Triangular layout three-head laser cutting blanking optimized layout method and system

Abstract

The invention belongs to the related technical field of layout optimization, and discloses a layout optimization method and system for triangularly-arranged three-head laser cutting blanking. According to the method, the characteristics of three-head laser cutting are combined, after a part sequence is initialized, pre-partitioning of longitudinal trisection of a plate is firstly carried out, and then a heuristic positioning algorithm which is based on the leftmost and lowermost preferentially leftmost and searches for an optimal positioning point according to the variance of the contour perimeter of the part in the region is designed for part positioning. And finally, an improved genetic-taboo mixed sequencing algorithm is used for optimizing the layout sequence of the parts, and layout optimization of three-head laser partition cutting blanking is achieved through algorithm iteration. Through the three-head laser cutting blanking line, the number of times that parts are broken can be reduced while the calculation complexity is not increased, the cutting quality of the three-head laser cutting blanking line is improved, and the laser cutting efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field related to layout optimization, and more specifically, relates to a method and system for optimizing layout of blanking with three heads of laser cutting in a herringbone shape. Background Art

[0002] At present, the three-head laser cutting blanking line with a herringbone layout has been applied in China. This technology uses three laser heads to cut at the same time, which greatly improves the efficiency and capacity of laser cutting. However, when enterprises currently use the three-head laser blanking technology, they still use the same layout optimization method as the traditional single-head laser blanking. In order to meet the conditions for the three laser heads to cut at the same time, when planning the cutting path of the three-head laser, it is necessary to divide the layout diagram into three areas for cutting path planning within the area, and to check the status of each laser head to ensure that the three laser heads will not exceed the limit when cutting. This leads to the three-head laser blanking technology facing the problem that the parts are interrupted too many times by the partition line, and the automatic partition often needs to adjust the parameters manually to pass the verification. These problems have a very adverse impact on the production of enterprises. Therefore, it is necessary to optimize the process of automatic layout of parts, combine the characteristics of three-head laser cutting, reduce the number of times parts are interrupted, improve the cutting quality of the three-head laser cutting blanking line and improve the efficiency of laser cutting, so as to improve the benefits of enterprises. Summary of the invention

[0003] In view of the above defects or improvement needs of the prior art, the present invention provides a method and system for optimizing the layout of blanking in a three-head laser cutting process with a herringbone layout, which solves the technical problem that the existing layout method only considers the use of a single laser head and ignores the characteristics of a three-head laser, resulting in parts being easily interrupted too many times by partition lines.

[0004] To achieve the above object, according to one aspect of the present invention, a method for optimizing blanking of three-head laser cutting in a herringbone layout is provided, the method comprising the following steps:

[0005] Constructing an initial population including a plurality of individuals, each individual including the arrangement order and rotation angle of all the parts to be arranged; arranging all the parts to be arranged in each individual on the plate to be processed according to the respective arrangement order and rotation angle, so as to obtain the initial arrangement result of each individual;

[0006] The initial population is optimized and iterated using a genetic algorithm. During the iteration process, the arrangement result of each individual in the initial population is updated and the fitness function is calculated using the arrangement result of each individual. The iteration is repeated multiple times until a preset maximum number of iterations is reached, so as to obtain the minimum value of the fitness function and the arrangement result of the individual corresponding thereto, which is the best arrangement result.

[0007] The initial sample arrangement result of each individual is carried out in the following manner:

[0008] Divide the plate to be processed into three equal parts using partition lines, define reference vertices of each part to be arranged, and arrange the reference vertices of the parts to be arranged with an arrangement order of 1 on the plate to be processed;

[0009] For the part pi to be arranged with an arrangement order of i, calculate the collision-free area of ​​the part pi to be arranged; determine whether the reference vertex of the part pi to be arranged can reduce the gap area between the arranged parts when it is at each location in the collision-free area; if so, the position corresponding to the smallest gap area between all the arranged parts after the part pi to be arranged is the positioning point of the reference vertex of the part pi to be arranged; if not, calculate the collision-free area of ​​the arranged parts and obtain the boundary vertices of the collision-free area of ​​the arranged parts, calculate the number of times the part pi to be arranged is interrupted by the partition line when the reference point of the part pi to be arranged is positioned at the boundary vertices of each collision-free area, and use the location with the least interruption as the positioning point of the reference vertex of the part pi to be arranged; i=i+1, until the positioning of all the reference points of the parts to be arranged in the individual is completed, that is, the arrangement is completed; i is an integer greater than 0.

