KR102162088B1 - Method for auto nest building to using genetic algorithm smart auto nesting of programs - Google Patents

Abstract

The present invention relates to a method for performing auto-nesting by a smart auto-nesting program using a genetic algorithm. The method comprises the steps of: (S10) converting drawings for a base material (1) and a material (2) which are written in 2D CAD drawings into a scalable vector graphics (SVG) format and storing the same on a server; and verifying, on the server, whether or not shape information is normally formed in the SVG drawings of the base material (1) and the material (2) having been converted into the SVG format. According to the present invention, the waste of a base material space can be minimized.

Description

Method for auto nesting to using genetic algorithm smart auto nesting of programs}

The present invention relates to a method of auto-nesting with a smart auto-nesting program using a genetic algorithm, and converts a drawing written in 2D CAD into a drawing in SVG format, and uses genetic algorithms, IFP algorithms, and NFP algorithms to achieve maximum in one parent material. Since a lot of materials are arranged, the consumption of the base material is reduced, and the material is automatically arranged within the rim of the base material through genetic algorithms, IFP algorithms and NFP algorithms, so the maximum amount of materials can be arranged on the base material, thereby wasting space for the base material. It is about a method of auto-nesting with a smart auto-nesting program using genetic algorithms that can minimize and do not require manual labor.

In general, the nesting algorithm is an optimal member placement algorithm, and it is an element technology that has a great ripple effect on improving price competitiveness and productivity in a wide variety of industries such as the shipbuilding industry, automobile industry, and clothing industry that cut, process and assemble a member's original plate. can do. Efficient batching algorithms have very high economical utilization in the sense of economic utilization of the base material, thus reducing production time and improving productivity. The general goal of optimal placement is to minimize the scrap ratio of the materials used, but in recent years, it has been widely used for purposes such as placement considering various constraints or 3D placement for efficient loading of goods in warehouses and containers. It is being utilized.

Until now, many industries have focused on developing nesting algorithms. However, the research on the development of the nesting algorithm applicable to the shipbuilding industry focused on the excellence of work rather than a direct study on the nesting methodology.

In the case of nesting using such a genetic algorithm, materials can be placed within a limited frame of the parent material using a genetic algorithm, but there is a problem in that it is difficult to control when it is out of or overlapped with the border.

In addition, when a large amount of materials are arranged on the base material, there is a problem in that an additional arrangement is required manually for detailed arrangement.

Prior Patent 1 relates to a bitmap conversion unit that obtains member information from a design drawing for nesting, and converts members and a thick plate on which the members are to be placed into a bitmap format of a preset resolution using the obtained member information; A virtual diagonal line is set on the thick plate on which the members converted in bitmap form are to be placed, and the distance between the piercing points of the members sequentially selected according to the arrangement order and the set virtual diagonal lines is the minimum, and the remaining members are A member arranging unit for sequentially arranging the selected member at a position having the greatest possible arrangement; And for the arranged members, cutting the member by calculating the starting position at which to start cutting and the distance between the cutting points of the arranged members, and selecting the point where the calculated distance is the minimum as the cutting point, Among them, a configuration including a member cutting portion for cutting a member by sequentially selecting a cutting point having a minimum adjacent distance is described.

However, Prior Patent 1 converts the thick plate into a bitmap so that the members to be cut on the thick plate are sequentially arranged along an imaginary diagonal, so that the members arranged on the thick plate are arranged at an optimal distance and position to save the remaining space of the thick plate. There is a problem that is difficult to minimize.

In addition, since the members are sequentially arranged, the position of the members overlapping or deviating from the thick plate must be corrected through a separate manual operation, which takes a long time and has a complicated problem.

Prior patent 2 creates a plurality of parent material arrangement plans in which the target member is disposed at various positions for a base material having a dispositioned member when placing the next target member of the displaced member, and for a base material not having the displaced member A design plan generation step of generating a base material layout plan in which the target member is placed, and a plurality of design proposals after the target member is disposed, a design proposal limiting step of limiting a design plan for subsequent member placement among the plurality of generated designs, and all After arranging the members, a configuration is described including a design plan selection step of selecting a design with the best yield among the generated designs.

However, Prior Patent 2 is a configuration in which a number of drawings in which the target members are arranged on the base material are generated and the drawing with the best yield is selected. Since the drawing with the highest yield must be selected from among the plurality of drawings created for selecting the optimum position of the member, It is difficult to select an accurate drawing, and since the drawing is not automatically selected, it takes a long time to select.

