CN113592174B - Knowledge-driven two-dimensional polygonal cloth piece automatic discharging method - Google Patents
Knowledge-driven two-dimensional polygonal cloth piece automatic discharging method Download PDFInfo
- Publication number
- CN113592174B CN113592174B CN202110855739.0A CN202110855739A CN113592174B CN 113592174 B CN113592174 B CN 113592174B CN 202110855739 A CN202110855739 A CN 202110855739A CN 113592174 B CN113592174 B CN 113592174B
- Authority
- CN
- China
- Prior art keywords
- polygon
- dimensional
- dimensional polygonal
- polygons
- sample
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000004744 fabric Substances 0.000 title claims abstract description 184
- 238000007599 discharging Methods 0.000 title claims abstract description 59
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000005457 optimization Methods 0.000 claims abstract description 30
- 208000028659 discharge Diseases 0.000 claims description 32
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000004422 calculation algorithm Methods 0.000 abstract description 34
- 238000004519 manufacturing process Methods 0.000 abstract description 11
- 239000000463 material Substances 0.000 abstract description 3
- 230000006872 improvement Effects 0.000 abstract description 2
- 230000008569 process Effects 0.000 description 11
- 230000001788 irregular Effects 0.000 description 7
- 239000004753 textile Substances 0.000 description 5
- 230000006835 compression Effects 0.000 description 4
- 238000007906 compression Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 238000000926 separation method Methods 0.000 description 3
- 238000005520 cutting process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000002922 simulated annealing Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000009194 climbing Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000012821 model calculation Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
- G06Q10/043—Optimisation of two dimensional placement, e.g. cutting of clothes or wood
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Manufacturing & Machinery (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Treatment Of Fiber Materials (AREA)
Abstract
The invention discloses a two-dimensional polygonal cloth piece automatic discharging method based on knowledge driving. The invention comprises the following steps: calculating to obtain external critical polygons between every two polygons; the method comprises the steps of arranging a plurality of polygons in a descending order, calculating an inscribed critical polygon of the polygon to be arranged for the fabric and an circumscribed critical polygon of the polygon to be arranged, and calculating an optimal placement position by using a local fitness model to obtain a primary material arrangement result of the polygons; and optimizing by utilizing the local fitness model and the knowledge driving model to obtain optimized discharging results of a plurality of polygons, continuously optimizing, and obtaining the optimal discharging results in preset time. The method is suitable for adopting a more concise and efficient method on the optimization model, has high algorithm iteration efficiency, shorter consumed time and remarkable improvement of the utilization rate, is more efficient on the aspect of large sample data sets, can obtain a more satisfactory result in a short time, and is more suitable for the actual production environment.
Description
Technical Field
The invention belongs to a two-dimensional polygonal discharging method in the field of computer graphics, and particularly relates to a knowledge-driven two-dimensional polygonal cloth piece automatic discharging method.
Background
The problem of emission of two-dimensional irregular polygons is a combinatorial optimization problem, requiring a given set of two-dimensional polygons to be placed in a rectangular container, where each polygon is not necessarily a convex polygon, requiring that all polygons do not overlap or exceed the rectangular container with other polygons, such a layout being feasible. Rectangular containers have a fixed width, while their length can be modified to place all polygons therein. The aim is to find a viable layout that minimizes the container length. The emission problem of two-dimensional irregular polygons varies somewhat with the rotation of the polygon: (1) allow any angle of rotation, (2) allow a limited number of angles, (3) not allow rotation. Wherein case (2) is processed. Note that the case (3) is a special case of (2) in which the given rotation angle of each polygon is 90 ° or 180 °. The problem of two-dimensional irregular polygonal emission has many applications in the materials industry, such as the paper and textile industry, where raw materials are typically provided in rolls. In the textile industry, rotation is typically limited to 180 ° because the textile has a texture and possibly a stretch pattern. It is well known that even without rotation, the emission problem of two-dimensional irregular polygons is also an NP (nonlinear polynomial non-deterministic polynomial) problem.
At present, many scholars have studied the emission problem of two-dimensional irregular polygons, and the research direction is mainly divided into three categories: the problem is solved by the critical polygon, the problem of the polygon emission mode, and the problem of polygon emission optimization (comprising emission positions and emission sequences).
The solving method of the critical polygon and the research of the polygon emission mode problem are relatively mature, and more research focuses are mainly focused on the polygon emission optimization problem at present.
Adamicz and Albano propose an algorithm that divides a given set of polygons into a plurality of polygon subsets, then generates a rectangular shell for each subset, wherein the polygons in the subset are compactly placed therein (i.e., wasting a bit of space), and finally finds the layout of these matrices. Albano and Sapuppo present an algorithm that places given polygons one by one in the lower left corner according to the order in which they are input, and then they use tree search to obtain a good sequence. Some methods of finding good sequences are based on local searches.
Mathematical programming is also used for two-dimensional irregular polygonal emission problems. For example, milenkovic et al propose a linear programming based compression and separation algorithm. Given the feasible layout of a given polygon in a container, the compression algorithm will continuously transform the polygons in the current layout to minimize the length of the container and output a locally optimal solution. If the layout of a given polygon is not feasible, then the separation algorithm will continuously transform the polygons in the current layout to eliminate overlap of the polygons.