[0010] Further preferably, the relationship of the fitness function is as follows:

[0011]

[0012] Z i =Max(ΣP i ) / (L×W)

[0013] Among them, F i is the fitness value of the ith individual, Z i is the evaluation value of the ith individual, Pi is the area of ​​the ith part, L is the row length, W is the plate width, and n is the total number of individuals.

[0014] Further preferably, the reference vertex of the parts to be arranged is the lower left vertex of the parts to be arranged.

[0015] Further preferably, the reference vertices of the parts to be arranged with an arrangement order of 1 are arranged on the plate to be processed using the leftmost and bottommost positioning principle.

[0016] Further preferably, when there are multiple vertices corresponding to the place with the least number of interruptions, the reference vertex of the part pi to be arranged is calculated respectively at each vertex with the least number of interruptions, and the variance of the part contour circumference in the three-divided area after all arranged parts are divided by each partition line is calculated, and the vertex corresponding to the smallest variance is used as the positioning point of the reference vertex of the part pi to be arranged.

[0017] Further preferably, when there are multiple vertices corresponding to the minimum variance, the leftmost and bottommost vertex among all vertices with the minimum variance is selected as the positioning point of the reference vertex of the part pi to be arranged according to the leftmost and bottommost positioning principle.

[0018] Further preferably, when the initial population is optimized and iterated using a genetic algorithm, when a local optimal solution is encountered during the iteration process, a taboo search algorithm is used to perform a taboo search on the current population to form a new population, so as to jump out of the local optimal solution.

[0019] Further preferably, there are 50 individuals in the initial population, of which 10 individuals are special individuals, and the special individuals are divided into two groups, wherein the arrangement order of the parts to be arranged in the 5 special individuals in one group is sorted according to the area size of the parts to be arranged, and the rotation angle is randomly generated, and the arrangement order of the parts to be arranged in the 5 special individuals in the other group is sorted according to the circumference size of the parts to be arranged, and the rotation angle is randomly generated; and the arrangement order and rotation angle of the parts to be arranged in the remaining 40 individuals are both randomly generated.

[0020] Further preferably, the gap area is equal to the area of ​​the outer envelope polygon of the arranged parts minus the sum of the areas of all the arranged parts, and the collision-free area is calculated using a critical polygon algorithm.

[0021] According to another aspect of the present invention, a blanking optimization arrangement system for three-head laser cutting in a herringbone layout is provided, the system comprising an actuator for executing the above-mentioned blanking optimization arrangement method for three-head laser cutting in a herringbone layout.

[0022] In general, the above technical solutions conceived by the present invention have the following beneficial effects compared with the prior art:

[0023] 1. The present invention takes into account the characteristics of three-head laser cutting blanking when positioning parts. The plate is first pre-divided into three equal parts longitudinally, and a positioning strategy based on the part gap area is designed for processing. The part with the least number of interruptions is selected as the individual nesting result in the nesting, which reduces the number of interruptions of the part by the partition line in the obtained nesting diagram, improves the part cutting quality, and can obtain a higher material feeding speed, thereby improving production efficiency;

[0024] 2. In the subsequent part positioning process, the algorithm of the present invention calculates the number of times the part is interrupted by the partition line, the variance of the part contour perimeter in each partition, and locates the final positioning point according to the leftmost and bottommost principle, so as to screen out the best positioning point. The positioning point obtained by this method reduces the number of times the part is interrupted by the partition line while ensuring the material utilization rate;