Prior Patent 3 is to develop an arrangement algorithm suitable for nesting in the shipbuilding industry, applying a genetic algorithm to the part of optimizing the arrangement order, and an improved NFP (no fit polygon) method to arrange for a given order. The configuration using the) method is described.

However, prior patent 3 has a problem in that the arrangement is made with only the genetic algorithm and the NFP, and thus the region of the base material is not considered when nesting because it is arranged regardless of the relationship with the base material.

In addition, there is a problem in that it is difficult to calculate the required space between the material and the material when cutting the material, and since it is not possible to arrange it after alignment according to the limit of the angle and the size of the material, there is a problem that must be rearranged manually.

Prior Patent 1: Korean Patent Publication No. 10-1752099 (2017.06.22.) Prior Patent 2: Korean Patent Registration No. 10-1871095 (2018.06.19.) Prior Patent 3: Research on Optimal Arrangement of Arbitrary Shape Members Using Genetic Algorithm and No Fit Polygon Method / Byung-Hang Yoo, Busan: Graduate School, Pukyong National University, 2002.8

The present invention converts a drawing written in 2D CAD into a drawing in SVG format to arrange as many materials as possible in one parent material through genetic algorithms, IFP algorithms and NFP algorithms, so that the consumption of the base material is reduced, and genetic algorithms, IFP algorithms And NFP algorithm to automatically arrange materials within the rim of the base material, so the maximum amount of materials can be arranged on the base material, minimizing waste of base material space, and a smart auto-nesting program using genetic algorithms that do not require manual work. Its purpose is to provide a method of auto-nesting.

In the method of auto-nesting with a smart auto-nesting program using a genetic algorithm according to the present invention to achieve the above object, a drawing of a base material and a material created as a 2D CAD drawing on a server is converted to a SVG (Scalable Vector Graphics) format. Converting and storing; Verifying, at a server, whether shape information is normally performed in the SVG (Scalable Vector Graphics) drawing of the base material and the material converted to the SVG format; Inputting shape information for the verified SVG (Scalable Vector Graphics) drawing to a server; Arranging the shape information of the SVG drawing input to the server such that as many materials are located in the base material as much as possible through a genetic algorithm; Arranging the position of the material in the aligned base material so as not to overlap within a range that does not deviate from the base material through an IFP algorithm (Inner Fit Polygon Algorithm) and an NFP algorithm (No Fit Polygon Algorithm); If the materials are arranged so that they do not overlap in the base material, calculating and arranging Fitness so that the materials are arranged in a minimum space by adjusting the positions of the materials so that as many materials as possible are arranged; The step of arranging, arranging so as not to overlap, and arranging so that the material can be located in the base material as much as possible by repeating the step of arranging and arranging fitness multiple times, and arranging the SVG drawing completed through the arranging step It characterized in that it comprises the step of storing, converting and storing the 2D CAD drawing to be applied to the facility.

In the step of verifying in the server, it is characterized in determining whether the SVG drawing is in a proper format, determining whether the closed shape of the SVG drawing is normal, and determining the relationship between the base material and the material in the SVG drawing.

In the step of inputting the shape information to the server, the shape, quantity, plate availability, identification ID of the base material and material, and the gap difference between the base material and the material displayed on the SVG drawing are input to the server.

The aligning step is characterized by arranging the materials in the base material by calculating the order and angle in the server so that as many materials as possible are located within the range of the base material.

The present invention converts a drawing written in 2D CAD into a drawing in SVG format so that as many materials as possible are arranged in one parent material through genetic algorithm, IFP algorithm, and NFP algorithm, there is an effect of reducing the consumption of the base material.

In addition, since materials are automatically arranged within the rim of the base material through genetic algorithms, IFP algorithms and NFP algorithms, the maximum amount of materials can be arranged on the base material, minimizing the waste of base material space, and eliminating the need for manual work. have.

1 is a diagram showing a procedure of a method of auto-nesting with a smart auto-nesting program using a genetic algorithm according to an embodiment of the present invention.
FIG. 2 is a diagram illustrating that materials of an SVG drawing auto-nested through a method of auto-nesting with a smart auto-nesting program using a genetic algorithm according to the present invention are arranged on a base material.
FIG. 3 is a view showing that more materials of the SVG drawing auto-nested through the auto-nesting method using the genetic algorithm according to the present invention are arranged on the base material than in FIG. 4.
4 is a view showing that the material of the SVG drawing auto-nested by the method of auto-nesting with a smart auto-nesting program using a genetic algorithm according to the present invention is optimized and arranged in a base material.

