Segmented tire changing process optimization method based on genetic algorithm

Abstract

The invention provides a segmented tire replacement process optimization method based on a genetic algorithm, and the method comprises the steps: firstly obtaining production data in a tire position map, and extracting the position and height of each tire frame; depicting the occupied area of the jig frame according to the current situation of a production site, and determining a displacement domain of movement of the ship segments relative to the jig frame according to the boundary range of the jig frame, so as to determine the movable range of the segments on the site; and then the movement amount of the segments relative to the jig frame is optimized and solved based on a genetic algorithm, and the optimal movement amount is obtained by taking the minimum sum of the heights, needing to be adjusted, of the jig frames when the segments are replaced every time as the target. In the solving process based on the genetic algorithm, firstly, an initial population is randomly generated, genes of chromosomes in the initial population represent the movement amount of each segment, and individuals with high fitness in the initial population are selected as parent chromosomes; and then repeating crossover and variation processes, and finally reserving an individual with a relatively high fitness value as an optimal movement amount value.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing technology, and in particular to a method for optimizing a segmented tire changing process based on a genetic algorithm. Background Technology

[0002] In shipbuilding, the precise positioning of different ship sections on the jig is a crucial step in ensuring construction quality. Due to the complex spatial curvature of the section bottom surfaces, there are significant height differences between each positioning point and the ground reference plane. When different sections are placed on the jig, the jig needs to be adaptively adjusted to conform to the section's alignment, such as... Figure 1 As shown, during the tire changing process, the precise adjustment of the tire frame height not only affects the alignment of the segmented lines, but is also a decisive factor affecting the control of subsequent welding deformation and the overall assembly accuracy.

[0003] When assessing the workload of jig adjustment from an engineering management perspective, a multi-dimensional quantitative analysis system needs to be established. First, the structural dimensional parameters of the jig (including longitudinal span and transverse width) constitute the physical boundary conditions for segmental displacement adjustment, directly determining the adjustable range of the segment in three-dimensional space. Second, the configuration of the number of segment positioning points reflects both the structural scale of the segment and the complexity of the support system. Third, the cluster of height coordinates of the positioning points fully records the shape characteristics of the bottom surface of the segment. These three core elements interact to form the assessment model for the workload of jig adjustment.

[0004] In engineering, jig drawings are generally used to clearly define the positioning points of segments during the manufacturing process, ensuring that they maintain the correct position and posture during construction. By changing the relative position of the segment and the jig, such as adjusting a certain amount of movement along the jig's layout direction or height direction, the current segment can be made to fit the jig as closely as possible, thereby reducing the amount of jig adjustment. However, how to efficiently and accurately find the optimal amount of movement is a problem that technicians urgently need to solve. Summary of the Invention

[0005] In view of the shortcomings of the prior art described above, the present invention provides a method for optimizing the segmented tire changing process based on a genetic algorithm, comprising the following steps:

[0006] S1: Obtain production data from the tire position diagram and convert it into a data table. Extract the location and height of each tire frame. Multiple tire frames are arranged in the left-right direction.

[0007] S2: Describe the footprint of the jig based on the current production site conditions, and determine the displacement domain of the ship section relative to the jig based on the boundary range of the jig.

[0008] S3: Determine the number of segments. When changing to different segments, the height of each tire frame needs to be adjusted to fit the bottom line of the segment. Based on the genetic algorithm, optimize the movement of the segment relative to the tire frame. The goal is to minimize the sum of the height adjustments required for each tire frame when changing segments, and obtain the optimal movement amount.

[0009] Optionally, the jig includes angle steel and a flexible jig.

[0010] Optionally, step S3 includes:

[0011] Define the amount of displacement of the segment relative to the frame within the displacement domain. The amount of displacement includes the horizontal displacement x and the vertical displacement y.

[0012] An initial population is randomly generated, consisting of multiple chromosomes. The genes in the first chromosome are arranged in sequence as X1, X2, X3, ..., Xm, Y1, Y2, Y3, ..., Ym. Here, X1 represents the horizontal displacement x of the first segment relative to the frame, Xm represents the horizontal displacement x of the m-th segment relative to the frame, Y1 represents the vertical displacement y of the first segment relative to the frame, and Ym represents the vertical displacement y of the m-th segment relative to the frame.