[0025] 3. Since there will be a problem of large blank areas in the layout diagram during the part positioning process, when it is detected that the gap area between parts decreases instead of increases after the parts are placed in a certain position, the algorithm can directly place the parts in this position by triggering a special amnesty, without having to perform the aforementioned positioning process. This can make the parts layout result more compact, thereby reducing the layout length and improving material utilization;

[0026] 4. The initial population of the present invention includes 5 special individuals sorted by part area and with random part rotation angles, 5 special individuals sorted by part circumference and with random part rotation angles, and 40 random individuals with random part order and rotation angles. By adding a certain number of special individuals to the initial population, the algorithm can obtain a convergent solution faster, while ensuring that the number of random individuals accounts for a high proportion and increasing the randomness of the algorithm, making it more likely that the algorithm will obtain the global optimal solution;

[0027] 5. A single genetic algorithm has disadvantages such as slow convergence speed and poor climbing ability. By combining it with a taboo search algorithm, the local search ability of the genetic algorithm can be enhanced. First, use the improved genetic algorithm to perform a global search on the initial population. When it is detected that the algorithm cannot obtain a better solution after several consecutive generations, use the taboo search algorithm to perform a taboo search on the population trapped in the local optimal state. After obtaining a new population through the taboo search, return to use the improved genetic algorithm to search and find the optimal individual in the new population. Such a hybrid algorithm can not only provide the taboo search algorithm with a good initial population, but also bring stronger local search capabilities to the improved genetic algorithm, thereby improving the quality of search results and facilitating faster finding of the global optimal solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a flow chart of a method for optimizing blanking arrangement of three-head laser cutting in a herringbone layout constructed according to a preferred embodiment of the present invention;

[0029] Figure 2 It is a schematic diagram of pre-dividing a plate into three equal parts longitudinally according to a preferred embodiment of the present invention;

[0030] Figure 3 is a flow chart of a BLVP-CA positioning algorithm constructed according to a preferred embodiment of the present invention;

[0031] Figure 4 The parts to be arranged and the schematic diagram thereof are constructed according to the preferred embodiment of the present invention, wherein the best positioning point is the vertex 6;

[0032] Figure 5 It is a schematic diagram of the parts to be arranged and the optimal positioning point is vertex 4 constructed according to the preferred embodiment of the present invention;

[0033] Figure 6 1 is a schematic diagram of the nesting results of parts with and without considering the amnesty situation constructed according to the preferred embodiment of the present invention, wherein (a) is the nesting result of parts without considering the amnesty situation, and (b) is the nesting result of parts with the amnesty situation considered;

[0034] Figure 7 It is a flow chart of an improved genetic-tabu hybrid algorithm constructed according to a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0036] like Figure 1 As shown, a method for optimizing blanking arrangement of three-head laser cutting in a herringbone shape is provided, the method comprising:

[0037] (1) Initialize the part sequence and angle to generate the initial population;

[0038] Initialize the part sequence and angle, generate the initial population, and the generated initial population includes two groups of 10 special individuals. One group of 5 special individuals is sorted by part area size and the part rotation angle is random. The other group of 5 special individuals is sorted by part circumference size and the part rotation angle is random. In addition to these two groups of special individuals, there are 40 individuals with randomly generated part sequence and rotation angle.

[0039] (2) Pre-partition the sheet into three equal parts in the longitudinal direction; execute the BLVP-CA positioning algorithm to complete the part positioning;

[0040] like Figure 2 As shown, after the plate is pre-divided into three equal parts longitudinally, an upper partition line L1 and a lower partition line L2 are obtained, and the three areas are named area A, area B and area C from top to bottom.

