Artificial intelligence-based mobius ring production method and mobius ring stent

Abstract

The application relates to the technical field of artificial intelligence, and provides a Mobius ring production method based on artificial intelligence and a Mobius ring support. A parameterized model of the Mobius ring support is established, the parameterized model is configured with multiple key process parameters, and a multi-objective optimization AI algorithm is used to calculate an optimal blanking scheme and a welding point distribution scheme according to product production requirements. The first optimization target of the optimization target of the multi-objective optimization AI algorithm is to minimize material waste, the second optimization target is to meet the radian consistency of the inner ring and the outer ring after forming and the 180-degree twisting requirement, and the third optimization target is to uniformly distribute the welding positions, so as to guarantee the stability and strength when the support displays on both sides. The material belt is blanked according to the optimal blanking scheme, and after the Mobius ring support is assembled, the hardening welding support bars are welded according to the welding point distribution scheme, so that the labor experience cost is reduced, and the production efficiency of the Mobius ring support is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical fields of artificial intelligence, Mobius ring support and the like, and in particular to a Mobius ring production method based on artificial intelligence and a Mobius ring support. BACKGROUND

[0002] In the production of Mobius ring lamps and display screens, it is necessary to first design a Mobius ring support, and then assemble a display module or a light-emitting module to the Mobius ring support to form a Mobius ring lamp or a display screen. In the production of the Mobius ring support, a plurality of strip materials need to be spliced to obtain a continuous long strip material with a first end connected to a second end. In order to smoothly connect the first end to the second end of the long strip material and form a twist, the length of each inner arc and outer arc needs to be accurately calculated when the strip material is cut. The Mobius ring support formed includes an inner ring and an outer ring. In order to realize double-sided display, a plurality of hardening welding positions are selected at the edge positions of the inner ring and the outer ring to weld support strips, so as to realize hardening treatment of the entire Mobius ring support and support double-sided display. In practice, the cutting of the strip material and the selection of the hardening welding positions mainly rely on human experience. When producing Mobius ring supports of different models and lengths, a lot of experience cost needs to be paid, which is not conducive to the efficient production of the Mobius ring support.

[0003] In summary, the existing production technology of the Mobius ring support has the technical problems of high labor experience cost and low production efficiency. SUMMARY

[0004] In view of the above-mentioned problems of the prior art, the present application provides a Mobius ring production method based on artificial intelligence and a Mobius ring support, so as to reduce the labor experience cost and improve the production efficiency of the Mobius ring support.

[0005] In a first aspect, the present application provides a Mobius ring production method based on artificial intelligence, comprising:

[0006] A parameterized model of the Mobius ring support is established, and the parameterized model is configured with a plurality of key process parameters, including a single-section strip length, a twist angle, an inner ring radius, an outer ring radius, and a maximum number of hardening welding support strips that can be welded;

[0007] According to the production requirements of the finished product, a multi-objective optimization AI algorithm is used to calculate an optimal cutting scheme and a welding point distribution scheme for the plurality of key process parameters; the optimization objectives of the multi-objective optimization AI algorithm include a first optimization objective, a second optimization objective, and a third optimization objective, the first optimization objective is to minimize material waste, the second optimization objective is to meet the requirements of the arc consistency of the inner ring and the outer ring after forming and the 180° twist, and the third optimization objective is to evenly distribute the welding positions to ensure the stability and strength when the support is double-sided displayed;

[0008] According to the optimal blanking scheme, the material belt is blanked, and after the material belt obtained by blanking is assembled to form a closed Mobius ring support, the hardening welding support bar is welded according to the welding point distribution scheme.

[0009] In a second aspect, the present application provides a Mobius ring support produced by the above-mentioned Mobius ring production method based on artificial intelligence.