[0013] Based on the gene values ​​in the first chromosome, a preset step size is added to each gene value to generate the second chromosome, and so on, to generate multiple chromosomes.

[0014] Optionally, step S3 further includes:

[0015] Define the objective function as follows:

[0016]

[0017] In the formula: m is the total number of segments involved in tire changing, and n is the number of tire frames; L mn L represents the ground clearance of the nth tire frame when the mth segment is fitted with the upper tire and is in contact with the tire frame. (m-1)n This represents the ground clearance of the nth tire frame when the upper tire is placed in the (m-1)th segment and is in contact with the tire frame;

[0018] The fitness function is F = a / f;

[0019] The fitness values ​​of chromosomes in the initial population are calculated, and individuals with high fitness in the initial population are selected as parent chromosomes. Single-point gene crossover is performed on the selected parent chromosomes to generate offspring chromosomes. Mutation is then performed, and the values ​​of single-point genes in the offspring chromosomes are randomly adjusted to enhance diversity. The selection, crossover, and mutation process is repeated until the predetermined number of iterations is reached or the stopping condition is met. Finally, individuals with high fitness values ​​are retained as the optimal movement value.

[0020] Optionally, the amount of movement also includes the angle value of the segment's rotation relative to the frame within the allowable range of the site, the angle value being a multiple of 90°, including 90°, 180°, and 270°; correspondingly, a gene is added to the chromosome to represent the angle value.

[0021] Optionally, the numerical values ​​represented by each gene in the chromosome are converted into binary codes, with each bit of the binary code serving as a child node, thus using multiple child nodes to replace a single gene.

[0022] As described above, this invention provides a method for optimizing a segmented tire-changing process based on a genetic algorithm. This method first obtains production data from the tire position map, extracting the location and height of each tire frame. Then, it delineates the tire frame's footprint based on the current production site conditions, and determines the displacement domain of the ship segment relative to the tire frame based on the tire frame's boundary range, thus determining the segment's movable range on the site. Next, it optimizes the segment's movement relative to the tire frame using a genetic algorithm, aiming to minimize the sum of height adjustments required for each tire frame during segment replacement, thereby obtaining the optimal movement amount. In the genetic algorithm's solution process, an initial population is first randomly generated, containing multiple chromosomes. Genes arranged sequentially on the chromosomes represent the movement amount of each segment. Genes at the same location on different chromosomes represent different movement amounts. Individuals with high fitness in the initial population are selected as parent chromosomes. The crossover and mutation process is then repeated until a predetermined number of iterations is reached or a stopping condition is met. Finally, individuals with higher fitness values ​​are retained as the optimal movement amount.

[0023] This invention uses a genetic algorithm for mathematical modeling, achieving better accuracy and reducing the likelihood of getting trapped in local optima compared to greedy algorithms and gradient descent. Compared to exhaustive search, its computational speed increases exponentially, significantly improving problem-solving efficiency. Furthermore, this method offers several improved solutions, providing greater flexibility in analyzing specific problems. Attached Figure Description

[0024] Figure 1 This is a diagram showing the comparison of the changes in the tire profile during segmented tire replacement.

[0025] Figure 2 This is a schematic diagram of segmented fetal position data.

[0026] Figure 3 The diagram shows the basic logic of a genetic algorithm.

[0027] Figure 4 The diagram shows the convergence process of the computational iteration. Detailed Implementation

[0028] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention.

[0029] It should be noted that the illustrations provided in this embodiment are only schematic representations of the basic concept of the present invention. Therefore, the illustrations only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0030] Smart manufacturing refers to the use of advanced information technologies, such as the Internet of Things, big data, and artificial intelligence, to optimize production processes, making manufacturing more intelligent, efficient, and flexible. It encompasses multiple aspects, including automated production lines, intelligent monitoring systems, and predictive maintenance. Genetic algorithms, as an optimization tool, are widely used in various stages of smart manufacturing, such as production planning, logistics, and quality control, improving overall efficiency and competitiveness.