Bennell and dowslot combine the lower left approach with a linear programming based compression algorithm to obtain a better performing algorithm. Gomes and Oliveira combine a heuristic algorithm in the lower left corner with a linear programming based compression and separation algorithm. They further incorporated the method into a simulated annealing algorithm. Burke et al developed a left-hand underfill algorithm and used it in conjunction with hill climbing or tabu search to quickly obtain a high quality solution. Egeblad et al developed an efficient method of finding the best position of a given polygon by translating the polygon in the given direction, minimizing its overlap area with the current layout, and using it to guide local searches.
The scholars mostly apply intelligent optimization algorithms, such as genetic algorithms, simulated annealing algorithms, particle swarm algorithms, tabu search algorithms and the like, on the polygon emission optimization problem. The intelligent optimization algorithm has good effect on the problem of polygon emission optimization, but the intelligent optimization algorithm is characterized by long time consumption and high randomness of the early optimization result, so that the use samples in the research of most students are based on small sample models, but are not suitable for the actual production environment. For example, in textile fields such as sofas and clothes, there are often hundreds of fabrics to be discharged, and at this time, the required discharge time is difficult to estimate by using an intelligent optimization algorithm, which may lead to stagnation of actual production. Therefore, a two-dimensional irregular polygon automatic discharging method with higher efficiency, shorter time consumption and more definite rule destination is needed at present.
Disclosure of Invention
In view of the above-mentioned drawbacks of the prior art, an object of the present invention is to provide an automatic two-dimensional polygonal sheet discharging method based on knowledge driving. The invention can reduce the time required by polygon emission optimization, does not sacrifice the utilization rate of the sample, can be suitable for large-scale sample discharging, and has higher application prospect in actual production.
The invention adopts the following technical scheme:
the invention comprises the following steps:
s1: according to the two-dimensional polygon cloth sample, calculating to obtain the external critical polygons between every two-dimensional polygon cloth samples;
s2: the two-dimensional polygonal cloth sample is arranged in a descending order according to the areas of the two-dimensional polygonal cloth sample, and the positions in the fabric are placed one by one according to the arrangement order; calculating an inscribed critical polygon of the two-dimensional polygonal cloth sample to be arranged relative to the fabric and an circumscribed critical polygon of the two-dimensional polygonal cloth sample to be arranged relative to the arranged two-dimensional polygonal cloth sample according to the two-dimensional polygonal cloth sample to be arranged, and calculating and judging the placement position by using a local fitness model to obtain the optimal placement position of the two-dimensional polygonal cloth sample to be arranged and obtain the primary discharge results of a plurality of two-dimensional polygonal cloth samples;
s3: according to the primary discharging results of the two-dimensional polygonal sheet samples, optimizing the arrangement sequence and the optimal placement position of the two-dimensional polygonal sheet samples by utilizing a local fitness model and a knowledge driving model to obtain optimized discharging results of the two-dimensional polygonal sheet samples, and continuously optimizing the optimized discharging results of the two-dimensional polygonal sheet samples;
s4: and taking the optimal discharging result obtained in the preset time as the final discharging result of the plurality of two-dimensional polygonal cloth sample.
The S1 specifically comprises the following steps:
s11: performing de-duplication operation on the plurality of two-dimensional polygonal cloth samples to obtain a repeated polygonal cloth set S1 and a de-duplication polygonal cloth set S2, wherein the repeated polygonal cloth set S1 is a set formed by repeated two-dimensional polygonal cloth samples in the plurality of two-dimensional polygonal cloth samples, and the de-duplication polygonal cloth set S2 is a set formed by removing redundant repeated two-dimensional polygonal cloth samples in the plurality of two-dimensional polygonal cloth samples;
s12: the two-dimensional polygonal cloth sample in the repeated polygonal cloth set S1 or the de-repeated polygonal cloth set S2 is marked as (P) i ) I=1, 2, …, L, i is the sequence number of the two-dimensional polygonal patch sample in the repeating polygonal patch set S1 or the deduplication polygonal patch set S2, L is the total number of the two-dimensional polygonal patch samples in the repeating polygonal patch set S1 or the deduplication polygonal patch set S2; calculating to obtain external critical polygons of the repeated polygon piece set S1 and the de-duplicated polygon piece set S2, and obtaining external critical polygons between every two-dimensional polygon piece samples;
for repeated polygonal cloth piecesSet S1, calculate two-dimensional polygon patch sample (P i ) External critical polygons to itself; for the deduplication polygon patch set S2, a two-dimensional polygon patch sample (P i ) With the rest two-dimensional polygonal cloth sample P 1 ，P 2 …P i-1 ，P i+1 …P L And the critical polygons are externally connected between every two of the two.