[0010] Compared with the prior art, the present application has the following advantages:

[0011] The present application provides a Mobius ring production method based on artificial intelligence and a Mobius ring support. By establishing a parameterized model of the Mobius ring support, the parameterized model is configured with a plurality of key process parameters, including single-section material belt length, twist angle, inner ring radius, outer ring radius, and maximum weldable number of hardening welding support bars. According to the production requirements of the finished product, a multi-objective optimization AI algorithm is used to calculate an optimal blanking scheme and a welding point distribution scheme. The optimization objectives of the multi-objective optimization AI algorithm include a first optimization objective, a second optimization objective, and a third optimization objective. The first optimization objective is to minimize material waste, the second optimization objective is to meet the arc consistency and 180° twist requirements of the formed inner ring and outer ring, and the third optimization objective is to ensure the stability and strength of the support when displaying on both sides. According to the optimal blanking scheme, the material belt is blanked, and after the material belt obtained by blanking is assembled to form a closed Mobius ring support, the hardening welding support bar is welded according to the welding point distribution scheme, thereby reducing the cost of human experience and improving the production efficiency of the Mobius ring support. BRIEF DESCRIPTION OF DRAWINGS

[0012] The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application. In the drawings, which are not necessarily drawn to scale, like reference numerals describe similar components throughout the several views. The specific embodiments of the present application will now be described with reference to the drawings:

[0013] Figure 1 is a flowchart of the Mobius ring production method based on artificial intelligence according to an embodiment of the present application;

[0014] Figure 2 is another flowchart of the Mobius ring production method based on artificial intelligence according to an embodiment of the present application;

[0015] Figure 3 is a structural schematic diagram of a Mobius ring support of an embodiment of the present application. DETAILED DESCRIPTION

[0016] In order for those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings of the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present application.

[0017] Embodiment one

[0018] Referring to Figures 1-3 , the present embodiment provides a Mobius ring production method based on artificial intelligence, comprising the following steps:

[0019] S101, a parameterized model of a Mobius ring support is established, the parameterized model is configured with a plurality of key process parameters, the plurality of key process parameters include a single section of a material belt length, a torsion angle, an inner ring radius, an outer ring radius, and a maximum weldable number of hardening welding support bars;

[0020] S102, according to the production requirements of finished products, a multi-objective optimization AI algorithm is used to calculate an optimal unloading scheme and a welding point distribution scheme for the plurality of key process parameters; the optimization objectives of the multi-objective optimization AI algorithm include a first optimization objective, a second optimization objective, and a third optimization objective, the first optimization objective is to minimize material waste, the second optimization objective is to meet the arc consistency of the inner ring and the outer ring after forming and the 180° torsion requirement, and the third optimization objective is to evenly distribute the welding positions to ensure the stability and strength when the support displays on both sides; wherein the first optimization objective, the second optimization objective, and the third optimization objective can be represented by setting corresponding objective functions. After the definition of the first optimization objective, the second optimization objective, and the third optimization objective in the present embodiment is clear, the setting method of the specific objective function can be known by those skilled in the art, and the present embodiment will not be described here.

[0021] S103, according to the optimal unloading scheme, the material belt is unloaded, and after the material belt obtained by unloading is assembled to form a closed Mobius ring support, the hardening welding support bar is welded according to the welding point distribution scheme.

