pieces are combined with the fitness function to find the optimal placement point in the convex hull region for insertion. If there are still unplaced initial remaining pieces after the insertion attempt, jump to step one to continue execution; the initial remaining pieces are inserted in this order. When all initial remaining pieces have completed the placement attempt or there is no placement area, the initial solution generation process ends. S4. Remaining piece layout: After the initial solution is generated, if there are still unplaced remaining pieces, it is necessary to select individual remaining pieces in the final set of remaining pieces in order based on the layout of the current initial solution and select their placement positions in combination with the fitness function; perform secondary optimization of the remaining pieces layout. Step S3 specifically includes the following steps: S31. Select the corresponding cut piece according to the pattern arrangement order, and calculate the non-overlapping constraint space between the cut piece and the already arranged area through NFP to generate a set of geometrically feasible candidate placement points. In this process, the lower left placement strategy is adopted: prioritize the selection of the point with the smallest horizontal coordinate, and if there are multiple points, further select the point with the smallest vertical coordinate. S32, combined with the grouping vector, calculate the subsequent placement position of the cut piece using formula (1) until the remaining number of cut pieces is zero or there is no place to place them; Equation (1); in, It is the displacement vector of the cut piece. and All are integers; S33. The temporary layout generated based on the grouping vector may have underutilized space. Therefore, it is necessary to calculate the convex hull region formed by all placed pieces to reflect the space occupancy of the current layout and provide a basis for optimizing space utilization. The construction of the convex hull region first selects the point with the smallest x-coordinate from all placed pieces as the starting point. If there are multiple points with the same x-coordinate, the point with the smallest y-coordinate is selected from the point set. Then, starting from the current point, all vertices are traversed to find the point with the smallest polar angle connecting it to the current point as the next convex hull point. This process is iterated until the starting point is returned, and the convex hull is finally constructed. Through the convex hull region, potential insertion space can be identified, laying a geometric foundation for the subsequent placement and layout optimization of pieces, thereby improving space utilization and layout efficiency. S34. Select the initial remaining pieces in the order of insertion and find the optimal placement point in the convex hull region using the fitness function. The introduction of the fitness function ensures that the placement of the initial remaining pieces can make the most of the remaining space while avoiding overlap with already placed pieces. This process continues until all remaining pieces have completed the insertion attempt. S35, if there are still unplaced initial remaining pieces after completing the above steps, jump to step one to continue execution; when all initial remaining pieces have been placed or there is no placeable area, the initial solution generation process ends; Step S2, in the solution process, includes the following steps: S21, First, fix the first piece of fabric in sequence, and use the NFP geometric tool to calculate the possible placement position of the second piece of fabric; S22, then calculate the minimum distance to move the second piece horizontally to the right from the position of the first piece to a non-overlapping position. The vector can then be obtained. ; S23. Finally, the set of possible placement points for the third piece of fabric is calculated using NFP. From this set, the point that does not overlap with either the first or second piece of fabric and has the minimum displacement is selected. This will give you the second vector. That is, grouping vectors .

2. The intelligent pattern layout optimization method considering quality zoning constraints according to claim 1, characterized in that, Step S4 specifically includes the following steps: S41. According to the arrangement order of the remaining pieces, select the remaining pieces from the last set of remaining pieces, calculate the candidate placement point set of the remaining pieces in their placement area, and select the fitness from them according to formula (2). Place the pieces at the highest possible point; when placing them, there may be overlap between the remaining pieces and the area to be arranged. Equation (2); in Indicates cut piece With cut pieces The overlapping area of ​​the smallest envelope rectangle; This represents the area of ​​the smallest envelope rectangle of the cut piece divided by the actual area of ​​the cut piece. Indicates cut piece With cut pieces The distance between the center of gravity divided by the cut piece With cut pieces The distance between the centers of the square; Indicates cut piece The area of ​​the smallest envelope rectangle of the expanded outline outside the fabric; The constant coefficient is between 0 and 2, and the adjustment coefficient is... Different nesting results can be produced. This represents the weight of the fit between the cut piece and the fabric boundary. The larger the value, the more likely it is to be selected from the edge of the fabric. S42. Calculation of the overlap value between the remaining piece and the position to be placed. The calculation of the overlap value is achieved through the following steps: traverse all the edges of the two overlapping polygons of the position to be placed and the polygon of the remaining piece, calculate the intersection points of the edges and store them in a temporary array, identify the points in the two polygons, form a new intersecting polygon, and finally use the shoelace theorem formula to calculate the area of ​​the intersecting polygon, which is the overlap value between the polygon of the position to be placed and the polygon of the remaining piece. By iterating through all the remaining polygons of the positions to be placed and repeating the above calculation, the overlap value of the current layout can be obtained. S43. Exchange and Move Operations: To reduce the overlap value to 0 and obtain the optimal layout, exchange and move operations are used for optimization. The exchange operation randomly selects two pieces of the same grade to swap positions. In the initial solution, pieces that overlap with the remaining pieces are swapped. In the current layout, other pieces of the same grade as the overlapping pieces are selected and swapped to reduce or eliminate the overlap value. If the selected remaining pieces still overlap after the exchange, the overlapping pieces are moved to a placeable area in the current layout other than their original positions, thus minimizing the overlap value. The exchange and move operations are repeated until the number of operations exceeds the limit or the overlap value of the current layout is 0. S44. If all remaining pieces have been placed, the algorithm ends and outputs the final solution; otherwise, jump to S41 and continue execution until all remaining pieces have been placed.

3. The intelligent pattern layout optimization method considering quality zoning constraints according to claim 1, characterized in that, In step S1, regarding the processing of size information, it is first necessary to accurately record the input leather size information, the size and position information of each grade area on the leather, and the size, quantity, and grade information of irregular cut pieces. This size information is represented in the form of a set of points stored in counterclockwise order; the area of ​​these polygonal areas can be calculated using the shoelace theorem formula. Based on the correspondence between the cut piece grade and the leather area grade, the placement area for each cut piece can be further calculated. The shoelace theorem formula is: ; In terms of grouping and sorting, all pieces are first sorted from highest to lowest grade; if two pieces are of the same grade, their areas are further compared, and the piece with the larger area is prioritized.

4. The intelligent pattern layout optimization method considering quality zoning constraints according to claim 3, characterized in that, In step S43, the combination of exchange and move operations constitutes the variable neighborhood search algorithm.

5. The intelligent pattern layout optimization method considering quality zoning constraints according to any one of claims 1-4, characterized in that, It also includes the step of generating the cutting path: S5: Extract the independent contour of each part, including the outer contour and internal holes, calculate the geometric features of each contour to prepare for subsequent path planning, and initialize the list of unprocessed contours. S6, take the contour closest to the origin as the starting cutting contour, take the lower left corner of the piece as the cutting point, and remove the contour from the list of unprocessed contours; S7: Select the piece of fabric closest to the cutting point from the list of unprocessed outlines as the pre-cut piece of fabric, and remove the outline from the list of unprocessed outlines. S8. If the unprocessed outline list is not empty, continue to step 3; if it is empty, the algorithm ends.

Images