The step S2 is specifically as follows:
s21: the two-dimensional polygonal cloth sample is arranged in a descending order according to the areas of the two-dimensional polygonal cloth sample, and the positions in the fabric are placed one by one according to the arrangement order; firstly, placing a first two-dimensional polygonal cloth sample into a fabric and taking the first two-dimensional polygonal cloth sample as a arranged polygon, and taking the next two-dimensional polygonal cloth sample to be arranged as a polygon to be arranged;
s22: calculating an inscribed critical polygon of the polygon to be arranged for the fabric, marking the inscribed critical polygon as an inscribed critical polygon track IFP, and marking an circumscribed critical polygon of the polygon to be arranged for all the arranged polygons as an circumscribed critical polygon track NFP;
s23: performing difference calculation on the external critical polygon track NFP through the internal critical polygon track IFP to obtain a track polygon of the current polygon to be arranged;
s24: calculating the fitting degree of each vertex of the track polygon by using a local fitting degree function, taking the vertex with the largest fitting degree as an optimal placement position, placing the current polygon to be arranged to the optimal placement position, and updating the arranged polygon and the polygon to be arranged;
s25: and repeating S22-S24 to discharge the residual two-dimensional polygonal cloth sample, so as to obtain primary discharge results of the two-dimensional polygonal cloth samples, wherein the primary discharge results comprise the primary length of the fabric.
The step S3 is specifically as follows:
s31: determining the primary length of the fabric according to the primary discharging result of the two-dimensional polygonal cloth sample, wherein the length to be discharged of the fabric is smaller than the primary length, and arranging the two-dimensional polygonal cloth sample in a descending order according to the areas of the two-dimensional polygonal cloth sample;
s32: placing a first two-dimensional polygonal cloth sample into the fabric and taking the first two-dimensional polygonal cloth sample as a arranged polygon, and taking a plurality of two-dimensional polygonal cloth samples behind the arranged polygon as polygons to be arranged;
s33: calculating the optimal placement positions of all the polygons to be arranged by using a local fitness model, calculating the credibility of the optimal placement positions of all the polygons to be arranged by using a knowledge driving model, arranging the polygons to be arranged with the largest credibility into the sequence, returning the rest polygons to be arranged into the sequence, and updating the arranged polygons and the polygons to be arranged;
s34: repeating S33 to discharge the remaining two-dimensional polygonal cloth sample, so as to obtain optimized discharge results of the two-dimensional polygonal cloth samples, wherein the optimized primary discharge results comprise optimized lengths of the fabrics;
s35: judging the optimized discharging result, if the optimized length of the fabric is greater than the length to be discharged, failing the optimization, and advancing the sequence of two-dimensional polygonal cloth sample exceeding the length to be discharged to obtain a new arrangement sequence, and repeating S32-S34; if the optimized length of the fabric is equal to or smaller than the length to be arranged, the optimization is successful, the arrangement sequence is recorded, the length to be arranged is reduced, and the optimization is continued by repeating the steps S32-S34.
And in the step S4, the credibility of the optimal placement position of each polygon to be arranged is calculated by using a knowledge driving model, and the calculation is specifically performed by the following formula:
s＝c 1 *f+c 2 *w+c 3 *p
wherein s represents the reliability of the optimal placement position of the current polygon to be ranked, w represents the waiting times of the current polygon to be ranked, f represents the fitness of the optimal placement position of the current polygon to be ranked, p represents the lifting parameter of the polygon to be ranked, c 1 ，c 2 ，c 3 First, second and third constant coefficients, respectively.
The beneficial effects of the invention are as follows:
1. the invention obtains the feasible solution of the polygon to be arranged through the critical polygon, judges the arrangement position quality of the polygon to be arranged by means of the local fitness function fitness, and judges the quality of the polygon in the feasible solution candidate set by means of the knowledge driving model. The realization process is simple and efficient, the codes are easy to realize, and the expandability is high. The knowledge driven model concept target is more definite, and the iterative optimization target is more accurate.
2. The invention is more suitable for practical production environment, especially for textile products, has excellent expression effect on large sample data (100 or more), has controllable overall discharging time, and optimizes iteration parameters more accurately and compactly compared with the traditional algorithm for carrying out polygon discharging optimization processing by using an intelligent optimization algorithm in the optimal solution within preset time, and is lighter in code realization and easier to carry into common numerical control cutting machines. The method is more suitable for the actual production environment, and requires the discharging time to be as short as possible, and meanwhile, the obtained result is as good as possible without delaying the subsequent production line.
Drawings
For further explanation of the description of the present invention, the following describes the embodiments of the present invention in further detail with reference to the accompanying drawings. It is to be understood that these drawings are solely for purposes of illustration and are not intended as a definition of the limits of the invention.
Fig. 1 is a block diagram of an algorithm of the present invention.
FIG. 2 is a schematic diagram of an algorithm for solving critical polygons by the Minkowski method.
FIG. 3 is a flow chart of a critical polygon initialization algorithm between all polygons.
Fig. 4 is a flowchart of a polygon deduplication algorithm.
Fig. 5 is a flowchart of a polygon-critical polygon solution algorithm.
Fig. 6 is a flowchart for obtaining a polygon trace.
Fig. 7 is a schematic diagram of a fit acquisition mode.
Fig. 8 is a sample wafer initial discharge flow chart.
Fig. 9 is a sample optimized discharge flow chart.
Fig. 10 shows the discharge result of sample 1.
Fig. 11 shows the discharge result of sample 2.
Fig. 12 shows the discharge result of sample 3.
Detailed Description
Other advantages and effects of the present invention will become apparent to those skilled in the art from the following disclosure of the present invention, which is to be read in light of the specific examples. The invention may be practiced or carried out in other embodiments and details within the scope and range of equivalents of the specific embodiments and ranges of equivalents, without departing from the spirit of the invention.