[0022] It should be noted that the Mobius ring support has unique application value in the field of lamps, display screens and the like, and usually requires processes such as splicing, twisting and welding of the material belt to form a ring structure that can display on both sides, is beautiful and has reasonable stress. However, traditional production often relies on manual experience to determine the blanking scheme and welding position, and when faced with a variety of models, sizes and thicknesses of materials, it is easy to cause problems such as material waste, twisting out of place or insufficient strength. In the embodiment, through the parameterized model and the multi-objective optimization AI algorithm, a relatively optimal or approximately optimal production scheme can be automatically and quickly found within a given range of process parameters, maximizing production efficiency and product quality. In step S101, the parameterized model is configured with a plurality of key process parameters, including single-section material belt length, twisting angle, inner ring radius, outer ring radius, and maximum weldable number of hardening welding support bars. Among them, the formation of the Mobius ring needs to consider both the geometric shape (such as the twisting angle, the inner and outer ring radius, etc.) and the support bar related parameters (such as the upper limit of the number). If the systematic management of these process parameters is missing, subsequent production will face the risk of a large number of manual trial and error and difficulty in maintaining consistency. Concentrating all parameters in an adjustable parameterized model can greatly improve the efficiency of design, modification and subsequent optimization. Different specifications of Mobius rings (different in length, twisting angle, etc.) only need to adjust the corresponding parameters in the parameterized model to quickly generate a digital model, without the need to start modeling from scratch, not only meeting batch efficient production and reducing labor experience costs, but also supporting flexible customization. In step S102, according to the production requirements of the finished product, a multi-objective optimization AI algorithm is used to calculate an optimal blanking scheme and a welding point distribution scheme for the plurality of key process parameters. The optimization objectives of the multi-objective optimization AI algorithm include a first optimization objective, a second optimization objective and a third optimization objective, the first optimization objective is to minimize material waste, the second optimization objective is to meet the consistency of the inner and outer ring arcs and the 180° twisting requirement after forming, and the third optimization objective is to evenly distribute the welding positions to ensure the stability and strength of the support when displaying on both sides. Among them, when producing the Mobius ring support, attention needs to be paid to: material utilization rate (reduce waste); forming accuracy (meet 180° twisting and ensure the consistency of inner and outer ring arcs); welding support stability (evenly distributed, do not deform or loosen when supporting both sides). These objectives often restrict each other: for example, in order to save materials, it may not be easy to form or the support bar distribution may be poor; in order to meet the high strength support, it may increase the amount of material splicing and cause material waste. The multi-objective optimization AI algorithm (such as genetic algorithm, particle swarm algorithm, etc.) can perform global search and iteration in a complex multi-dimensional space, and can find a Pareto optimal solution that takes into account all parties better than manual experience or single-objective optimization. Traditional methods usually rely on the personal experience of technicians to adjust the blanking scheme or repeatedly explore the welding position; thus, for new models or large-scale production, the efficiency and consistency are difficult to guarantee.In this embodiment, by inputting the product production demand (such as the product circumference), the multi-objective optimization AI algorithm can automatically calculate the optimal cutting scheme and the optimal welding point distribution, significantly shortening the trial and error process and improving product consistency. In addition, different enterprises or products have different requirements for the twisting accuracy, appearance, and support strength of the Mobius ring. Multi-objective optimization can flexibly set weights or limit conditions and automatically output the optimal scheme in different scenarios without the need to rewrite the logic. In step S103, the material strip is cut according to the optimal cutting scheme, and after the material strip obtained by cutting is assembled to form a closed Mobius ring support, the hardening welding support bar is welded according to the welding point distribution scheme. In step S102, the multi-objective optimization AI algorithm has given a cutting scheme and welding position that meet the multi-objective balance. In step S103, the cutting is performed according to the optimized specified material strip quantity, length, cross-sectional cutting, etc. The cut material strip is continuously spliced and twisted to form a Mobius ring in a closed loop. Then, the hardening support bar is installed and welded at the AI-suggested welding point coordinates (i.e., the welding point distribution scheme) to ensure stable double-sided display. The Mobius ring often needs to emit light / display on both sides in the field of lamps or display screens, so the distribution of the welding support bar is particularly important. If the distribution is too sparse or uneven, the double-sided display panel may be loose or deformed; if the distribution is too dense or the position is not appropriate, material is wasted and the appearance is damaged. In step S103, according to the optimal welding layout, the strength of the finished product can be ensured, and a more optimal appearance and double-sided function can be achieved.

[0023] In further embodiments, the product production demand includes target size requirements, tolerances and surface requirements, installation methods, load weights or optical layout requirements, currently available material specifications, thickness and quantity requirements. It should be noted that the multi-objective optimization AI algorithm needs to find the optimal scheme under the premise of meeting the product production demand. Among them, the target size requirements (such as the total circumference, the outer / inner ring diameter or radius, and the 180° twisting angle) directly determine the geometric characteristics that the material strip should exhibit after forming. The AI algorithm needs to take these parameters as the basis when calculating the segmented cutting or twisting; at the same time, the distribution of the welding position also needs to match the geometric arc of the final ring shape to ensure that the finished product splicing is accurate. In addition, the tolerance and surface requirements can be used in multi-objective optimization to form accuracy and appearance. The tolerance requirement determines the allowable range of cutting and welding position accuracy; the surface requirement affects whether more welding positions of the hardening welding support bar are acceptable, thereby affecting the optimization of the number and position of the welding bar. In addition, the installation method, load weight or optical layout requirement determines the actual function (such as a display module, a lamp module, etc.) that the Mobius ring finally bears, thereby affecting the distribution of the welding position, the number of support bars, and the selection of the material strip thickness. In addition, the currently available material specifications, thickness and quantity requirements consider arranging the optimal cutting layout based on the availability of known materials to minimize waste or short materials.