Hereinafter, specific embodiments for carrying out the present invention will be described with reference to the drawings. Embodiments of the present invention are for explaining one invention, and the scope of rights is not limited to the illustrated embodiments, and the illustrated drawings are limited to the drawings because only the essential content is enlarged and ancillary items are omitted for clarity of the invention. And should not be interpreted.

The present invention is a step of converting and storing the drawings of the base material (1) and the material (2) created as 2D CAD drawings in a server into SVG (Scalable Vector Graphics) format (S10); Verifying at the server whether the shape information is normally performed in the SVG (Scalable Vector Graphics) drawing of the base material 1 and the material 2 converted to the SVG format (S20); Inputting shape information for the verified SVG (Scalable Vector Graphics) drawing to a server (S30); Arranging the shape information of the SVG drawing input to the server so that the material 2 is located as much as possible in the base material 1 through a genetic algorithm (S40); Arranging the position of the material (2) within the aligned base material 1 so that it does not overlap within a range that does not deviate from the base material 1 through an IFP algorithm (Inner Fit Polygon Algorithm) and an NFP algorithm (No Fit Polygon Algorithm) ( S50); If the material (2) is arranged so that it does not overlap within the base material (1), the position of the material (2) is adjusted so that the material (2) is placed in the minimum space, so that the maximum number of materials (2) is placed. Calculating and placing (S60); The step of arranging the material 2 to be located in the base material 1 as much as possible by repeating the step of aligning, the step of arranging so as not to overlap, and the step of calculating and arranging fitness (S60) (S70) and It characterized in that it comprises a step (S80) of storing the SVG drawing that has been arranged through the arranging step (S70), and converting and storing it into a 2D CAD drawing so that it can be applied to the facility.

In the step of verifying at the server (S20), it is determined whether the SVG drawing is in a proper format, it is determined whether the closed shape of the SVG drawing is normal, and the relationship between the base material (1) and the material (2) in the SVG drawing is determined. It characterized in that the discrimination.

In the step of inputting the shape information to the server (S30), the shape, quantity, plate availability, identification ID of the base material 1 and material 2, and the base material ( It is characterized by inputting the gap difference between 1) and the material (2) in the server.

The aligning step is characterized by arranging the materials 2 in the base material 1 by calculating the order and angle in the server so that the material 2 is located as much as possible within the range of the base material 1.

1 is a diagram showing the sequence of a method of auto-nesting with a smart auto-nesting program using a genetic algorithm according to an embodiment of the present invention, a base material (1) and a material (2) created as a 2D CAD drawing on a server Step of converting and saving the drawing for the SVG (Scalable Vector Graphics) format (S10), whether the shape information has been properly performed in the SVG (Scalable Vector Graphics) drawing of the base material (1) and material (2) converted to SVG format The step of verifying whether or not in the server (S20), the step of inputting the shape information of the verified SVG (Scalable Vector Graphics) drawing to the server (S30), the shape information of the SVG drawing input to the server through a genetic algorithm 1) The step of aligning the material (2) as much as possible (S40), the position of the material (2) in the base material (1) to be aligned using IFP algorithm (Inner Fit Polygon Algorithm) and NFP algorithm (No Fit Polygon Algorithm). The step of arranging so as not to overlap within the range that does not deviate from the base material (1) through (S50), if the material (2) is arranged so that it does not overlap within the base material (1), the position of the material (2) is adjusted to minimize space The steps of calculating and arranging Fitness so that the material (2) is arranged and the maximum number of materials (2) are arranged (S60), the step of aligning, the step of arranging so as not to overlap, and the step of calculating and arranging Fitness (S60) ) By repeating a number of times to arrange the material (2) so that it can be located in the base material (1) as much as possible (S70) and the arrangement (S70) to save the arrangement completed SVG drawing as a 2D CAD drawing. It includes a step (S80) of converting and storing to be applied to the facility.

2 is a view showing that the material of the SVG drawing auto-nested through the method of auto-nesting with a smart auto-nesting program using a genetic algorithm according to the present invention is arranged on a parent material, and FIG. 3 is a gene according to the present invention. It is a diagram showing that the material of the SVG drawing auto-nested by the method of auto-nesting with a smart auto-nesting program using an algorithm is arranged on the base material more than that of FIG. 4, and FIG. 4 is a smart This is a drawing showing that the material of the SVG drawing that is auto-nested through the auto-nesting method is optimized and arranged in the base material.