According to the embodiment of the invention, a sofa sample data set is obtained from an actual production environment, the materials are discharged by an applied algorithm, the sample set comprises 142 two-dimensional polygonal cloth sample, the average vertex number of the two-dimensional polygonal cloth sample is 36, and the specific discharging process is as follows:
and acquiring dxf files of the sofa sample data set, acquiring txt files representing coordinate values of each vertex of the polygon through data conversion, and starting automatic discharging. The method comprises the steps of providing a discharging time of 3 minutes, analyzing characteristics of sample data, selecting proper parameters from an established parameter library for subsequent model calculation, and reducing a point set of a sample polygon with excessive number of vertexes (one point is omitted if the distance between two points is too short so as not to influence the whole), so that the calculation complexity of the subsequent critical polygon is reduced, and the algorithm efficiency is improved.
As shown in fig. 1, the method comprises the following steps:
according to the method, the positions to be arranged of the polygons are obtained through the critical polygons, the positions to be arranged are judged to be good or bad by means of a knowledge driving model, the optimal arrangement positions are obtained through calculation, all the polygons are automatically calculated in sequence, and automatic arrangement of the two-dimensional polygons is completed.
S1: and calculating and obtaining circumscribed critical polygons between every two-dimensional polygonal cloth samples by using a Minkowski difference method according to a sofa sample data set, as shown in figures 2, 3 and 5. Minkowski methodno-fit polygon, noted NFP), characterized by a simple algorithm implementation, the algorithm principle referring to fig. 2, fig. 2 a) representing two polygonal patch samples by vector line segments, fig. 2 b) representing all vector line segments of the two polygonal patch samples, fig. 2 c) sequentially connecting all vector line segments of the two polygonal patch samples end to obtain a critical polygon, fig. 2 c) NFP AB I.e. the critical polygon that is sought. Initializing critical polygons among all polygons, finding, storing, counting and eliminating repeated polygons in a polygon array from an original array, then firstly solving NFP of the repeated polygons for each polygon under each angle combination, and finally solving NFP among different polygons by utilizing multiple threads. Specific flow refers to the critical polygon initialization algorithm flow diagram among all polygons of fig. 3.
S1 specifically comprises the following steps:
s11: performing a de-duplication operation on the plurality of two-dimensional polygonal cloth samples, as shown in fig. 4, to obtain a repeated polygonal cloth set S1 and a de-duplication polygonal cloth set S2, where the repeated polygonal cloth set S1 is a set formed by repeated two-dimensional polygonal cloth samples in the plurality of two-dimensional polygonal cloth samples, and the de-duplication polygonal cloth set S2 is a set formed by removing redundant repeated two-dimensional polygonal cloth samples in the plurality of two-dimensional polygonal cloth samples; for example, the plurality of two-dimensional polygonal tile samples are denoted { A1, A1, B1, B2, C1, C1}, then S1= { A1, C1}, S2= { A1, B1, B2, C1}.
S12: the two-dimensional polygonal cloth sample in the repeated polygonal cloth set S1 or the de-repeated polygonal cloth set S2 is marked as (P) i ) I=1, 2, …, L, i is the sequence number of the two-dimensional polygonal patch sample in the repeating polygonal patch set S1 or the deduplication polygonal patch set S2, L is the total number of the two-dimensional polygonal patch samples in the repeating polygonal patch set S1 or the deduplication polygonal patch set S2; calculating to obtain external critical polygons of the repeated polygon cloth piece set S1 and the de-duplication polygon cloth piece set S2 by using a Minkowski difference method, and obtaining external critical polygons between every two-dimensional polygon cloth piece samples;
for repetition ofThe polygon patch set S1 calculates a two-dimensional polygon patch sample (P i ) External critical polygons to itself; for the deduplication polygon patch set S2, a two-dimensional polygon patch sample (P i ) With the rest two-dimensional polygonal cloth sample P 1 ，P 2 …P i-1 ，P i+1 …P L And the critical polygons are externally connected between every two of the two.