[0024] It should be noted that after obtaining the target size requirements, tolerances and surface requirements of the finished product, installation methods, load weight or optical layout requirements, current available material specifications, thickness and quantity requirements, and other specific finished product production requirements and the plurality of key process parameters, the specific finished product production requirements can be used as the optimization limiting variables of the multi-objective optimization AI algorithm, and the plurality of key process parameters are input into the trained multi-objective optimization AI algorithm model, and the multi-objective optimization AI algorithm model calculates the optimal blanking scheme and welding point distribution scheme according to the optimization limiting variables and the plurality of key process parameters, including single-section material belt length, twisting angle, inner ring radius, outer ring radius, material belt cross-sectional size, material belt thickness, material belt material, hardened welding support bar size, hardened welding support bar material, and hardened welding support bar maximum weldable quantity. The multi-objective optimization AI algorithm can use genetic algorithm, particle swarm algorithm or Bayesian optimization algorithm.

[0025] In further embodiments, the multi-objective optimization AI algorithm uses a genetic algorithm model. When using the genetic algorithm model to calculate the optimal blanking scheme and welding point distribution scheme, the following steps are included: determining the decision variables required by the genetic algorithm model, which include the specific lengths and splicing sequences of the plurality of sections of the material belt under the premise of satisfying the total circumference of the ring, the distribution coordinates of the hardened welding support bars, and the number of hardened welding support bars that does not exceed the maximum weldable quantity of the hardened welding support bars; the genetic algorithm model randomly generates a certain number of initial schemes within the constraint range according to the decision variables, the optimization limiting variables of the finished product production requirements, and the plurality of key process parameters; each initial scheme includes material belt segmentation and length combination, hardened welding support bar distribution coordinates and number; the genetic algorithm model evaluates and processes each initial scheme through the first optimization target, the second optimization target, and the third optimization target to obtain evaluation results saved in the fitness information of each initial scheme; through selection processing, the initial schemes with good fitness (i.e., less material waste, high forming precision, and reasonable welding point distribution) are retained, and through crossover processing, two excellent initial schemes exchange part of the decision variables (such as certain section of material belt length setting and welding point layout) with each other to generate new offspring schemes; through mutation processing, part of the initial schemes are randomly adjusted in a small range to increase the algorithm exploration range, such as slightly changing the length of a certain section of material belt, adding / removing welding points in certain ring sections, etc.; the evaluation processing, selection processing, crossover processing, and mutation processing are repeatedly performed to iteratively evolve, eliminate inferior initial schemes, and retain and improve superior schemes; if the initial schemes as a whole have converged or reached the preset excellent level after a certain generation or exceed the maximum number of generations, the genetic algorithm model stops iterative evolution and outputs several optimal blanking schemes and welding point distribution schemes.

[0026] It should be noted that in the production of the Mobius ring support, the length and splicing order of the several sections of the material belt, the number and distribution position of the support bars, all belong to a combination optimization problem. Genetic algorithm (GA) is good at global search by iteratively evolving the population individuals, can effectively avoid falling into local optimum, and can handle continuous or discrete parameters. In this embodiment, the first optimization target pursues less material waste, the second optimization target pursues high forming precision, and the third optimization target pursues reasonable distribution of welding points. Genetic algorithm can handle multiple targets in the same iteration, output one or more sets of better solutions, and can flexibly configure various screening mechanisms (such as Pareto sorting). The selection, crossover, mutation and other steps of genetic algorithm are relatively simple in principle and easy to implement at the software level; it can also be easily connected with other algorithm components (such as mechanical simulation or CAD interface), which is suitable for industrial application scenarios. In addition, in this embodiment, the decision variables include the specific length and splicing order of the several sections of the material belt under the premise of meeting the total circumference of the ring, the distribution coordinates of the hardened welded support bars, and the number of hardened welded support bars not exceeding the maximum weldable number of the hardened welded support bars. In the Mobius ring, the total circumference of the ring is a basic constraint, and the combination of the length of each section of the material belt and the splicing order directly determines the cutting method and material utilization rate; the distribution coordinates and number of hardened welded support bars affect the strength of the overall structure, the installation position of the double-sided display, and the welding cost. By setting these factors as decision variables, genetic algorithm can freely explore different segmentation and welding point layout schemes, and then find better solutions that take into account material waste, forming precision, and reasonable stress in iterative evolution. In addition, the genetic algorithm model randomly generates a certain number of initial schemes within the constraint range according to the decision variables, the optimization limit variables of the product production requirements, and the multiple key process parameters. Random initialization is a common practice for genetic algorithm, which can distribute initial solutions in the solution space to avoid local optimization from a single point. At the same time, it is necessary to ensure that the initial scheme does not violate the basic constraints (such as total circumference, tolerance requirements, available material specifications, and maximum weldable number) to ensure the basic feasibility of each individual. In addition, the genetic algorithm model evaluates each initial scheme through the first optimization target, the second optimization target, and the third optimization target to obtain evaluation results saved in the fitness information of each initial scheme. Genetic algorithm needs to use fitness to measure the performance of each scheme in multiple targets. Only in this way, the algorithm can retain excellent individuals with less material waste, more accurate forming, and more reasonable support distribution in the selection process, and gradually approach better solutions. The evaluation results are written into the fitness information, which can be selected, crossed, and mutated according to the multi-objective strategy (such as Pareto sorting or weight addition) to eliminate the inferior.In addition, by selection processing, the initial scheme with good fitness is retained; by crossover processing, two good initial schemes exchange part of the decision variables with each other; and by mutation processing, part of the initial scheme is adjusted in a small range at random. One of the core ideas of the genetic algorithm is selection. In multi-objective optimization, the solutions with good performance can be retained in combination with Pareto sorting or fitness ranking; crossover reflects that part of the decision variables (such as the length of the material belt and the coordinates of the welding points) of two good solutions are spliced together, so as to generate a new scheme with more potential; and mutation reflects that a small amount of random disturbance is allowed, such as adding or removing welding points in a certain ring segment, changing the length of the material belt, and the like, so as to avoid premature convergence of the algorithm and maintain the diversity of the population. The repeated execution of these “selection-crossover-mutation” steps can simulate the process of biological evolution, continuously improve the scheme, and improve the overall fitness of the population. In addition, if the initial scheme as a whole converges or reaches a preset good level after a certain generation or exceeds the maximum number of generations, the genetic algorithm model stops iteration and evolution and outputs several better cutting schemes and welding point distribution schemes. The genetic algorithm is essentially an iterative search, and the population gradually approaches the high fitness area after multiple rounds of “evaluation-selection-crossover-mutation”; when the algorithm meets the convergence condition (such as a small change in the objective function) or reaches the required precision in engineering, it is not necessary to waste computing resources any more, and finally multiple groups of better solutions are output, so that the producer can select and execute the optimal solution in combination with the actual production focus (material cost, time, strength, etc.).

[0027] In some preferred embodiments, when the better cutting scheme and the welding point distribution scheme are calculated, a mechanical simulation module is used in combination with the multi-objective optimization AI algorithm to automatically evaluate the influence of the length of the material belt and the distribution of the welding position on the strength of the finished product. It should be noted that after the Mobius ring is turned and formed, the inner ring and outer ring structures of the support will generate a specific stress distribution; if the welding position and the material belt splicing method are not properly selected, stress concentration and insufficient strength may occur in some ring segments. Therefore, in addition to achieving material saving or appearance precision in cutting layout and the number of welding points, attention should also be paid to the final mechanical performance to ensure that the finished product can bear the weight of the lamp or display screen and meet the safety requirements in the use environment. When the multi-objective optimization AI algorithm is used, the algorithm focuses on material utilization, forming precision (twist angle and ring circumference error), and welding distribution uniformity. Through the mechanical simulation module, the strength of the finished product as a whole or in part can also be included in the optimization evaluation: when the multi-objective optimization AI algorithm generates a group of material belt segments and welding coordinate schemes, the mechanical simulation module can automatically test whether there is a high stress area, excessive deformation, or potential failure risk under the stress state.

[0028] In some preferred embodiments, when the parametric model of the Möbius ring stent is established, the parametric modeling of the Möbius ring stent is performed using SolidWorks software to obtain the parametric model of the Möbius ring stent. It should be noted that SolidWorks, as a common three-dimensional design software, can enable designers to quickly create and modify the geometric structure of the Möbius ring stent. Under the parametric modeling function, only by inputting or adjusting the key parameters (such as the inner and outer ring radii, the twist angle, the width of the material belt, etc.) can the entire three-dimensional model be automatically updated and the structural changes be intuitively presented, thereby greatly reducing the workload of manual modeling and reducing the design error rate.