[0041] The flowchart of BLVP-CA positioning algorithm is as follows: Figure 3 As shown, the specific steps include:

[0042] Step 1: Get the parts sequence list and the parts rotation angle, and use the lower left vertex of the parts as the reference point P for part positioning. ref ;

[0043] Step 2: Use the critical polygon algorithm to calculate the collision-free region CFR of the parts to be arranged;

[0044] Step 3: Determine whether there is an amnesty situation. If so, prioritize the arrangement of the amnesty situation according to the positioning strategy based on the part gap area. The amnesty situation means that the reference vertex of the part can reduce the gap area of ​​the arranged parts when it is at various places in the non-collision zone. If there is no gap area reduction, proceed to the subsequent steps;

[0045] If it is determined that there is an amnesty, the amnesty situation will be prioritized for nesting according to the positioning strategy based on the gap area of ​​the parts. The gap area refers to the size of the gap between the parts. The smaller the gap area between the parts, the greater the fit and the better the nesting result. The method to calculate the gap area is to first calculate the outer envelope polygon area of ​​all the arranged parts, and then calculate the total area of ​​the parts within the outer envelope polygon. The difference between the two is the approximate value of the gap area.

[0046] Step 4: Calculate the sum of the number of times the parts to be arranged are interrupted by the upper partition line and the lower partition line when they are positioned at each boundary vertex of the CFR, that is, the number of intersections between the upper partition line L1 and the lower partition line L2 and the contour of the parts to be arranged, using n1, n2, n3, ..., n i They respectively represent the sum of the number of times the part is interrupted by the upper and lower partition lines when the part to be arranged is positioned at vertex i;

[0047] Step 5: Filter out the vertices that make the sum of the number of times the part to be arranged is interrupted by the upper partition line and the lower partition line the least. If there is only one vertex with the least number of interruptions, select this vertex as the best positioning point for the current part to be arranged. If there is more than one vertex with the least number of interruptions, calculate the perimeter variance values ​​of all arranged part contours in each area when the part to be arranged is positioned at these vertices. 1A It indicates the contour perimeter of all parts in area A after the parts to be arranged are positioned at vertex 1, C 1B It indicates the contour perimeter of all parts in area B after the parts to be arranged are positioned at vertex 1, C 1C It indicates the contour perimeter of all parts in area C after the part to be arranged is positioned at vertex 1. C is calculated in the same way. 2A , C 2B , C 2C ,......,C iA , C iB , C iC The formula for calculating the variance is as follows:

[0048]

[0049] like Figure 4The shown part to be nested and its schematic diagram with the best positioning point being vertex 6. A part P1 has already been placed on the sheet. The CFR of the part P2 to be nested is the red boundary in the figure and the gray area it encloses. When the part P2 to be nested is positioned at the 6 vertices of each boundary of the CFR, the number of times it is interrupted by the upper partition line and the lower partition line are respectively denoted as n1, n2, n3, n4, n5, n6. It can be obtained that n1 = n3 = n6 = 0, n2 = n4 = n5 = 2. First, it is determined that vertices 1, 3, and 6 can enable the part to be nested without crossing regions, that is, the number of times the part is interrupted by the partition line is the minimum value of 0. Then, calculate the perimeter and its variance C of the part contour in the three regions after the part is positioned at vertices 1, 3, and 6 1A , C 1B , C 1C , C 3A , C 3B , C 3C , and C 6A , C 6B , C 6C , It can be obtained So the best positioning point of the part P2 to be nested is vertex 6.

[0050] Step6: Screen out the vertex with the smallest perimeter variance value of all the arranged part contours in each region after the part to be nested is arranged. If there is only one vertex with the smallest variance value, select this vertex as the best positioning point of the current part to be nested. If there is still more than one vertex with the smallest variance value, select the leftmost and lowest (prioritizing the leftmost) vertex as the best positioning point.