In addition, when the cloth sample is allowed to have a preset rotation angle, the two-dimensional polygonal cloth sample P is obtained i The rotation is performed and then the calculation is performed. In particular, the preset rotation angle is set to 180 DEG
S2: the two-dimensional polygonal cloth sample is arranged in a descending order according to the areas of the two-dimensional polygonal cloth sample, and the positions in the fabric are placed one by one according to the arrangement order; calculating an inscribed critical polygon of the two-dimensional polygonal cloth sample to be arranged relative to the fabric and an circumscribed critical polygon of the two-dimensional polygonal cloth sample to be arranged relative to the arranged two-dimensional polygonal cloth sample according to the two-dimensional polygonal cloth sample to be arranged, and calculating and judging the placement position by using a local fitness model to obtain the optimal placement position of the two-dimensional polygonal cloth sample to be arranged and obtain the primary discharge results of a plurality of two-dimensional polygonal cloth samples;
s2 is specifically as shown in fig. 8:
s21: the two-dimensional polygonal cloth sample is arranged in a descending order according to the areas of the two-dimensional polygonal cloth sample, and the positions in the fabric are placed one by one according to the arrangement order; firstly, placing a first two-dimensional polygonal cloth sample into a fabric and taking the first two-dimensional polygonal cloth sample as a arranged polygon, and taking the next two-dimensional polygonal cloth sample to be arranged as a polygon to be arranged;
s22: calculating an inscribed critical polygon of the polygon to be arranged for the fabric, marking the inscribed critical polygon as an inscribed critical polygon track IFP, and marking an circumscribed critical polygon of the polygon to be arranged for all the arranged polygons as an circumscribed critical polygon track NFP;
s23: obtaining a track polygon of a current polygon to be arranged by differentiating an external critical polygon track NFP through an internal critical polygon track IFP, as shown in FIG. 6, wherein a of FIG. 6 is a critical polygon of the polygon to be arranged for the polygon to be arranged, b of FIG. 6 is all critical polygons which are taken out in a merging way, c of FIG. 6 is an internal critical polygon of the polygon to be arranged relative to a fabric, d of FIG. 6 is an external critical polygon track NFP through an internal critical polygon track IFP, and a track polygon of the current polygon to be arranged is obtained by differentiating the external critical polygon track NFP, and e of FIG. 6 is a polygon which is placed at a proper position;
s24: calculating the fitness of each vertex of the track polygon by using a local fitness function fitness, wherein as shown in fig. 7 (a), the vertex with the largest fitness is taken as an optimal placement position, the current polygon to be arranged is placed at the optimal placement position, and the arranged polygon and the polygon to be arranged are updated;
the local fitness function fitness is obtained through weighting fitness and position parameters, wherein constant coefficients are obtained through polygonal preprocessing, and sample data characteristics are analyzed. Degree of fit: s is(s) 1 ＝c 1 (the intersection area of the flaring profile between the parts + the intersection area of the part and the boundary flaring profile)/(the circumference of the part x the used length of the fabric); position parameters: s is(s) 2 ＝c 2 X L/used length of fabric. Note that: c 1 ，c 2 ，c 3 The constant coefficient is given, and L is the length of the fabric when the utilization rate of the fabric is 100%. When the polygons have been placed in a certain number, the weight of the position parameter should be increased, because the position parameter is an index that actually affects the utilization of the final result, and the fitness is to make the polygons fit more. The discharge process is thus divided into two parts, the former part taking into account more of the fit of the polygon and the latter part taking into account more of the position parameters of the polygon. Evaluation index formula: fitness= (1) s 1 +0.01s 2 When the used length of the fabric is less than L/c 3 ；(2)s 1 +s 2 When the used length of the fabric is more than or equal to L/c 3 . The process of obtaining the intersection area of the flaring profile between the parts and the intersection area of the part and the boundary flaring profile in the fit is shown in fig. 7 (b).
S25: and repeating S22-S24 to discharge the residual two-dimensional polygonal cloth sample, so as to obtain primary discharge results of the two-dimensional polygonal cloth samples, wherein the primary discharge results comprise the primary utilization rate of the fabric and the primary length of the fabric.
S3: according to the primary discharging results of the two-dimensional polygonal sheet samples, optimizing the arrangement sequence and the optimal placement position of the two-dimensional polygonal sheet samples by utilizing a local fitness model and a knowledge driving model to obtain optimized discharging results of the two-dimensional polygonal sheet samples, and continuously optimizing the optimized discharging results of the two-dimensional polygonal sheet samples;
s3 is specifically as shown in fig. 9:
s31: determining the primary length of the fabric according to the primary discharging result of the two-dimensional polygonal cloth sample, wherein the length to be discharged of the fabric is smaller than the primary length, and arranging the two-dimensional polygonal cloth sample in a descending order according to the areas of the two-dimensional polygonal cloth sample;
s32: placing a first two-dimensional polygonal cloth sample into the fabric and taking the first two-dimensional polygonal cloth sample as a arranged polygon, and taking a plurality of two-dimensional polygonal cloth samples behind the arranged polygon as polygons to be arranged; in the embodiment, four two-dimensional polygonal cloth samples are used as polygons to be arranged.
S33: calculating the optimal placement positions of all the polygons to be arranged by using a local fitness model, calculating the credibility of the optimal placement positions of all the polygons to be arranged by using a knowledge driving model, arranging the polygons to be arranged with the largest credibility into the sequence, returning the rest polygons to be arranged into the sequence, and updating the arranged polygons and the polygons to be arranged;
wherein, the credibility function according to the knowledge driving model is established, and the knowledge 1 is according to the following three important knowledge: the polygons which participate in the discharging are easier to be discharged to the proper position; knowledge 2: firstly, arranging polygons with large areas firstly, and arranging the polygons with small areas in a blank way; knowledge 3: if the length of the discharge web is reduced, the tail-end misalignment is overcome.
According to knowledge 2, polygons with large areas are usually arranged first in the process of arranging, then polygons with small areas are arranged in an empty way, but some possible better solutions, such as better solutions in the case of arranging the small polygons first, are missed, so that as a compromise scheme, when the ith polygon is arranged, not only one polygon is considered, but a plurality of subsequent polygons are considered, and a most suitable polygon is selected as the best solution for arranging at the moment according to the credibility function of the knowledge driving model. In the specific implementation process, instead of adopting the mode at the beginning of discharging, a plurality of polygons are discharged first and then the mode is adopted, because when the number of the arranged polygon samples is too small, the intersection area of the polygons and the fabric expansion outline is too much relied on when the matching degree is obtained, which is not an intended effect. For example, four polygons are firstly discharged, then comparison discharge is carried out by traversing the polygons, and a specific optimized discharging flow is shown in fig. 9.