[0031] The main idea of ​​genetic algorithms is to draw on Darwin's evolutionary model of natural selection. By referencing biological evolution, the problem to be solved is simulated as a process of biological evolution. Through operations such as replication, crossover, and mutation, the next generation of solutions is generated, and solutions with low fitness function values ​​are gradually eliminated, while solutions with high fitness function values ​​are added. After N generations of evolution, it is very likely that an individual with a very high fitness function value will evolve, which is the optimal result of the objective function value.

[0032] Traditional solution methods (such as gradient descent and Newton's method) often have limitations when dealing with nonlinear and nonconvex problems, and are prone to getting trapped in local optima. Genetic algorithms, as a global optimization algorithm that simulates natural selection and genetic mechanisms, have the following advantages: they do not depend on the gradient information of the problem and are suitable for discontinuous and nonconvex problems; they have global search capabilities and can effectively avoid getting trapped in local optima; they are robust and suitable for complex problems with multiple objectives and constraints.

[0033] The main idea of ​​this invention is to introduce a genetic algorithm to solve for the optimal movement of ship sections relative to the frame, so as to achieve efficient and accurate solution.

[0034] This invention provides a method for optimizing a segmented tire-changing process based on a genetic algorithm, comprising the following steps:

[0035] S1: Obtain production data from the fetal position chart and convert it into a table, such as... Figure 2 As shown, the location and height of each tire frame are extracted.

[0036] Specifically, the jig position diagram, serving as an instructional drawing for the segmented jig installation, details the location and specific height of each jig frame, indirectly reflecting the hull structure's lines and facilitating quick identification and installation by workers. The jig frame is the basic unit constituting the jig position; multiple jig frames are arranged along the left-right direction and generally take the form of angle steel or flexible jig frames. They are primarily used to support the bottom surface of the ship structure during jig installation, and their height is consistent with the bottom lines of the segment to ensure a close fit.

[0037] S2: Describe the footprint of the jig based on the current production site conditions, and determine the displacement domain of the ship section relative to the jig based on the boundary range of the jig. This determines the movable range of the section on the site.

[0038] S3: Determine the number of segments. When changing to different segments, the height of each tire frame needs to be adjusted to fit the bottom line of the segment. Based on a genetic algorithm, optimize the movement of the segments relative to the tire frame, aiming to minimize the sum of height adjustments required for each segment change, thus obtaining the optimal movement amount. For example... Figure 3 As shown, this step specifically includes:

[0039] Define the amount of movement of the segment relative to the jig within the allowable range of the site (displacement domain). The amount of movement includes horizontal displacement x (displacement in the left and right direction) and vertical displacement y (displacement in the up and down direction). Each segment of different size has its corresponding range of movement, which represents the limit of movement of the segment in the jig position, to prevent the segment from exceeding the jig position range.

[0040] An initial population is randomly generated, consisting of multiple chromosomes. The genes in the first chromosome are arranged in sequence as X1, X2, X3, ..., Xm, Y1, Y2, Y3, ..., Ym. Here, X1 represents the horizontal displacement x of the first segment relative to the frame, Xm represents the horizontal displacement x of the m-th segment relative to the frame, Y1 represents the vertical displacement y of the first segment relative to the frame, and Ym represents the vertical displacement y of the m-th segment relative to the frame.

[0041] Based on the gene values ​​in the first chromosome, a preset step size is added to each gene value to generate the second chromosome, and so on, generating multiple chromosomes. This yields the initial population, which includes various scenarios where each segment has different displacement amounts. The preset step size is selected according to actual needs, such as 0.2m or 0.5m.

[0042] As an example, the population initialization function can be expressed as:

[0043]

[0044] Among them, the chromosome length (gene loci) is 2N, a 1,1 a 1,2 a 1,3 ... a 1,N a 1,N+1 a 1,N+2 a 1,N+3 ... a 1,2N They correspond to X1, X2, X3, ..., Xm, Y1, Y2, Y3, ..., Ym respectively.

[0045] Define the objective function as follows:

[0046]

[0047] In the formula: m is the total number of segments involved in tire changing, and n is the number of tire frames; L mn L represents the ground clearance of the nth tire frame when the mth segment is fitted with the upper tire and is in contact with the tire frame. (m-1)n This represents the ground clearance of the nth tire frame when the upper tire is placed and aligned with the tire frame in the (m-1)th segment. For example, when the upper tire is placed and aligned with the tire frame in the first segment, the ground clearance of the first tire frame is 1m. Because the alignment of the second segment is different from the first segment, the height of the first tire frame in the second segment is adaptively adjusted to 1.5m when the upper tire is placed and aligned with the tire frame. Therefore, f increases by 0.5m. The total adjustment is obtained by summing the adjustments of each tire frame when the second segment replaces the first segment. This process is repeated for each of the m segments to obtain the total adjustment. The sum of these total adjustments is f.

[0048] The segment movement exceeding the displacement domain does not meet actual production conditions (segments cannot be suspended in the air), therefore, a penalty function is not considered; instead, the initial conditions of the population are directly constrained. Let the displacement domain be A, and the range of values ​​for decision variables X1, X2, X3, ..., Xm, Y1, Y2, Y3, ..., Ym be A. The fitness function is:

[0049] F = a / f

[0050] 'a' is a parameter; by adjusting 'a', the fitness value can be scaled to a reasonable range.

[0051] The fitness values ​​of each chromosome in the initial population are calculated, and individuals with high fitness are selected as parent chromosomes. Single-point gene crossover is performed on the selected parent chromosomes to generate offspring chromosomes. Mutation is then performed, randomly adjusting the values ​​of single-point genes in the offspring chromosomes to enhance diversity. A roulette wheel selection method can be used to further select individuals, ensuring that even highly fit individuals may be inherited, thus increasing genetic diversity.

[0052] Specifically, individual fitness serves as the selection criterion, and the selection strategy employs a roulette wheel selection method. The process of reproduction involves two steps: crossover and mutation. Crossover means that each individual is produced by the reproduction of both the father and mother, and the offspring's DNA inherits a portion of the father's DNA and a portion of the mother's DNA. The mating point is randomly generated and can be any position on the chromosome.

[0053] Through crossover, offspring chromosomes acquire characteristics from both the father and mother. However, offspring themselves may mutate, resulting in DNA that originates from neither the father nor the mother, but rather from a random mutation at a certain location. This mutation may lead to a shift beyond the defined domain, requiring modification of the mutated chromosome to conform to the domain's constraints.

[0054] Then repeat the selection, crossover, and mutation process, such as Figure 4 As shown, until the predetermined number of iterations is reached or the stopping condition is met, individuals with higher fitness values ​​are ultimately retained as the optimal movement value.

[0055] Furthermore, the amount of movement can also include the angle value of the segment's rotation relative to the frame within the allowable range of the site. This angle value is a multiple of 90°, such as 90°, 180°, or 270°. In this case, simply adding a gene to the chromosome to represent the angle value, using values ​​0, 1, and 2 to represent 90°, 180°, and 270° respectively, can achieve this. Rotating the angle can make the segment fit more closely to the frame, reducing the amount of frame height adjustment. For example, if the frame is arranged with the left side higher than the right, and the segment's bottom is also lower on the left and higher on the right before rotation, rotating the segment by 180 degrees will make both the frame and the segment fit more closely.

[0056] Furthermore, the numerical values ​​represented by each gene in the chromosome are converted into binary codes. Each bit of the binary code (taking the value 0 or 1) serves as a child node, thus using multiple child nodes to replace a single gene. Subsequent crossover and mutation operations are then performed on these child nodes. Using binary encoding allows for more accurate calculations, but it increases the computational load and reduces operating efficiency. The choice here can be made flexibly based on the actual situation.

[0057] Furthermore, based on integer genetic algorithms and traditional crossover mutation methods, various methods such as interval sampling, multi-point mutation, and superior gene recombination can be used. The specific application depends on the parameter dimensions, accuracy requirements, and constraints of the actual problem, and the algorithm can be improved according to actual needs.

[0058] In summary, this invention provides a method for optimizing a segmented tire-changing process based on a genetic algorithm. This method first obtains production data from the tire position map, extracting the location and height of each tire frame. Then, it delineates the tire frame's footprint based on the current production site conditions, and determines the displacement domain of the ship segment relative to the tire frame based on the tire frame's boundary range, thus defining the segment's movable range on the site. Next, it optimizes the segment's movement relative to the tire frame using a genetic algorithm, aiming to minimize the sum of height adjustments required for each tire frame during segment replacement, thereby obtaining the optimal movement amount. In the genetic algorithm's solution process, an initial population is first randomly generated, containing multiple chromosomes. Genes arranged sequentially on the chromosomes represent the movement amount of each segment. Genes at the same location on different chromosomes represent different movement amounts. Individuals with high fitness in the initial population are selected as parent chromosomes. The crossover and mutation process is then repeated until a predetermined number of iterations is reached or a stopping condition is met. Finally, individuals with higher fitness values ​​are retained as the optimal movement amount.

[0059] This invention uses a genetic algorithm for mathematical modeling, achieving better accuracy and reducing the likelihood of getting trapped in local optima compared to greedy algorithms and gradient descent. Compared to exhaustive search, its computational speed increases exponentially, significantly improving problem-solving efficiency. Furthermore, this method offers several improved solutions, providing greater flexibility in analyzing specific problems.

[0060] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A method for optimizing a segmented tire-changing process based on a genetic algorithm, characterized in that, Includes the following steps: S1: Obtain production data from the tire position diagram and convert it into a data table. Extract the location and height of each tire frame, and arrange multiple tire frames in the left-right direction. S2: Describe the footprint of the jig based on the current production site conditions, and determine the displacement domain of the ship section relative to the jig based on the boundary range of the jig. S3: Determine the number of segments. When changing to different segments, the height of each tire frame needs to be adjusted to fit the bottom line of the segment. Based on the genetic algorithm, optimize the movement of the segment relative to the tire frame. The goal is to minimize the sum of the height adjustments required for each tire frame when changing segments, and obtain the optimal movement amount.

2. The segmented tire-changing process optimization method based on genetic algorithm according to claim 1, characterized in that: The frame includes angle steel and flexible frame.

3. The segmented tire-changing process optimization method based on genetic algorithm according to claim 1, characterized in that, Step S3 includes: Define the amount of displacement of the segment relative to the frame within the displacement domain. The amount of displacement includes the horizontal displacement x and the vertical displacement y. An initial population is randomly generated, consisting of multiple chromosomes. The genes in the first chromosome are arranged in sequence as X1, X2, X3, ..., Xm, Y1, Y2, Y3, ..., Ym. Here, X1 represents the horizontal displacement x of the first segment relative to the frame, Xm represents the horizontal displacement x of the m-th segment relative to the frame, Y1 represents the vertical displacement y of the first segment relative to the frame, and Ym represents the vertical displacement y of the m-th segment relative to the frame. Based on the gene values ​​in the first chromosome, a preset step size is added to each gene value to generate the second chromosome, and so on, to generate multiple chromosomes.

4. The segmented tire-changing process optimization method based on genetic algorithm according to claim 3, characterized in that, Step S3 also includes: Define the objective function as follows: In the formula: m is the total number of segments involved in tire changing, and n is the number of tire frames; L mn L represents the ground clearance of the nth tire frame when the mth segment is fitted with the upper tire and is in contact with the tire frame. (m-1)n This represents the ground clearance of the nth tire frame when the upper tire is placed in the (m-1)th segment and is in contact with the tire frame; The fitness function is F = a / f; The fitness values ​​of chromosomes in the initial population are calculated, and individuals with high fitness in the initial population are selected as parent chromosomes. Single-point gene crossover is performed on the selected parent chromosomes to generate offspring chromosomes. Mutation is then performed, and the values ​​of single-point genes in the offspring chromosomes are randomly adjusted to enhance diversity. The selection, crossover, and mutation process is repeated until the predetermined number of iterations is reached or the stopping condition is met. Finally, individuals with high fitness values ​​are retained as the optimal movement value.

5. The segmented tire-changing process optimization method based on genetic algorithm according to claim 4, characterized in that: The amount of movement also includes the angle values ​​of the segmented rotation relative to the frame within the allowable range of the site, with the angle values ​​being multiples of 90°, including 90°, 180°, and 270°; correspondingly, a gene is added to the chromosome to represent the angle value.

6. The segmented tire-changing process optimization method based on genetic algorithm according to claim 4, characterized in that: The numerical values ​​represented by each gene in the chromosome are converted into binary codes, and each bit of the binary code is used as a child node, thus using multiple child nodes to replace a single gene.

Images