[0051] As Figure 5 shown, the part to be nested and its schematic diagram with the best positioning point being vertex 4. Part P2 is the part to be nested. Calculate its collision-free region CFR as the red solid line and the gray area it encloses. There are four vertices 1, 2, 3, and 4. The sheet partition lines are the red and green dotted lines in the figure, dividing the sheet into three regions A, B, and C from top to bottom. According to the above algorithm steps: First, calculate the number of times n1, n2, n3, n4 that the part P2 is interrupted by the partition line when positioned at vertices 1, 2, 3, and 4. It is obtained that n1 = n3 = n4 = 2 < n2 = 4. So vertices 1, 3, and 4 are screened out for the subsequent steps. Then, calculate the perimeter and its variance C of the part contour in the three regions after the part is positioned at vertices 1, 3, and 4 1A , C 1B , C 1C , C 3A , C 3B , C 3C , and C 4A , C4B , C 4C , obtain Finally, according to the leftmost and bottommost priority and leftmost strategy, vertex 4 is selected as the best positioning point of part P2 among vertex 3 and vertex 4.

[0052] If it is judged that there is an amnesty situation in the above Step3, then the nesting of the amnesty situation is preferentially carried out according to the positioning strategy based on the part clearance area. For example, Figure 6 As shown, the parts to be nested are nested in the order of P5 and P6. When not considering the amnesty situation, part P5 first fills the blank area in the upper left corner, which causes the placement position of part P6 to be only more to the right, greatly increasing the nesting length. When considering the amnesty situation, since part P5 can be placed in the part clearance below, this will make the clearance area between parts smaller. Therefore, the amnesty rule is triggered to place part P5 at the leftmost and bottommost position in the cavity. At this time, part P6 can be nested into the blank area in the upper left corner. After parts P5 and P6 are nested, the nesting length does not increase, and the obtained nesting result is significantly better.

[0053] (4) Optimize and iterate the initial population using the genetic algorithm. During the iteration process, calculate the fitness function using the nesting results of each individual in the population, record the current best nesting result, and judge whether the maximum number of iterations is reached. If the maximum number of iterations is not reached, then after performing the improved genetic-tabu hybrid algorithm to optimize the nesting order, return to step (3); if the maximum number of iterations is reached, then output the best nesting drawing.

[0054] For example, Figure 7 As shown, an improved genetic-tabu hybrid algorithm is used to optimize the nesting order and rotation angle of parts, and the nesting optimization of three-head laser zoning cutting and blanking is realized, including the following steps:

[0055] Step1: Initialize the population X and set the maximum number of iterations M of the improved genetic algorithm;

[0056] Step2: Use the improved genetic algorithm to iteratively optimize the initial population to obtain a new population X* after selection, crossover, and mutation;

[0057] Step3: Let the current iteration number of the improved genetic algorithm be n (n < M). If a better result has not been obtained for several consecutive generations, then call the tabu search algorithm to perform tabu search on the population X* to obtain a new population XL; if the improved genetic algorithm has not been unable to find a better solution for several consecutive generations, then jump out of the program after reaching the maximum number of iterations and output the current optimal result;

[0058] Step 4: If the taboo search algorithm obtains a better solution, the algorithm will go to Step 2 with the new population XL to continue genetic iteration and repeat the above steps. Otherwise, the program will be exited and the current optimal result will be output.

[0059] Preferably, the improved genetic algorithm has made the following improvements to each step of the genetic algorithm:

[0060] (1) Coding: A double-layer decimal coding scheme is used, and a double-layer coding structure is used to express the arrangement order and rotation angle of the parts;

[0061] (2) Initial population: The generated initial population includes two groups of 10 special individuals. One group of 5 special individuals is sorted by part area and the part rotation angle is random. The other group of 5 special individuals is sorted by part circumference and the part rotation angle is random. In addition to these two groups of special individuals, there are 40 individuals whose part order and rotation angle are randomly generated.