In addition, considering some special cases, a certain polygonal shape is special compared with other polygons, and the calculated value of the local fitness function fitness is not ideal all the time in the discharging process, so that the placing sequence of the special polygon can be continuously delayed, but because the special polygon is relatively large in area, a more optimal placing position is more difficult to find in the subsequent process. To avoid this, when evaluating the placement position goodness, it is not possible to rely on the fitness only, denoted as f, and it is necessary to add an index to the confidence function of the knowledge driven model, the waiting times waitTime denoted as w, and the polygon to be excluded lifting parameter performance improvement denoted as p. w represents the number of times the polygon has participated in the comparison (i.e. the number of times the polygon is selected to wait for discharge in the candidate set), in order to obtain p, the highest obtained fitness f of the polygon that fails in competition needs to be recorded during each round of polygon comparison, the fitness f is updated during each round of iterative comparison, if the highest obtained fitness f is obtained for a round of iteration New type Higher than the recorded history optimal fitness f, denoted as f History optimization The polygon to be arranged is considered to be significantly lifted, and the fit degree p is equal to f New type -f History optimization If the polygon to be arranged is not significantly improved, p is 0, and the result of the credibility is not influenced, so that the credibility considers the local part of the sample waferThe fitness also considers individual special samples, no polygon exists to continuously enter the candidate set to be discharged for many times, and when the waiting times are improved or the individual performance is greatly improved, a larger probability is selected as a final solution in the discharging.
In addition, according to knowledge 3, if the sample is discharged in an open boundary container, the situation that the tail of the polygon placement result is uneven (for example, the sample is initially discharged), so that in each iterative optimization process, the sample is discharged in a fixed rectangle, and the phenomenon that the tail of the sample discharge result is uneven is greatly solved.
In addition, according to the knowledge 1, if the polygons are not placed in the discharging process, that is, the iterative optimization fails, the order of the polygons which cannot be placed in the initial sequence of the polygons is considered to be improved, so that the polygons can participate in the discharging process earlier, and the polygons are more easily discharged to the proper positions.
S34: repeating S33 to discharge the remaining two-dimensional polygonal cloth sample to obtain optimized discharge results of the two-dimensional polygonal cloth samples, wherein the optimized primary discharge results comprise the optimized utilization rate of the fabric and the optimized length of the fabric;
s35: judging the optimized discharging result, if the optimized length of the fabric is greater than the length to be discharged, failing the optimization, and advancing the sequence of two-dimensional polygonal cloth sample exceeding the length to be discharged to obtain a new arrangement sequence, and repeating S32-S34; if the optimized length of the fabric is equal to or smaller than the length to be arranged, the optimization is successful, the arrangement sequence is recorded, the length to be arranged is reduced, and the optimization is continued by repeating the steps S32-S34.
S4: and taking the optimal discharging result obtained in the preset time as a final discharging result of a plurality of two-dimensional polygonal cloth sheet samples, wherein the optimal discharging result is a primary discharging result or an optimal discharging result after multiple times of optimization.
And S4, calculating the credibility of the optimal placement positions of the polygons to be arranged by using a knowledge driving model, and specifically calculating by using the following formula:
s＝c 1 *f+c 2 *w+c 3 *p
wherein s represents the reliability of the optimal placement position of the current polygon to be ranked, w represents the waiting times of the current polygon to be ranked, i.e. the waiting times become the polygon to be ranked but are not ranked, f represents the fitting degree of the optimal placement position of the current polygon to be ranked, p represents the lifting parameter of the polygon to be ranked, c 1 ，c 2 ，c 3 First, second and third constant coefficients, respectively.
And reading an optimal discharging result in a preset time, outputting the polygon vertex coordinates as txt files, converting the txt files into dxf files, and feeding the dxf files back to a final discharging result of the terminal for subsequent fabric cutting.
The result of discharging the sofa sample (sample 1) is shown in FIG. 10, and the utilization rate of the discharging result is 92.57% (discharging time: 3 min).
Similarly, the result of discharging (utilization 93.12%) of the other sofa sample (sample 2) is shown in fig. 11, and the result of discharging (utilization 83.32%) of the one complex sample (sample 3) is shown in fig. 12.
In summary, the embodiment of the invention can obtain a better discharging result in a short time, and compared with the traditional intelligent optimization algorithm, the algorithm is more concise and efficient. In the face of more regular sample data such as sofa-like sample pieces (sample 1, sample 2), excellent results can be obtained in a short time; and simultaneously, the method can also show excellent discharging results for a more complex sample (sample 3). The algorithm can be suitable for actual production environments, and has great significance in improving production workshop efficiency and reducing production cost.
The foregoing embodiments are merely illustrative of the principles and functions of the present invention, and are not in limitation thereof. Modifications and variations may be made to the above-described embodiments by those skilled in the art without departing from the spirit and scope of the invention. Therefore, all equivalent modifications and changes that can be made without departing from the spirit and technical spirit of the present invention are intended to be covered by the appended claims.