Referring to FIG. 1, the converting and storing step (S10) is a step of converting and storing drawings of the base material (1) and the material (2) created as a 2D CAD drawing in the server into SVG (Scalable Vector Graphics) format. It is (S10).

In this case, the 2D CAD drawing is a drawing showing the shapes of the base material 1 and the material 2, respectively, and the server converts them to SVG (Scalable Vector Graphics) format so that they are all arranged on one drawing.

When the 2D CAD drawing is converted to the SVG format drawing through the converting and storing step (S10), the SVG (Scalable) of the base material (1) and the material (2) converted to SVG format through the step (S20) of verifying in the server Vector Graphics) The server verifies whether the shape information in the drawing has been properly performed.

In the step of verifying at the server (S20), it is determined whether the SVG drawing is in a proper format, it is determined whether the closed shape of the SVG drawing is normal, and the relationship between the base material (1) and the material (2) in the SVG drawing is determined. Is determined.

The SVG drawing that has not passed the verification in the verification step (S20) in the server is converted and saved in the step (S10) of re-converting and storing the modified other drawing of the corresponding part in the server, and this is verified again. .

At this time, the SVG (Scalable Vector Graphics) format is used to draw graphics in an XML language similar to XHTML, and the SVG image using this is either manually specifying all necessary lines and shapes, or modifying an existing raster image, or both. It can be made by combining methods.

In addition, in the SVG format image, the image and its components may be transformed, combined, and filtered, and the original form may be completely changed through this.

This SVG format is an XML-based standard graphic format, and animation can be applied to all elements and attributes of SVG, and it is configured based on vector, so it is not affected by resolution.

In addition, it has the advantage of being able to freely control all elements by using the Jvascript SVG DOM API, but there is a characteristic that rendering is slower as the document complexity increases.

When the SVG drawing is verified through the verifying step (S20) in the server, shape information for the SVG (Scalable Vector Graphics) drawing verified through the inputting step (S30) to the server is input to the server.

In the step of inputting the shape information to the server (S30), the shape, quantity, plate availability, identification ID of the base material (1) and material (2), and the base material (2) displayed in the SVG drawing The gap difference between (1) and material (2) is entered into the server.

The reason for inputting the shape information of the base material (1) and the material (2) to the server is that, when the material of the SVG drawing is arranged through nesting, the information is changed according to the use of the processed material when it is actually processed. It is necessary so that you can easily find the material (2) by entering it.

2 to 4, when the shape information of the SVG drawing is input to the server through the step of inputting the shape information to the server (S30), the shape information of the SVG drawing input to the server is stored through the alignment step. Through the genetic algorithm (Genetic Algorithm), the material (2) is arranged to be located in the parent material (1) as much as possible.

At this time, the genetic algorithm is a search algorithm for solving an optimization problem by applying the principle of evolution, which is an engineering algorithm that applies the evolutionary process in which only the most suitable individuals survive in the natural world (J Holland). 1992)

Since this genetic algorithm was introduced by Holland in 1975, it has been actively applied to the traveling salesman problem, the combination optimization problem, the scheduling problem, and the transportation problem.

That is, the angle can be changed through rotation of the material 2 disposed within the rim of the base material 1 through the genetic algorithm.

In the step of aligning, the order and angle to be entered so that as many materials 2 are located within the range of the base material 1 are calculated by the server to arrange the materials 2 in the base material 1.

When the material (2) is aligned so as to be located in the base material (1) as much as possible through the alignment step, the position of the material (2) in the base material (1) aligned through the step of arranging not to overlap is determined by the IFP algorithm (Inner Fit Polygon). Algorithm) and NFP algorithms (No Fit Polygon Algorithm) are arranged so that they do not overlap within a range that does not deviate from the base material (1).

At this time, the IFP algorithm (Inner Fit Polygon Algorithm) checks whether the material 2 deviates within the frame of the base material 1, and adjusts the position of the material 2 so that it is located within the border of the base material 1. .

The IFP algorithm (Inner Fit Polygon Algorithm) is an algorithm that calculates an optimal path so that the material 2 does not deviate from the edge of the base material 1.

And the NFP algorithm (No Fit Polygon Algorithm) adjusts the position so that the material 2 does not overlap each other within the rim of the base material 1.,

The NFP algorithm (No Fit Polygon Algorithm) is an algorithm to find a path so that the material 2 does not overlap each other within the border of the base material 1.