[0062] (3) Fitness function: After all parts are arranged, the set of codes for the arrangement order of parts and their rotation angles constitutes an individual in the population. The following objective function formula is used to evaluate the individuals in the population:

[0063] Z=Max(ΣP i ) / (L×W)

[0064] And use the following formula to convert the evaluation value into fitness value:

[0065]

[0066] In the formula, F i represents the fitness value of the i-th individual, Z i Represents the evaluation value of the ith individual. The percentage of the difference between each individual's evaluation value and the lowest evaluation value in the sum of all individual differences is the individual's fitness value;

[0067] (4) Selection operator: The selection operation is performed by combining roulette selection with elite selection. Its proportional selection operator can be expressed by the following formula:

[0068]

[0069] Unlike roulette wheel selection based purely on probability, the method of retaining some elites directly based on their fitness values ​​is called elite selection. For example, the population size of the genetic algorithm is N, and the elite retention rate is P e , then N*P will be retained in one round of selection e An elite individual.

[0070] (5) Crossover operator: Three crossover operators are used. In the population, one-third uses single-point crossover, one-third uses double-point crossover, and the remaining one-third uses subpath exchange crossover. The three crossover operators are as follows:

[0071] Single-point crossover: First, two individuals are selected as parent 1 and parent 2. The single-point crossover is the k-th coding position. Then the offspring individual has the same first k codes as the parent individual, and the subsequent data is arranged according to the coding order of the other parent individual.

[0072] Double-point crossover: First, two individuals are selected as parent 1 and parent 2. The double-point crossover is the k1th and k2th coding positions. Then, the coding of one parent between k1 and k2 is arranged in the order of the corresponding coding of the other parent. The coding of the parent individual that is not between k1 and k2 remains unchanged, and a new offspring individual is obtained.

[0073] Subpath exchange crossover: select two individuals as parent 1 and parent 2, randomly select a code in one parent individual, rearrange the selected code in the order corresponding to the same code in the other parent individual, and keep the other unselected codes unchanged to obtain a new offspring individual.

[0074] (6) Mutation operator: Four different mutation operators are used: gene preposition, gene postposition, gene flipping and gene swapping.

[0075] The present invention designs a double-layer decimal coding scheme, the upper layer represents the order of parts being arranged, and the lower layer represents the rotation angle of the parts. The initial population consisting of special individuals + random individuals is designed for the genetic algorithm. The selection strategy of elite selection + roulette is adopted in the selection operation, a part of elite individuals are directly retained, and crossover and mutation individuals are generated in proportion; then, for the crossover and mutation operations of the genetic algorithm, 4 mutation operators and 3 crossover operators are designed, the 4 mutation operators are gene preposition, gene postposition, gene flipping and gene exchange, and the 3 crossover operators are single-point crossover, double-point crossover and sub-path exchange crossover; the improved genetic algorithm has stronger global search ability and can retain excellent genes through the elite selection strategy.

[0076] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for optimizing blanking arrangement of three-head laser cutting in a herringbone layout, characterized in that: The method comprises the following steps: Constructing an initial population including a plurality of individuals, each individual including the arrangement order and rotation angle of all the parts to be arranged; arranging all the parts to be arranged in each individual on the plate to be processed according to the respective arrangement order and rotation angle, so as to obtain the initial arrangement result of each individual; The initial population is optimized and iterated using a genetic algorithm. During the iteration process, the arrangement result of each individual in the initial population is updated and the fitness function is calculated using the arrangement result of each individual. The iteration is repeated multiple times until a preset maximum number of iterations is reached, so as to obtain the minimum value of the fitness function and the arrangement result of the individual corresponding thereto, which is the best arrangement result. The initial sample arrangement result of each individual is carried out in the following manner: Divide the plate to be processed into three equal parts using partition lines, define reference vertices of each part to be arranged, and arrange the reference vertices of the parts to be arranged with an arrangement order of 1 on the plate to be processed; For the part pi to be arranged with arrangement order i, calculate the collision-free area of ​​the part pi to be arranged; Determine whether the reference vertex pi of the part to be arranged can reduce the gap area between the arranged parts when it is located in each of the collision-free areas; If so, the position corresponding to the smallest gap area between all the arranged parts after the arrangement of the part pi to be arranged is the positioning point of the reference vertex of the part pi to be arranged; if not, calculate the collision-free area of ​​the arranged parts and obtain the boundary vertices of the collision-free area of ​​the arranged parts, and calculate the number of times the part pi to be arranged is interrupted by the partition line when the reference point of the part pi to be arranged is positioned at the boundary vertices of each collision-free area, and the position with the least interruption is used as the positioning point of the reference vertex of the part pi to be arranged; i=i+1, until the positioning of all the reference points of the parts to be arranged in the individual is completed, that is, the arrangement is completed; i is an integer greater than 0.

2. The method for optimizing blanking arrangement of three-head laser cutting in a herringbone layout as claimed in claim 1, characterized in that: The relationship of the fitness function is as follows: Among them, F i is the fitness value of the ith individual, Z i is the evaluation value of the ith individual, P i is the area of ​​the ith part, L is the row length, W is the plate width, and n is the total number of individuals.

3. The method for optimizing blanking arrangement of three-head laser cutting in a herringbone layout as claimed in claim 1, characterized in that: The reference vertex of the part to be arranged is the lower left vertex of the part to be arranged.

4. The method for optimizing blanking arrangement of three-head laser cutting in a herringbone layout as claimed in claim 3, characterized in that: The reference vertices of the parts to be arranged with an arrangement order of 1 are arranged on the plate to be processed using the leftmost and bottommost positioning principle.

5. The method for optimizing blanking arrangement of three-head laser cutting in a herringbone layout as claimed in claim 1, characterized in that: When there are multiple vertices corresponding to the place with the least number of interruptions, the reference vertex of the part pi to be arranged is calculated respectively at each vertex with the least number of interruptions. The variance of the part contour circumference in the three-divided area after all arranged parts are divided by each partition line is calculated, and the vertex corresponding to the smallest variance is used as the positioning point of the reference vertex of the part pi to be arranged.

6. The method for optimizing blanking arrangement of three-head laser cutting in a herringbone layout as claimed in claim 5, characterized in that: When there are multiple vertices corresponding to the minimum variance, the leftmost and bottommost vertex among all vertices with the minimum variance is selected as the positioning point of the reference vertex pi of the part to be arranged according to the leftmost and bottommost positioning principle.

7. The method for optimizing blanking arrangement of three-head laser cutting in a herringbone layout as claimed in claim 1, characterized in that: When the initial population is optimized and iterated using the genetic algorithm, when a local optimal solution is encountered during the iteration process, a taboo search algorithm is used to perform a taboo search on the current population to form a new population, thereby jumping out of the local optimal solution.

8. The method for optimizing blanking arrangement of three-head laser cutting in a herringbone layout as claimed in claim 1, characterized in that: There are 50 individuals in the initial population, of which 10 individuals are special individuals. The special individuals are divided into two groups. The arrangement order of the parts to be arranged in the 5 special individuals in one group is sorted according to the area size of the parts to be arranged, and the rotation angle is randomly generated. The arrangement order of the parts to be arranged in the 5 special individuals in the other group is sorted according to the circumference size of the parts to be arranged, and the rotation angle is randomly generated; in the remaining 40 individuals, the arrangement order and rotation angle of the parts to be arranged are both randomly generated.

9. The method for optimizing blanking arrangement of three-head laser cutting in a herringbone layout as claimed in claim 1 or 2, characterized in that: The gap area is equal to the area of ​​the outer envelope polygon of the arranged parts minus the sum of the areas of all the arranged parts, and the collision-free area is calculated using a critical polygon algorithm.

10. A three-head laser cutting blanking optimization layout system with a herringbone layout, characterized in that: The system includes an actuator, which is used to execute the optimized layout method for blanking with three heads of laser cutting in a herringbone layout as described in any one of claims 1 to 9.

Images