Claims (3)
1. The two-dimensional polygonal cloth piece automatic discharging method based on knowledge driving is characterized by comprising the following steps of:
s1: according to the two-dimensional polygon cloth sample, calculating to obtain the external critical polygons between every two-dimensional polygon cloth samples;
s2: the two-dimensional polygonal cloth sample is arranged in a descending order according to the areas of the two-dimensional polygonal cloth sample, and the positions in the fabric are placed one by one according to the arrangement order; calculating an inscribed critical polygon of the two-dimensional polygonal cloth sample to be arranged relative to the fabric and an circumscribed critical polygon of the two-dimensional polygonal cloth sample to be arranged relative to the arranged two-dimensional polygonal cloth sample according to the two-dimensional polygonal cloth sample to be arranged, and calculating and judging the placement position by using a local fitness model to obtain the optimal placement position of the two-dimensional polygonal cloth sample to be arranged and obtain the primary discharge results of a plurality of two-dimensional polygonal cloth samples;
s3: according to the primary discharging results of the two-dimensional polygonal sheet samples, optimizing the arrangement sequence and the optimal placement position of the two-dimensional polygonal sheet samples by utilizing a local fitness model and a knowledge driving model to obtain optimized discharging results of the two-dimensional polygonal sheet samples, and continuously optimizing the optimized discharging results of the two-dimensional polygonal sheet samples;
s4: taking the optimal discharging result obtained in the preset time as the final discharging result of a plurality of two-dimensional polygonal cloth sample;
the step S2 is specifically as follows:
s21: the two-dimensional polygonal cloth sample is arranged in a descending order according to the areas of the two-dimensional polygonal cloth sample, and the positions in the fabric are placed one by one according to the arrangement order; firstly, placing a first two-dimensional polygonal cloth sample into a fabric and taking the first two-dimensional polygonal cloth sample as a arranged polygon, and taking the next two-dimensional polygonal cloth sample to be arranged as a polygon to be arranged;
s22: calculating an inscribed critical polygon of the polygon to be arranged for the fabric, marking the inscribed critical polygon as an inscribed critical polygon track IFP, and marking an circumscribed critical polygon of the polygon to be arranged for all the arranged polygons as an circumscribed critical polygon track NFP;
s23: performing difference calculation on the external critical polygon track NFP through the internal critical polygon track IFP to obtain a track polygon of the current polygon to be arranged;
s24: calculating the fitting degree of each vertex of the track polygon by using a local fitting degree function, taking the vertex with the largest fitting degree as an optimal placement position, placing the current polygon to be arranged to the optimal placement position, and updating the arranged polygon and the polygon to be arranged;
s25: repeating S22-S24 to discharge the residual two-dimensional polygonal cloth sample, and obtaining primary discharge results of a plurality of two-dimensional polygonal cloth samples, wherein the primary discharge results comprise primary lengths of fabrics;
the step S3 is specifically as follows:
s31: determining the primary length of the fabric according to the primary discharging result of the two-dimensional polygonal cloth sample, wherein the length to be discharged of the fabric is smaller than the primary length, and arranging the two-dimensional polygonal cloth sample in a descending order according to the areas of the two-dimensional polygonal cloth sample;
s32: placing a first two-dimensional polygonal cloth sample into the fabric and taking the first two-dimensional polygonal cloth sample as a arranged polygon, and taking a plurality of two-dimensional polygonal cloth samples behind the arranged polygon as polygons to be arranged;
s33: calculating the optimal placement positions of all the polygons to be arranged by using a local fitness model, calculating the credibility of the optimal placement positions of all the polygons to be arranged by using a knowledge driving model, arranging the polygons to be arranged with the largest credibility into the sequence, returning the rest polygons to be arranged into the sequence, and updating the arranged polygons and the polygons to be arranged;
s34: repeating S33 to discharge the remaining two-dimensional polygonal cloth sample, so as to obtain optimized discharge results of the two-dimensional polygonal cloth samples, wherein the optimized primary discharge results comprise optimized lengths of the fabrics;
s35: judging the optimized discharging result, if the optimized length of the fabric is greater than the length to be discharged, failing the optimization, and advancing the sequence of two-dimensional polygonal cloth sample exceeding the length to be discharged to obtain a new arrangement sequence, and repeating S32-S34; if the optimized length of the fabric is equal to or smaller than the length to be arranged, the optimization is successful, the arrangement sequence is recorded, the length to be arranged is reduced, and the optimization is continued by repeating the steps S32-S34.
2. The knowledge-driven two-dimensional polygonal cloth piece automatic discharging method according to claim 1, wherein the S1 specifically comprises:
s11: performing de-duplication operation on the plurality of two-dimensional polygonal cloth samples to obtain a repeated polygonal cloth set S1 and a de-duplication polygonal cloth set S2, wherein the repeated polygonal cloth set S1 is a set formed by repeated two-dimensional polygonal cloth samples in the plurality of two-dimensional polygonal cloth samples, and the de-duplication polygonal cloth set S2 is a set formed by removing redundant repeated two-dimensional polygonal cloth samples in the plurality of two-dimensional polygonal cloth samples;
s12: the two-dimensional polygonal cloth sample in the repeated polygonal cloth set S1 or the de-repeated polygonal cloth set S2 is marked as (P) i ) I=1, 2, …, L, i is the sequence number of the two-dimensional polygonal patch sample in the repeating polygonal patch set S1 or the deduplication polygonal patch set S2, L is the total number of the two-dimensional polygonal patch samples in the repeating polygonal patch set S1 or the deduplication polygonal patch set S2; calculating to obtain external critical polygons of the repeated polygon piece set S1 and the de-duplicated polygon piece set S2, and obtaining external critical polygons between every two-dimensional polygon piece samples;
for the repeated polygon patch set S1, a two-dimensional polygon patch sample (P i ) External critical polygons to itself; for the deduplication polygon patch set S2, a two-dimensional polygon patch sample (P i ) With the rest two-dimensional polygonal cloth sample P 1 ,P 2 …P i-1 ,P i+1 …P L And the critical polygons are externally connected between every two of the two.