When the position of the material 2 is placed within the rim of the base material 1 through the step of arranging so as not to overlap, the minimum space by adjusting the position of the material 2 through the step (S60) of calculating and arranging fitness As the material (2) is placed, fitness is calculated and placed so that as many materials (2) as possible are placed.

If the material 2 is arranged as much as possible in the base material 1 through the step (S60) of calculating and arranging the fitness, the step of arranging through the step (S70) of arranging to be located as much as possible, so as not to overlap By repeating the step of arranging and the step (S60) of calculating and arranging Fitness (S60) a number of times, the material 2 is arranged so as to be located in the base material 1 as much as possible.

The reason for arranging so that the material 2 can be located in the base material 1 as much as possible through the arranging step (S70) in this way is that the base material 1 is consumed as much as possible and a large number of materials 2 are used as the base material 1 This is to reduce the additional consumption of the base material 1 by placing it in one.

At this time, when it is determined that as many materials 2 as possible within the border of the base material 1 are arranged, the steps of aligning, arranging not to overlap, and calculating and arranging Fitness (S60) are repeated. When is generated several times, the server automatically determines that it has been arranged and ends.

When the material 2 is arranged to be located in the base material 1 as much as possible through the arranging step (S70), the SVG drawing is stored in the server through the step (S80) to be applied to the facility, and 2D Converted to CAD drawing and saved so that it can be applied to the facility.

The present invention made in this way converts a drawing written in 2D CAD into a drawing in SVG format, so that as many materials as possible are arranged in one base material through genetic algorithm, IFP algorithm, and NFP algorithm, there is an effect of reducing the consumption of the base material. .

In addition, since materials are automatically arranged within the rim of the base material through genetic algorithms, IFP algorithms and NFP algorithms, the maximum amount of materials can be arranged on the base material, minimizing the waste of base material space, and eliminating the need for manual work. have.

The method of auto-nesting with the smart auto-nesting program using the genetic algorithm as described above is not limited to the configuration and operation method of the above-described embodiments. The above embodiments may be configured so that all or part of each of the embodiments may be selectively combined to make various modifications.

1: base material 2: material

Claims (4)

Converting and storing the drawings of the base material (1) and the material (2) created as 2D CAD drawings in the server into SVG (Scalable Vector Graphics) format (S10);
Verifying at the server whether the shape information is normally performed in the SVG (Scalable Vector Graphics) drawing of the base material 1 and the material 2 converted to the SVG format (S20);
Inputting shape information for the verified SVG (Scalable Vector Graphics) drawing to a server (S30);
Arranging the shape information of the SVG drawing input to the server so that the material 2 is located as much as possible in the base material 1 through a genetic algorithm (S40);
Arranging the position of the material (2) within the aligned base material 1 so that it does not overlap within a range that does not deviate from the base material 1 through an IFP algorithm (Inner Fit Polygon Algorithm) and an NFP algorithm (No Fit Polygon Algorithm) ( S50);
If the material (2) is arranged so that it does not overlap within the base material (1), the position of the material (2) is adjusted so that the material (2) is placed in the minimum space, so that the maximum number of materials (2) is placed. Calculating and placing (S60);
The step of arranging the material 2 to be located in the base material 1 as much as possible by repeating the step of aligning, the step of arranging so as not to overlap, and the step of calculating and arranging fitness (S60) (S70) and
Smart Autone using a genetic algorithm, characterized in that it comprises the step (S80) of storing the SVG drawing that has been arranged through the arranging step (S70), and converting and storing it into a 2D CAD drawing to be applied to the facility. How to auto-nesting with a sting program.
The method of claim 1,
In the step of verifying at the server (S20), it is determined whether the SVG drawing is in a proper format, it is determined whether the closed shape of the SVG drawing is normal, and the relationship between the base material (1) and the material (2) in the SVG drawing is determined. Method for auto-nesting with a smart auto-nesting program using a genetic algorithm, characterized in that to determine.
The method of claim 1,
In the step of inputting the shape information to the server (S30), the shape, quantity, plate availability, identification ID of the base material 1 and material 2, and the base material ( A method of auto-nesting with a smart auto-nesting program using a genetic algorithm, characterized in that the difference between 1) and the material (2) is input to the server.
The method of claim 1,
The sorting step uses a genetic algorithm, characterized in that the order and angle to be entered so that the material 2 is located within the range of the base material 1 as much as possible is calculated by the server to align the material 2 in the base material 1 How to auto-nesting with the smart auto-nesting program.

Images