[0029] In some preferred embodiments, after the hardening welding support bars are welded according to the welding point distribution scheme, the number of hardening welding support bars is increased or decreased to obtain an adjusted Möbius ring stent of the hardening welding support bars; the structural stability of the adjusted Möbius ring stent is compared with that of the Möbius ring stent before adjustment; if the structural stability of the adjusted Möbius ring stent is better than that of the Möbius ring stent before adjustment, the welding point distribution scheme is updated, and if the structural stability of the adjusted Möbius ring stent is worse than that of the Möbius ring stent before adjustment, the welding point distribution scheme is maintained. It should be noted that although the multi-objective optimization AI algorithm can theoretically output a better welding point distribution scheme, some details that are not fully considered may occur in the production site (such as material batch difference, welding process deviation, appearance or specific installation requirement change, etc.). In this embodiment, by moderately increasing or decreasing the number of hardening welding support bars after welding, the local area is quickly reinforced or simplified, so that the stability effect of the stent under actual working conditions can be tested and observed in the shortest time with the least modification. When the stability result of the adjusted stent is better (or worse) than that of the stent before adjustment, this test result can be fed back to the welding point distribution scheme in time. If it is better, the scheme is updated; if the effect is poor, the original scheme is retained, thereby providing a closed-loop mechanism for trial and error and improvement in the production line, which can continuously improve the welding point distribution strategy and accumulate experience in subsequent batch production or repeated manufacturing of the same model, thereby continuously improving the overall quality and reliability of the Möbius ring stent.

[0030] In some preferred embodiments, after the Mobius ring support with better structural stability is obtained, a reinforcing cover plate is welded at the welding position of the hardened welding support bar to cover the hardened welding support bar, obtaining a Mobius ring support finished product. It should be noted that after multiple tests or fine-tuning, a Mobius ring support with better structural stability has been obtained. At this time, by welding a reinforcing cover plate at the welding position of the hardened welding support bar, the support bar is more firmly covered with the main frame in a closed manner, further improving the strength or protection performance of the local connection, playing a role in aesthetics and decoration, avoiding external collisions or environmental corrosion during use of the support, and making the overall modeling more complete.

[0031] In some preferred embodiments, when the multi-objective optimization AI algorithm is used to calculate the optimal blanking scheme for the plurality of key process parameters, the single-section strip length includes a single-section strip length of an inner ring of the Mobius ring support and a single-section strip length of an outer ring of the Mobius ring support, and the single-section strip length of the inner ring of the Mobius ring support is smaller than the single-section strip length of the outer ring of the Mobius ring support. It should be noted that the inner ring of the Mobius ring support usually has a smaller radius of curvature and is more compact than the outer ring. If the inner ring strip length is consistent with the outer ring, a large arc error or local stress concentration may be formed during assembly and twisting, affecting the forming accuracy and overall strength. In this embodiment, the inner ring single-section strip is set to be shorter than the outer ring single-section strip, which helps to better fit the inner ring curved surface with a smaller radius, reduces the problem of pulling or difficulty in smooth splicing during twisting deformation, and thus improves the curve accuracy and stress uniformity of the final product.

[0032] In some preferred embodiments, in the optimal blanking scheme, the total number of single-section strips of the inner ring of the Mobius ring support and the total number of single-section strips of the outer ring of the Mobius ring support are the same. It should be noted that when the number of single-section strips of the inner ring and the outer ring is consistent, each inner ring strip can be one-to-one or synchronized with the corresponding outer ring strip during splicing and turning, avoiding the problem of misalignment of the inner ring and the outer ring in the assembly sequence.

[0033] In some preferred embodiments, before the hardened welding support bar is welded according to the welding point distribution scheme, whether the Mobius ring support is successfully closed at the beginning and the end is tested. It should be noted that when the Mobius ring support is formed into a 180° twist and attempts to butt joint at the beginning and the end, if there are problems such as length error, turning error, misalignment of the inner and outer rings, etc., the support may not be able to be smoothly closed. If the hardened welding is performed before the beginning and the end are completely and accurately butt jointed, once it is found that the closure is deviated, the welding point often needs to be cut off and re-turned or reworked. This not only wastes time, but also easily damages the material. Therefore, by testing whether the beginning and the end are successfully closed before welding the support bar, the risk of repeated disassembly can be reduced by avoiding the urgent welding on a structure with assembly errors or size errors.

[0034] Embodiment two

[0035] Referring to Figure 1 A Mobius ring support is produced by the artificial intelligence-based Mobius ring production method described in any of the above embodiments. A parameterized model of the Mobius ring support is established, which is configured with a plurality of key process parameters, including single-section material belt length, twisting angle, inner ring radius, outer ring radius, and maximum weldable quantity of hardening welding support bars. According to product production requirements, a multi-objective optimization AI algorithm is used to calculate an optimal blanking scheme and a welding point distribution scheme for the plurality of key process parameters. The optimization objectives of the multi-objective optimization AI algorithm include a first optimization objective, a second optimization objective, and a third optimization objective. The first optimization objective is to minimize material waste, the second optimization objective is to meet the requirements of the arc consistency of the inner ring and the outer ring after forming and the 180° twisting requirement, and the third optimization objective is to evenly distribute the welding positions to ensure the stability and strength of the support when displaying on both sides. The material belt is blanked according to the optimal blanking scheme, and after the material belt obtained by blanking is assembled to form a Mobius ring support with a closed head and tail, the hardening welding support bars are welded according to the welding point distribution scheme, thereby reducing the labor experience cost and improving the production efficiency of the Mobius ring support.

[0036] It should be noted that the above embodiments are only preferred specific embodiments of the present application, and the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which should be covered within the protection scope of the present application. The protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for producing a Mobius ring based on artificial intelligence, characterized by, include: A parametric model of the Möbius strip support is established. The parametric model is configured with a variety of key process parameters, including the length of a single strip, the torsion angle, the inner ring radius, the outer ring radius, and the maximum number of hardened weldable support strips. Based on the finished product production requirements, a multi-objective optimization AI algorithm is used to calculate the optimal material cutting scheme and weld point distribution scheme for the various key process parameters. The optimization objectives of the multi-objective optimization AI algorithm include a first optimization objective, a second optimization objective, and a third optimization objective. The first optimization objective is to minimize material waste. The second optimization objective is to meet the curvature consistency of the inner and outer rings and the 180° torsion requirement after forming. The third optimization objective is to ensure uniform distribution of welding positions to guarantee the stability and strength of the bracket when it is displayed on both sides. The material strip is cut according to the preferred cutting scheme, and after the material strip is assembled into a closed Möbius ring bracket, the hardened welding support strip is welded according to the welding point distribution scheme. The finished product production requirements include target size requirements, tolerance and surface requirements, installation method, load weight or optical layout requirements, and currently available material specifications, thickness and quantity requirements. Among these, the target size requirements determine the geometric characteristics the strip should exhibit after forming; the distribution of welding positions matches the final ring's geometric curvature to ensure accurate splicing of the finished product; tolerance requirements determine the allowable range of accuracy for strip cutting and welding positions; surface requirements affect the optimization of the number and position of welding strips; installation method, load weight or optical layout requirements determine the actual function of the Möbius strip, affecting the distribution of welding positions, the number of support strips, and the selection of strip thickness; currently available material specifications, thickness and quantity requirements consider optimizing the cutting and layout based on known material availability to avoid waste or material shortages. After obtaining the target size requirements, tolerance and surface requirements, installation method, load weight or optical layout requirements of the finished product, the currently available material specifications, thickness and quantity requirements, and the specific finished product production needs and the various key process parameters, the specific finished product production needs are used as optimization constraint variables for the multi-objective optimization AI algorithm. These requirements, along with the various key process parameters, are input into the trained multi-objective optimization AI algorithm model. The multi-objective optimization AI algorithm model calculates the optimal cutting scheme and weld point distribution scheme based on the optimization constraint variables and the various key process parameters, including the single-segment strip length, torsion angle, inner ring radius, outer ring radius, strip cross-sectional size, strip thickness, strip material, hardened welding support strip size, hardened welding support strip material, and the maximum number of hardened welding support strips that can be welded. The multi-objective optimization AI algorithm employs a genetic algorithm model. When using the genetic algorithm model to calculate the optimal material cutting scheme and weld point distribution scheme, the following steps are taken: determining the decision variables required by the genetic algorithm model. These decision variables include the specific lengths and splicing order of several material strip segments while satisfying the total circumference of the ring, the distribution coordinates of the hardened welding support strips, and the number of hardened welding support strips not exceeding the maximum weldable number of the hardened welding support strips. The genetic algorithm model, based on the decision variables, the finished product production requirements as optimization constraints, and the various key process parameters, randomly generates a certain number of initial schemes within the constraints. Each initial scheme includes material strip segmentation and length combinations, and the distribution coordinates and number of hardened welding support strips. The genetic algorithm model evaluates each initial scheme based on the first optimization objective, the... The second and third optimization objectives are evaluated to obtain evaluation results which are saved in the fitness information of each initial scheme. Through selection, initial schemes with good fitness are retained. Through crossover, two excellent initial schemes exchange some decision variables to generate new descendant schemes. Through mutation, some initial schemes are randomly adjusted within a small range to increase the algorithm's exploration range. The evaluation, selection, crossover, and mutation processes are repeated iteratively to evolve, eliminating poor initial schemes generation by generation and retaining and improving better schemes. If, after a certain generation or exceeding the maximum number of generations, the initial schemes have converged or reached a preset level of excellence, the genetic algorithm model stops iterative evolution and outputs several better material cutting schemes and solder joint distribution schemes.

2. The method for producing a Möbius strip based on artificial intelligence as described in claim 1, characterized in that, Also includes: When calculating the optimal material cutting scheme and the weld point distribution scheme, the mechanical simulation module is used in conjunction with the multi-objective optimization AI algorithm to automatically evaluate the impact of different strip lengths on the strength of the finished product by splicing the strips and the distribution of welding positions.

3. The method for producing a Möbius strip based on artificial intelligence as described in claim 1, characterized in that, The process of establishing a parametric model of the Möbius strip support includes: using SolidWorks software to perform parametric modeling of the Möbius strip support to obtain the parametric model of the Möbius strip support.

4. The method for producing a Möbius strip based on artificial intelligence as described in claim 1, characterized in that, The multi-objective optimization AI algorithm employs genetic algorithms, particle swarm optimization algorithms, or Bayesian optimization algorithms.

5. The method for producing a Möbius strip based on artificial intelligence as described in claim 1, characterized in that, After welding the hardened support strips according to the weld point distribution scheme, the process includes: increasing or decreasing the number of hardened support strips to obtain a Möbius strip bracket with adjusted hardened support strips; comparing the structural stability of the adjusted Möbius strip bracket with that of the original Möbius strip bracket; if the structural stability of the adjusted Möbius strip bracket is better than that of the original Möbius strip bracket, then the weld point distribution scheme is updated; if the structural stability of the adjusted Möbius strip bracket is worse than that of the original Möbius strip bracket, then the weld point distribution scheme is maintained.

6. The method for producing a Möbius strip based on artificial intelligence as described in claim 5, characterized in that, After adjusting to obtain a Möbius strip support with better structural stability, a reinforcing sealing plate is welded to the welding position of the hardened welded support strip to cover the hardened welded support strip, thus obtaining the finished Möbius strip support.

7. The method for producing a Möbius strip based on artificial intelligence as described in claim 1, characterized in that, When using a multi-objective optimization AI algorithm to calculate the optimal material cutting scheme for the various key process parameters, the single-segment material strip length includes the single-segment material strip length of the inner ring of the Möbius strip support and the single-segment material strip length of the outer ring of the Möbius strip support. The single-segment material strip length of the inner ring of the Möbius strip support is less than the single-segment material strip length of the outer ring of the Möbius strip support.

8. The method for producing a Möbius strip based on artificial intelligence as described in claim 7, characterized in that, In the preferred material feeding scheme, the total number of single-segment material strips in the inner ring of the Möbius strip support is the same as the total number of single-segment material strips in the outer ring of the Möbius strip support.

9. The method for producing a Möbius strip based on artificial intelligence as described in any one of claims 1-8, characterized in that, Before welding the hardened support strip according to the weld point distribution scheme, test whether the Möbius strip bracket can be successfully closed end to end.

10. A Möbius strip support, characterized in that, The Möbius strip support is manufactured using the artificial intelligence-based Möbius strip production method as described in any one of claims 1-9.

Images