3. The knowledge-driven two-dimensional polygon cloth automatic discharging method according to claim 1, wherein the confidence level of the optimal placement position of each polygon to be discharged is calculated by using a knowledge-driven model in S4, specifically by the following formula:
s＝c 1 *f+c 2 *w+c 3 *p
wherein s represents the reliability of the optimal placement position of the current polygon to be ranked, w represents the waiting times of the current polygon to be ranked, f represents the fitness of the optimal placement position of the current polygon to be ranked, p represents the lifting parameter of the polygon to be ranked, c 1 ，c 2 ，c 3 First, second and third constant coefficients, respectively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110855739.0A CN113592174B (en) | 2021-07-28 | 2021-07-28 | Knowledge-driven two-dimensional polygonal cloth piece automatic discharging method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110855739.0A CN113592174B (en) | 2021-07-28 | 2021-07-28 | Knowledge-driven two-dimensional polygonal cloth piece automatic discharging method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113592174A CN113592174A (en) | 2021-11-02 |
| CN113592174B true CN113592174B (en) | 2024-01-05 |
Family
ID=78251101
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110855739.0A Active CN113592174B (en) | 2021-07-28 | 2021-07-28 | Knowledge-driven two-dimensional polygonal cloth piece automatic discharging method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113592174B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115146332B (en) * | 2022-07-25 | 2024-04-12 | 广州市圆方计算机软件工程有限公司 | Wood ceiling material discharging optimization method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105989605A (en) * | 2016-03-04 | 2016-10-05 | 安徽工程大学 | Irregular part blanking layout positioning method |
| CN106971242A (en) * | 2017-03-21 | 2017-07-21 | 上海大学 | A kind of clothes Automatic Optimal discharging method |
| CN110457756A (en) * | 2019-07-15 | 2019-11-15 | 天津大学 | One kind being based on critical rupture stress plate automatic nesting method |
| CN110598893A (en) * | 2019-07-18 | 2019-12-20 | 山东大学 | Multi-specification part layout method and system |
-
2021
- 2021-07-28 CN CN202110855739.0A patent/CN113592174B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105989605A (en) * | 2016-03-04 | 2016-10-05 | 安徽工程大学 | Irregular part blanking layout positioning method |
| CN106971242A (en) * | 2017-03-21 | 2017-07-21 | 上海大学 | A kind of clothes Automatic Optimal discharging method |
| CN110457756A (en) * | 2019-07-15 | 2019-11-15 | 天津大学 | One kind being based on critical rupture stress plate automatic nesting method |
| CN110598893A (en) * | 2019-07-18 | 2019-12-20 | 山东大学 | Multi-specification part layout method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113592174A (en) | 2021-11-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Gao et al. | A semi-supervised convolutional neural network-based method for steel surface defect recognition | |
| CN110111297B (en) | Injection molding product surface image defect identification method based on transfer learning | |
| Pan et al. | Efficient nearest-neighbor computation for GPU-based motion planning | |
| CN110569901A (en) | Channel selection-based countermeasure elimination weak supervision target detection method | |
| CN111429415B (en) | Method for constructing efficient detection model of product surface defects based on network collaborative pruning | |
| CN109948742B (en) | Handwritten picture classification method based on quantum neural network | |
| CN113592174B (en) | Knowledge-driven two-dimensional polygonal cloth piece automatic discharging method | |
| CN108776845B (en) | Mixed fruit fly algorithm based on dual-target job shop scheduling | |
| CN110889552B (en) | Apple automatic boxing path optimization method based on optimal parameter genetic algorithm | |
| CN110399917B (en) | Image classification method based on hyper-parameter optimization CNN | |
| CN105976421A (en) | Rendering program online optimization method | |
| CN109063917A (en) | Sublevel cloth Nesting based on genetic Optimization Algorithm | |
| Zeng et al. | Steel sheet defect detection based on deep learning method | |
| CN116560313A (en) | Genetic algorithm optimization scheduling method for multi-objective flexible job shop problem | |
| Haochen et al. | CNN-based model for pose detection of industrial PCB | |
| CN114819728A (en) | Flexible workshop production scheduling method capable of self-adaptive local search | |
| CN109074348A (en) | For being iterated the equipment and alternative manner of cluster to input data set | |
| CN106780636B (en) | Sparse reconstruction method and device for image | |
| CN116561710B (en) | Welding parameter transfer learning prediction method based on data space conversion | |
| Zheng et al. | Data-driven optimization based on random forest surrogate | |
| CN112942837A (en) | Cantilever structure concrete 3D printing method and system | |
| CN110705650B (en) | Sheet metal layout method based on deep learning | |
| Arora et al. | Design of a production system using genetic algorithm | |
| CN114996781A (en) | Two-dimensional irregular part layout method and system based on actors-critics | |
| Zhang et al. | Research on steel surface defect detection based on YOLOv5 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |