Laser cutting track interpolation method, device and equipment and storage medium

By introducing a non-dominated sorting evolution mechanism for infeasible solution updates and B-spline interpolation into laser cutting trajectory interpolation, the cutting trajectory is optimized, solving the accuracy and efficiency problems of complex geometries and achieving high-precision and high-efficiency laser cutting results.

CN118940622APending Publication Date: 2024-11-12JIANGSU UNIV OF SCI & TECH
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202410978594.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-22
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

Existing laser cutting trajectory interpolation methods suffer from insufficient accuracy, low efficiency, and lack of interpretability when dealing with complex three-dimensional surfaces and complex geometries. In particular, in NURBS curve interpolation, the neglect of higher-order terms leads to truncation errors and speed fluctuations, affecting processing quality and equipment stability.

Method used

A non-dominated sorting evolution mechanism for updating infeasible solutions is adopted. By constructing a multi-objective optimization model and utilizing the multi-objective function information of infeasible solutions, B-spline interpolation is optimized. Combined with adaptive pre-interpolation and Simpson numerical integration method, velocity and acceleration are dynamically adjusted to optimize the cutting trajectory and reduce uncertainty and error in the interpolation process.

Benefits of technology

It improves the precision and efficiency of laser cutting, reduces the peak value of centripetal acceleration, reduces uncertainties in the interpolation process, meets the requirements of high precision and edge smoothness, and improves processing quality and equipment stability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118940622A_ABST
    Figure CN118940622A_ABST
Patent Text Reader

Abstract

The invention provides a laser cutting track interpolation method, device and equipment and a storage medium, and the method comprises the steps: importing discrete laser head data points in a laser cutting track, and carrying out B spline curve fitting to obtain a first laser cutting target track; setting a speed constraint condition to carry out interpolation on the first laser cutting target track to obtain a second laser cutting target track, and optimizing the second laser cutting target track by adopting a non-dominated sorting evolution mechanism updated by an infeasible solution based on a multi-target optimization model to obtain a third laser cutting target track; and the interpolation position of the third laser cutting target track is adjusted, a fourth laser cutting target track is output to serve as the final track path of the laser head, and the position coordinates of the laser head are output. According to the method, a non-dominated sorting evolution mechanism of infeasible solution updating is provided, B spline interpolation is improved, the cutting track is optimized, the machining precision and efficiency are improved, and the defects of traditional mechanical cutting and existing laser cutting technologies in the aspect of processing complex geometrical shapes are overcome.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of laser cutting manufacturing technology, and in particular to a laser cutting trajectory interpolation method, device, equipment and storage medium. Background Art

[0002] In the shipbuilding process, sample boxes are important tools for verifying the design and production process. Traditional sample box manufacturing methods usually rely on manual cutting or mechanical cutting, which have the problems of low precision, low efficiency and high cost. With the development of laser technology, laser cutting has gradually been applied in the field of shipbuilding due to its advantages of high precision, high efficiency and low cost. However, in the laser cutting process, how to optimize the cutting trajectory to improve processing efficiency and quality is still a problem to be solved.

[0003] The current CNC laser cutting trajectory interpolation technology mainly focuses on the optimization of plane cutting and simple curve cutting. For complex three-dimensional surfaces and complex geometric shapes of marine sample boxes, the trajectory interpolation method has not been fully studied. The existing interpolation methods are often inefficient and inaccurate when processing complex shapes. In the related art, the discrete tool position data points of the laser cutting trajectory are fitted to generate a smooth NURBS curve to improve the shortcomings of the laser cutting curve, but this method has the following two disadvantages:

[0004] (1) When dealing with complex acceleration and deceleration control and parameter densification, the internal calculation process and decision-making mechanism of the laser cutting NURBS curve interpolation method are relatively complex and lack interpretability, making the behavior and results of the method difficult to understand and predict. Although the speed fluctuation compensation method can reduce errors, its compensation mechanism and effect are not intuitive and transparent, making it difficult to accurately evaluate its performance and applicability;

[0005] (2) In the NURBS curve interpolation method, the parameter densification method generally uses Taylor series expansion. However, due to the nonlinear relationship between arc length and parameters, this method will produce errors. The deviation between the actual interpolation chord length and the theoretical interpolation arc length cannot be completely eliminated, resulting in reduced interpolation accuracy. However, this method does not directly explain how to accurately select the interpolation point, so the interpolation accuracy of this method cannot be guaranteed in high-order cases. When calculating the interpolation parameters only by relying on Taylor series expansion method and Adams differential equation method, it is easy to ignore the high-order terms, resulting in the selected interpolation point not being the optimal one in the solution or the interpolation point that meets the accuracy requirements. Therefore, truncation errors may occur during use, which directly affect the interpolation accuracy. During the assembly line operation, as the number of iterations increases, it is highly likely that the requirements of high-precision processing and edge smoothness cannot be met at some point.

[0006] Laser cutting trajectory interpolation method is an important part of CNC technology. By optimizing the cutting trajectory, it can effectively reduce cutting time, improve cutting accuracy and reduce material waste. Therefore, it is of great practical significance to study and develop a suitable laser cutting trajectory interpolation method. Summary of the invention

[0007] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a laser cutting trajectory interpolation method, device, equipment and storage medium, propose a non-dominated sorting evolution mechanism for updating infeasible solutions, use the multi-objective function information of infeasible solutions to update the non-dominated sorting process, improve B-spline interpolation, and improve processing accuracy and efficiency by optimizing the cutting trajectory, thereby making up for the shortcomings of traditional mechanical cutting and existing laser cutting technology in processing complex geometric shapes.

[0008] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0009] A laser cutting trajectory interpolation method, comprising:

[0010] Import discrete laser head data points in the laser cutting trajectory;

[0011] Perform B-spline curve fitting on the imported laser head data points to obtain the first laser cutting target trajectory;

[0012] Set a speed constraint condition, interpolate the first laser cutting target trajectory, and obtain a second laser cutting target trajectory;

[0013] A multi-objective optimization model is constructed, and based on the multi-objective optimization model, a non-dominated sorting evolution mechanism of updating infeasible solutions is adopted to optimize the second laser cutting target trajectory to obtain the third laser cutting target trajectory;

[0014] The interpolation position of the third laser cutting target trajectory is adjusted, and the fourth laser cutting target trajectory is output. The fourth laser cutting target trajectory is used as the final trajectory path of the laser head and the position coordinates of the laser head are output.

[0015] Furthermore, the method of importing discrete laser head data points in the laser cutting trajectory is:

[0016] Obtain the cutting line of the marine sample box, discretize the cutting line through computer-aided manufacturing, and obtain discrete laser head data points;

[0017] Set process constraints, including maximum bow height error, interpolation cycle, maximum cutting speed, centripetal acceleration, and maximum centripetal jerk.

[0018] Furthermore, a speed constraint condition is set, and the first laser cutting target trajectory is interpolated to obtain the second laser cutting target trajectory as follows:

[0019] The pre-interpolation feed speed is obtained according to the maximum bow height error and the maximum centripetal acceleration limit conditions;

[0020] The pre-interpolation feed speed is used as a speed constraint condition to obtain the ideal step length in each interpolation cycle, and the corresponding interpolation point coordinates are obtained based on the ideal step length in each interpolation cycle.

[0021] Furthermore, a multi-objective optimization model is constructed, and the second laser cutting target trajectory is optimized by adopting a non-dominated sorting evolution mechanism for updating infeasible solutions based on the multi-objective optimization model. The method for obtaining the third laser cutting target trajectory is as follows:

[0022] Obtaining a first multi-objective optimization function based on process constraints, and constructing a multi-objective optimization model for the second laser cutting target trajectory;

[0023] Converting the multi-objective optimization model into a second multi-objective optimization function based on control points;

[0024] For the second multi-objective optimization function based on control points, the non-dominated solution set is updated by using the infeasible solution, and the Pareto optimal solution set is found through the non-dominated sorting evolution mechanism of the non-dominated solution set.

[0025] The third laser cutting target trajectory is obtained based on the Pareto optimal solution set.

[0026] Furthermore, the method of converting the multi-objective optimization model into a second multi-objective optimization function based on control points is:

[0027] Parametric control points: express process constraints as constraints on control points;

[0028] Adaptive pre-interpolation: dynamically adjust the speed and acceleration of pre-interpolation according to the specific position of each control point and the geometric characteristics of the curve;

[0029] Dynamic solution and optimization: Combined with the adaptive Simpson numerical integration method, the tool position data points of the second laser cutting target trajectory are fitted, and based on the real-time calculated curve length, the acceleration and deceleration control parameters are dynamically solved and optimized.

[0030] Furthermore, for the second multi-objective optimization function based on control points, the non-dominated solution set is updated by using the infeasible solution, and the method of finding the Pareto optimal solution set through the non-dominated sorting evolution mechanism of the non-dominated solution set is as follows:

[0031] Population generation: randomly generate an initial population with a population size of N;

[0032] Non-dominated sorting: Divide the initial population into different levels according to non-dominated sorting;

[0033] Fitness assignment: Calculate the crowding degree of individuals in each non-dominated level and assign a fitness value;

[0034] Selection operation: Use the binary tournament selection method to select individuals according to their fitness values ​​to enter the crossover and mutation operations, repeating population generation and non-dominated sorting until the required number of individuals are selected as the parent population;

[0035] Crossover and mutation: Perform simulated binary crossover and polynomial mutation operations on the parent population to generate a mutant population;

[0036] Merge population: merge the parent population and the mutant population to form a merged population with a population size of 2N;

[0037] Infeasible solution processing: Select infeasible solutions based on the merged population;

[0038] Non-dominated sorting update: perform a new non-dominated sorting on the infeasible solutions and divide them into different levels, and select the top N individuals into the next generation population;

[0039] Get the number of iterations n. If the number of iterations n is less than the maximum number of iterations N, max , repeat the selection operation to update the non-dominated sort until the number of iterations n reaches the preset maximum number of iterations N max .

[0040] Furthermore, infeasible solution processing: The method of selecting infeasible solutions based on the merged population is:

[0041] Calculate the value of the second multi-objective optimization function for each individual in the merged population;

[0042] Setting a first threshold of the second multi-objective optimization function, obtaining all first entities whose second multi-objective optimization function values ​​satisfy the first threshold of the second multi-objective optimization function, treating each first entity as an infeasible solution and calculating the constraint violation degree of the first entity, and imposing a certain penalty on the first entity based on the constraint violation degree;

[0043] Setting a second threshold of the second multi-objective optimization function, the second threshold of the second multi-objective optimization function is better than the first threshold of the second multi-objective optimization function, and obtaining all second individuals whose second multi-objective optimization function values ​​satisfy the second threshold of the second multi-objective optimization function;

[0044] Take all first bodies and all second bodies as updated infeasible solutions.

[0045] A laser cutting trajectory interpolation device, comprising:

[0046] Laser head data point acquisition module, used to import discrete laser head data points in the laser cutting trajectory;

[0047] A first laser cutting target trajectory acquisition module is used to perform B-spline curve fitting on the imported laser head data points to obtain the first laser cutting target trajectory;

[0048] A second laser cutting target trajectory acquisition module is used to set speed constraint conditions, interpolate the first laser cutting target trajectory, and obtain a second laser cutting target trajectory;

[0049] A third laser cutting target trajectory acquisition module is used to construct a multi-objective optimization model, and optimize the second laser cutting target trajectory by adopting a non-dominated sorting evolution mechanism of updating infeasible solutions based on the multi-objective optimization model to obtain a third laser cutting target trajectory;

[0050] The final trajectory path acquisition module is used to adjust the interpolation position of the third laser cutting target trajectory, output the fourth laser cutting target trajectory, use the fourth laser cutting target trajectory as the final trajectory path of the laser head and output the laser head position coordinates.

[0051] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the laser cutting trajectory interpolation method is implemented when the processor executes the program.

[0052] A non-transitory computer-readable storage medium stores a computer program, which implements the above-mentioned laser cutting trajectory interpolation method when executed by a processor.

[0053] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0054] The laser cutting trajectory interpolation method provided by the present invention proposes a non-dominated sorting evolution mechanism for updating infeasible solutions, uses the multi-objective function information of infeasible solutions to update the non-dominated sorting process, and adaptively applies population division and individual replacement. Based on this, the B-spline interpolation is improved. Compared with the multi-order spline curves in similar studies, it provides a smoother global trajectory while reducing the centripetal acceleration peak, reducing the uncertainty in the interpolation process, making the speed fluctuation during the interpolation process smaller, improving the processing quality and equipment stability, so that the laser cutting trajectory interpolation method provided by the present invention can solve the problem that the NURBS curve interpolation algorithm is not interpretable.

[0055] The present invention improves the constraint processing rules when high-order terms exist, takes into account the multi-objective function value information of infeasible individuals, and broadens the evolution direction. Especially when a large number of infeasible solutions appear when high-order terms exist, it can better balance the objectives and constraints, reduce truncation errors, and improve interpolation accuracy, so that during the assembly line operation, the requirements of high-precision processing and edge smoothness can be met.

[0056] The present invention reduces the greediness of typical feasibility rules, provides an alternative evolution path from the infeasible area to the feasible area, can find the optimal solution more efficiently, and can solve the degradation of interpolation point positioning selection caused by long-term iteration in the laser cutting assembly line process within a suitable computing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The illustrative embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0058] Figure 1 A flow chart of a laser cutting trajectory interpolation method of the present invention;

[0059] Figure 2 It is a schematic diagram of optimizing the second laser cutting target trajectory based on the multi-objective optimization model of the present invention by adopting the non-dominated sorting evolution mechanism of updating infeasible solutions;

[0060] Figure 3 It is a schematic diagram of a laser cutting trajectory interpolation device of the present invention. DETAILED DESCRIPTION

[0061] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0062] In the following description, a large number of specific details are provided to provide a more thorough understanding of the present invention. However, it is apparent to those skilled in the art that the present invention can be implemented without one or more of these details. In other examples, in order to avoid confusion with the present invention, some technical features well known in the art are not described.

[0063] It should be understood that the present invention can be implemented in different forms and should not be interpreted as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to make the disclosure thorough and complete and to fully convey the scope of the present invention to those skilled in the art.

[0064] The infeasible solutions and feasible solutions proposed by the present invention are explained below.

[0065] An infeasible solution is a solution that does not satisfy the problem constraints. In multi-objective optimization problems, infeasible solutions usually have the following characteristics:

[0066] Constraint Violation: If a solution violates one or more constraints, then it is infeasible. For example, in the set of decision variables h, if C(h) < 0, then the solution is infeasible.

[0067] Penalty mechanism: In order to deal with infeasible solutions during the optimization process, a penalty mechanism is usually introduced to reduce the fitness of these solutions, so that the optimization algorithm tends to explore feasible solutions. For example, in the improved feasibility rule, the greediness of infeasible solutions is reduced by considering the information of multi-objective function values, so as to better explore the optimal solution.

[0068] Non-dominated level: Infeasible solutions may be given a lower level in the non-dominated sort. This means that infeasible solutions will be preferentially excluded or inferior when selecting the next generation of population.

[0069] A feasible solution is a solution that satisfies all constraints. In multi-objective optimization problems, a feasible solution has the following characteristics:

[0070] Constraint Satisfaction: A solution is feasible if it satisfies all constraints. For example, in a set of decision variables h, if C(h) = 0, then the solution is feasible.

[0071] Optimization objective: A feasible solution not only satisfies the constraints, but also performs well on the optimization objective function. The optimization algorithm tends to select the best solution from the feasible solutions to maximize or minimize the overall objective.

[0072] Evolutionary trend: During the optimization process, the algorithm will gradually evolve from an infeasible solution to a feasible solution, and then further optimize from the feasible solution to the optimal solution. This evolutionary trend ensures that the algorithm can find the optimal solution based on satisfying the constraints.

[0073] Example 1

[0074] Embodiment 1 provides a laser cutting trajectory interpolation method, such as Figure 1 As shown, including:

[0075] Step S1: importing discrete laser head data points in the laser cutting trajectory;

[0076] Step S2: performing B-spline curve fitting on the imported laser head data points to obtain the first laser cutting target trajectory;

[0077] Step S3: setting speed constraint conditions, interpolating the first laser cutting target trajectory, and obtaining the second laser cutting target trajectory;

[0078] Step S4: construct a multi-objective optimization model, and optimize the second laser cutting target trajectory by adopting a non-dominated sorting evolution mechanism for updating infeasible solutions based on the multi-objective optimization model to obtain a third laser cutting target trajectory;

[0079] Step S5: adjusting the interpolation position of the third laser cutting target trajectory, outputting the fourth laser cutting target trajectory, taking the fourth laser cutting target trajectory as the final trajectory path of the laser head and outputting the laser head position coordinates.

[0080] In the related technologies, when discussing NURBS (Non-Uniform Rational B-Spline) curve interpolation algorithms, the lack of interpretability mainly refers to the following aspects:

[0081] The algorithm is complex and all specific parameters are not given in the relevant references. The definition and calculation of NURBS curves involve high mathematical complexity, including the calculation of basis functions and parameterization, which makes the algorithm difficult to understand and explain intuitively. Especially for engineers or technicians without a deep mathematical background, it is difficult to understand the principles and implementation details of the NURBS curve interpolation algorithm.

[0082] There is no simple analytical relationship between the parameters of a NURBS curve (usually represented by u) and the actual arc length. This means that when interpolating, it is impossible to accurately predict the position and velocity changes of points on the curve directly through the parameter values. This nonlinear relationship increases the uncertainty in the interpolation process and affects the precise control and calculation of the interpolation points.

[0083] In NURBS curve interpolation, it is difficult to accurately predict the deceleration point during the acceleration and deceleration process, which poses a challenge to the speed and centripetal acceleration control in actual processing. Traditional acceleration and deceleration control algorithms may not be directly applicable to NURBS curve interpolation, and complex adjustments and optimizations are required. This control difficulty leads to large speed fluctuations during the interpolation process, affecting processing quality and equipment stability.

[0084] NURBS curve interpolation needs to ensure accuracy while improving computational efficiency as much as possible. However, due to the large amount of calculation, directly using the definitional calculation method will lead to low efficiency, and it is necessary to introduce recursive algorithms to improve efficiency. This trade-off also makes the interpretation of the algorithm more complicated, because it is necessary to understand the advantages and disadvantages of different calculation methods and applicable scenarios.

[0085] At the same time, in the process of NURBS curve fitting of laser cutting trajectory, the accuracy of the interpolation algorithm is crucial to the processing quality.

[0086] For the problem of "unable to solve high-order terms, resulting in truncation errors", a specific explanation is given:

[0087] The NURBS curve interpolation algorithm often uses the Taylor expansion method for approximate calculations. Taylor expansion approximates the true value of the curve through the derivative of the curve at a certain point, but this approximation is only valid when it is truncated to a certain order. When high-order derivatives are ignored, an error will occur between the interpolation point and the true curve. This error is called truncation error.

[0088] The truncation error comes from the neglect of the high-order terms of the Taylor expansion. The truncation error will have a significant impact on the processing accuracy and speed stability. Processing accuracy: Due to the existence of truncation error, the tiny line segments between adjacent interpolation points cannot accurately follow the actual curve, resulting in the generation of bow height error. Bow height error refers to the distance error between the midpoint of the interpolated straight line segment and the actual curve. This error will directly affect the processing accuracy.

[0089] Speed ​​stability: Truncation error can also cause speed fluctuations between interpolation points. In theory, the speed between interpolation points should be uniform, but due to the existence of truncation error, the speed between actual interpolation points will change, resulting in speed fluctuations, affecting the processing quality and the stability of the machine tool.

[0090] In order to improve the interpolation accuracy, more high-order derivatives need to be considered, but this will greatly increase the amount of calculation and reduce the real-time performance of interpolation. In practical applications, in order to find a balance between accuracy and efficiency, only a finite-order Taylor expansion can be used, which inevitably introduces truncation errors.

[0091] Since the high-order terms are ignored, there is an error between the actual calculated step length and the theoretical step length. This error will lead to uneven distribution of interpolation points, further aggravating speed fluctuations and machining errors.

[0092] The present embodiment provides a laser cutting trajectory interpolation method, which proposes a non-dominated sorting evolution mechanism for updating infeasible solutions, uses the multi-objective function information of infeasible solutions to update the non-dominated sorting process, and adaptively applies population division and individual replacement. Based on this, the B-spline interpolation is improved. Compared with the multi-order spline curves in similar studies, it provides a smoother global trajectory and reduces the centripetal acceleration peak, reducing the uncertainty in the interpolation process, making the speed fluctuation during the interpolation process smaller, improving the processing quality and equipment stability, so that the laser cutting trajectory interpolation method provided by the present invention can solve the problem that the NURBS curve interpolation algorithm is not interpretable.

[0093] The present invention improves the constraint processing rules when high-order terms exist, takes into account the multi-objective function value information of infeasible individuals, and broadens the evolution direction. Especially when a large number of infeasible solutions appear when high-order terms exist, it can better balance the objectives and constraints, reduce truncation errors, and improve interpolation accuracy, so that during the assembly line operation, the requirements of high-precision processing and edge smoothness can be met.

[0094] Specifically, this embodiment includes:

[0095] 1. Problem definition and modeling.

[0096] First, the objective function is determined based on the known path points of the trajectory through B-spline interpolation and time information. The constraints are converted into a form derived from control points, which can simplify the constraints on the kinematic parameters.

[0097] 2. Constraint simplification and objective function determination.

[0098] The established spatial kinematic parameter constraints (motion range and limit settings for laser cutting) are simplified using a limited number of control points. The objective function is determined by the time-dependent design variables.

[0099] 3. Introduce INSEA algorithm.

[0100] Based on the above model, the Non-dominated Sorting Evolutionary Algorithm (INSEA) for Updating Infeasible Solutions is introduced to optimize the above model and find the best Pareto solution. The main improvement of this algorithm is to use the multi-objective function information of infeasible solutions to update the non-dominated sorting and increase the evolutionary direction.

[0101] The present invention reduces the greediness of typical feasibility rules, provides an alternative evolution path from the infeasible area to the feasible area, can find the optimal solution more efficiently, and can solve the degradation of interpolation point positioning selection caused by long-term iteration in the laser cutting assembly line process within a suitable computing efficiency.

[0102] In step S1 of this embodiment, the method of importing discrete laser head data points in the laser cutting trajectory is:

[0103] Step S101: obtaining a cutting line of a marine sample box, and discretizing the cutting line through computer-aided manufacturing to obtain discrete laser head data points;

[0104] Step S102: setting process constraints, including maximum bow height error, interpolation cycle, maximum cutting speed, centripetal acceleration and maximum centripetal jerk.

[0105] Specifically, the bow height error δ is determined by calculating the maximum deviation between the straight line segment between adjacent interpolation points and the actual curve. The specific formula is:

[0106]

[0107] Where L = v s T s is the feed step length in a single interpolation cycle, T s is the interpolation period, v s is the interpolation period, and R is the radius of curvature of the curve.

[0108] The existence of bow height error will reduce processing accuracy and affect cutting quality. The bow height error must be limited to a certain range, so the maximum bow height error needs to be set to ensure that the processing accuracy requirements are met. The feed speed needs to be adjusted according to the maximum allowable bow height error to ensure processing accuracy.

[0109] Interpolation cycle T s It is the time interval set in each interpolation calculation process. This time interval is usually fixed, but can be adjusted according to specific processing requirements.

[0110] The interpolation cycle determines the interval of the interpolation points, which affects the feed step length L and the bow height error δ. A smaller interpolation cycle can increase the density of the interpolation points and reduce the bow height error, but it will increase the calculation burden. By optimizing the interpolation cycle, a balance can be found between machining accuracy and calculation efficiency.

[0111] Maximum cutting speed V max It is the highest speed that the machine tool can reach under stable processing conditions. This parameter needs to be set according to the performance of the machine tool and the characteristics of the processed material. The maximum cutting speed limits the feed speed in actual processing to ensure that the machine tool operates within a controllable range.

[0112] When machining high curvature areas, the actual feed rate may need to be lower than the maximum cutting speed to avoid overcutting and speed fluctuations.

[0113] Maximum centripetal acceleration J max It is the maximum rate of change that the machine tool can withstand when the centripetal acceleration changes. This parameter is also determined by the mechanical and control performance of the machine tool.

[0114] The maximum centripetal acceleration control is to ensure the smooth change of centripetal acceleration and further reduce machine vibration and processing errors. During the acceleration and deceleration process, it is necessary to ensure that the maximum centripetal acceleration is within the allowable range to avoid the degradation of processing quality due to excessive changes in centripetal acceleration.

[0115] In step S3 of this embodiment, the speed constraint condition is set, and the first laser cutting target trajectory is interpolated to obtain the second laser cutting target trajectory by:

[0116] Step S301: obtaining the pre-interpolation feed speed according to the maximum bow height error and the maximum centripetal acceleration constraint conditions;

[0117] Step S302: taking the pre-interpolation feed speed as a speed constraint condition to obtain an ideal step length in each interpolation cycle, and obtaining corresponding interpolation point coordinates based on the ideal step length in each interpolation cycle.

[0118] In step S4 of this embodiment, a multi-objective optimization model is constructed, and a non-dominated sorting evolution mechanism for updating infeasible solutions is adopted based on the multi-objective optimization model to optimize the second laser cutting target trajectory, and a method for obtaining the third laser cutting target trajectory is as follows:

[0119] Step S401: obtaining a first multi-objective optimization function based on process constraints, and constructing a multi-objective optimization model for the second laser cutting target trajectory.

[0120] Step S402: converting the multi-objective optimization model into a second multi-objective optimization function based on control points.

[0121] Step S403: for the second multi-objective optimization function based on the control point, the non-dominated solution set is updated by using the infeasible solution, and the Pareto optimal solution set is found through the non-dominated sorting evolution mechanism of the non-dominated solution set.

[0122] Step S404: obtaining a third laser cutting target trajectory based on the Pareto optimal solution set.

[0123] In step S401 of this embodiment, the first multi-objective optimization function can be set according to actual needs, and the first multi-objective optimization function to be optimized is defined, for example, the smoothness, accuracy and processing time of the cutting trajectory.

[0124] In step S402 of this embodiment, in the control of the laser cutting trajectory, the process constraint conditions are converted into a form related to the control points in order to simplify the constraint processing process. The specific method of converting the multi-objective optimization model into a second multi-objective optimization function based on the control points includes the following steps:

[0125] Step S4021: parameterize control points: express process constraints (such as maximum cutting speed, maximum acceleration, etc.) as constraints on control points.

[0126] The purpose of this is to apply constraints directly to the control points of the interpolation points so that they can be adjusted and controlled in real time during the trajectory interpolation process.

[0127] Step S4022: Adaptive pre-interpolation: dynamically adjust the speed and acceleration of pre-interpolation according to the specific position of each control point and the geometric characteristics of the curve.

[0128] This method can avoid excessive acceleration changes and speed fluctuations while ensuring cutting quality.

[0129] Step S4023: Dynamic solution and optimization: Combined with the adaptive Simpson numerical integration method, the tool position data points of the second laser cutting target trajectory are fitted, and based on the real-time calculated curve length, the acceleration and deceleration control parameters are dynamically solved and optimized.

[0130] This approach not only simplifies the handling of constraints, but also improves the accuracy and smoothness of the cutting.

[0131] In step S403 of this embodiment, for the second multi-objective optimization function based on the control point, the non-dominated solution set is updated using the infeasible solution, such as Figure 2 As shown in Figure 2, the method to find the Pareto optimal solution set through the non-dominated sorting evolution mechanism of the non-dominated solution set is:

[0132] Step S4031: Population generation: randomly generate an initial population P0 with a population size of N.

[0133] The generation rule of the initial population is the way to create the first population in the evolutionary algorithm. The so-called "chromosome" has the following main methods:

[0134] Random generation: Randomly generate a certain number of individuals in the defined solution space. This is the most common method and can ensure the diversity of the population.

[0135] Heuristic generation: Use knowledge of the problem domain or heuristic rules to generate initial solutions, thereby improving the quality of the initial population.

[0136] Hybrid generation: Combining random generation and heuristic generation methods to ensure diversity and improve the quality of initial solutions.

[0137] Random generation is adopted in the case of the present invention.

[0138] Step S4032: Non-dominated sorting: Divide the initial population into different levels F1, F2, ..., F n .

[0139] Step S4033: fitness assignment: calculate the crowding degree of individuals in each non-dominated level and assign a fitness value.

[0140] Step S4034: Selection operation: Use the binary tournament selection method to select individuals according to their fitness values ​​to enter the crossover and mutation operations, repeating population generation and non-dominated sorting until the required number of individuals are selected as the parent population P.

[0141] Binary Tournament Selection is a selection mechanism in genetic algorithms. The specific steps are as follows:

[0142] Randomly select two individuals: Randomly select two individuals from the initial population.

[0143] Fitness comparison: Compare the fitness values ​​of two individuals and select the individual with higher fitness value as the winner.

[0144] Step S4035: Crossover and mutation: Perform simulated binary crossover (SBX) and polynomial mutation operations on the parent population P to generate a mutant population Q.

[0145] Simulated Binary Crossover (SBX) specifically includes:

[0146] Select parents: Select two parent individuals (p1 and p2) from the parent population P

[0147] Calculate offspring genes: Generate mutant individuals using the following formula: Q1, Q2, ..., Q n

[0148] c1=0.5×[(1+β)×p1+(1-β)×p2]

[0149] c2=0.5×[(1-β)×p1+(1+β)×p2]

[0150] Among them, β is a random number that depends on the parent individuals. New individuals are generated based on the calculated offspring genes.

[0151] Polynomial Mutation:

[0152] Select mutation points: randomly select a gene in an individual, execute the neighborhood pairing strategy, as well as differential evolution and polynomial mutation to generate mutant individuals.

[0153] Generate sufficient mutant individuals through binary crossover and polynomial mutation to form a mutant population defined as Q;

[0154] Calculate the variation: Use the following formula to calculate the variation: Where r is a random number between [0,1] and η is the variation distribution index.

[0155] Generate mutant genes: adjust the selected gene value according to the mutation amount.

[0156] Step S4036: Merge populations: merge the parent population P and the mutant population Q to form a merged population with a population size of 2N.

[0157] Step S4037: Infeasible solution processing: selecting an infeasible solution based on the merged population.

[0158] Step S4038: non-dominated sorting update: perform a new non-dominated sorting on the infeasible solutions and divide them into different levels, and select the top N individuals into the next generation population;

[0159] Step S4039: Get the number of iterations n. If the number of iterations n is less than the maximum number of iterations N, max, repeat the selection operation to update the non-dominated sort until the number of iterations n reaches the preset maximum number of iterations N max .

[0160] In step S4037, infeasible solution processing: the method of selecting infeasible solutions based on the merged population is:

[0161] Calculate the value of the second multi-objective optimization function for each individual in the merged population;

[0162] Setting a first threshold of the second multi-objective optimization function, obtaining all first entities whose second multi-objective optimization function values ​​satisfy the first threshold of the second multi-objective optimization function, treating each first entity as an infeasible solution and calculating the constraint violation degree of the first entity, and imposing a certain penalty on the first entity based on the constraint violation degree;

[0163] Setting a second threshold of the second multi-objective optimization function, the second threshold of the second multi-objective optimization function is better than the first threshold of the second multi-objective optimization function, and obtaining all second individuals whose second multi-objective optimization function values ​​satisfy the second threshold of the second multi-objective optimization function;

[0164] Take all first bodies and all second bodies as updated infeasible solutions.

[0165] Specifically, the following strategy can be adopted: based on the value of the second multi-objective optimization function, the merged population R t All individuals in the algorithm are sorted, and those individuals that violate the constraints but have better values ​​in the second multi-objective optimization function are selected, that is, individuals whose values ​​of the second multi-objective optimization function meet the first threshold of the second multi-objective optimization function. Appropriate penalties are imposed on infeasible solutions according to the degree of constraint violation, but at the same time, some infeasible solutions that have advantages in the second multi-objective optimization function value are retained, that is, individuals whose values ​​of the second multi-objective optimization function meet the first threshold of the second multi-objective optimization function, so as to increase the diversity and exploration ability of the population.

[0166] Constraint Violation:

[0167] For each individual, calculate its constraint violation degree, that is, the degree of violation of the constraint. Usually, it is the sum of all the violations of the constraint.

[0168]

[0169] Among them, g i (x) is the i-th constraint function.

[0170] Punishment formulation:

[0171] For individuals that violate the constraints, a penalty term is introduced into the fitness function:

[0172] F(x)=f(x)+λV(x)

[0173] Among them, f(x) is the original objective function and λ is the penalty coefficient.

[0174] Example:

[0175] Suppose there is an objective function f(x)=x 2 , with the constraint x≥1.

[0176] For individual x=0.5, its constraint violation degree V(x)=max(0.1-0.5)=0.5.

[0177] Fitness value F(x) = x 2 +λV(x)=0.25+λ0.5.

[0178] In step S4039 of this embodiment, the infeasible solutions are subjected to a three-stage non-dominated sorting process through repeated iterations, specifically:

[0179] All individuals are infeasible solutions to feasible solutions, and the number of individuals is less than N;

[0180] The number of feasible solutions is less than N and the number of feasible solutions is greater than N;

[0181] The feasible solution individuals are greater than N and the non-dominated sorting produces the optimal solution.

[0182] In each stage, the algorithm selects the best solution from the total population as the parent population for the next generation by merging the parent and child populations.

[0183] Through step S4, the third laser cutting target trajectory is obtained relative to the second laser cutting target trajectory, and the interpolation point position is adjusted to meet the optimization target.

[0184] In step S5 of this embodiment, the method of adjusting the interpolation position of the third laser cutting target trajectory and outputting the fourth laser cutting target trajectory is:

[0185] Calculate interpolation point parameters: Use the optimized interpolation points to estimate the second-order Taylor expansion interpolation parameters to ensure the accuracy of the interpolation points.

[0186] Specifically, the method for estimating the second-order Taylor expansion interpolation parameters using the optimized interpolation points is:

[0187] Step S501: define the objective function and constraints. The objective function is the position of the interpolation point, which is estimated by the second-order Taylor expansion formula. The constraints include the limits of bow height error, feed speed, acceleration and jerk.

[0188] Step S502: Second-order Taylor expansion formula: treat the parameter u as a function u(t) that changes continuously with time t, and use the second-order Taylor expansion formula to estimate the interpolation parameters

[0189] Where Δt is the time increment, u(t) is the first-order derivative of parameter u, is the second derivative of parameter u.

[0190] Step S503: Calculate the bow height error:

[0191] Step S504: interpolation direction selection: To improve accuracy, forward interpolation is performed from the starting point to the end point of the curve segment, and reverse interpolation is performed from the end point to the starting point. The current speeds of the forward and reverse interpolations are compared to decide whether to perform forward interpolation or reverse interpolation, and to determine whether the convergence condition is met to ensure that the speed and acceleration of the forward and reverse interpolations at the convergence point are both zero.

[0192] Step S505: speed fluctuation analysis: evaluate the speed fluctuation during the interpolation process to ensure interpolation accuracy. The speed fluctuation of the second-order Taylor expansion interpolation parameter estimation algorithm should be as small as possible, and the maximum speed fluctuation should not exceed the allowable range (eg 0.001%).

[0193] Step S506: Specifically compare and judge the accuracy of the bow height error and the speed fluctuation rate.

[0194] Bow height error:

[0195] Compare the bow height error of the second-order Taylor expansion interpolation point with the maximum allowable bow height error of actual CNC machining. For example, the maximum bow height error of the second-order Taylor expansion interpolation parameter estimation algorithm is 6.84μm, which exceeds the maximum allowable bow height error of CNC machining by 5μm, so it is necessary to improve the algorithm or choose other methods.

[0196] Speed ​​Fluctuation Rate:

[0197] The speed fluctuation rate during the interpolation process is compared with the maximum allowable speed fluctuation rate. For example, the maximum speed fluctuation rate of the second-order Taylor expansion interpolation parameter estimation algorithm is 0.005%, which is much larger than the maximum allowable speed fluctuation rate of 0.001%. Therefore, the algorithm needs to be improved to reduce the speed fluctuation.

[0198] Comparison of actual interpolation and theoretical interpolation:

[0199] The distance between the actual interpolation point and the theoretical interpolation point is compared, and the accuracy is judged by calculating the deviation between the actual interpolation chord length and the theoretical interpolation arc length.

[0200] The role of trajectory length is:

[0201] Calculation of interpolation points:

[0202] The trajectory length is used to determine the position of the interpolation point. The interpolation algorithm calculates the coordinates of each interpolation point based on the total length of the trajectory and the interpolation step size.

[0203] In NURBS curve interpolation, the parameter space will be discretized, and the interpolation step s is determined by the feed speed u and the interpolation period t: s = v(u) × t.

[0204] Feed speed control:

[0205] The distance between interpolation points (i.e. part of the trajectory length) affects the calculation of feed rate. The actual feed rate direction is along the chord length, while the theoretical feed rate direction is along the arc length of the curve.

[0206] Feed rate fluctuation describes the deviation of the actual feed rate from the theoretical feed rate, which is caused by the path length.

[0207] Acceleration and deceleration control:

[0208] During the interpolation process, the acceleration and deceleration process is determined according to the trajectory length and curve shape. The control of acceleration and deceleration should be determined based on the speed, acceleration and trajectory length of the current interpolation point. The trajectory length and the speed of the interpolation start and end points will affect the type of convergence of the forward and reverse interpolation, thereby affecting the entire acceleration and deceleration control process.

[0209] The role of Adams and least squares solutions

[0210] A method for solving high-order Taylor formulas to obtain approximate estimated solutions.

[0211] The main function of the Adams formula in interpolation is to estimate the parameter value of the next interpolation point. By using the parameters and derivative values ​​of multiple previous interpolation points, the Adams formula can provide a more accurate interpolation point location. The combination of explicit and implicit Adams formulas can better control the truncation error in the interpolation process and improve interpolation accuracy.

[0212] The explicit fourth-order Adams prediction formula is as follows:

[0213] Among them, f i Indicates parameter u i The derivative of , h represents the step size.

[0214] The implicit fourth-order Adams correction formula is as follows

[0215] In summary, this embodiment has the following advantages:

[0216] 1. Social effect.

[0217] By introducing the "non-dominated sorting evolutionary algorithm for updating infeasible solutions" and B-splines as combined research content, the present invention has made significant progress in laser cutting trajectory optimization. The specific social effects are as follows:

[0218] Improve production efficiency: By optimizing the laser cutting trajectory and reducing the cutting time, the production efficiency of ship sample boxes is improved. This not only relieves production pressure, but also improves the efficiency of the overall manufacturing industry.

[0219] Improve product quality: Higher precision cutting technology reduces cutting errors, improves the quality of ship sample boxes, reduces subsequent processing steps, and improves product consistency and reliability.

[0220] Energy saving and environmental protection: Optimized cutting paths and reduced errors reduce energy consumption and material waste, which is in line with the concept of green manufacturing and has important environmental protection significance.

[0221] 2. Specific contact for the application of ship sample boxes.

[0222] The manufacturing of ship sample boxes has high requirements for cutting accuracy and efficiency. The simple use of NURBS curve interpolation method has certain limitations, such as poor interpretation and reduced accuracy after multiple iterations. By introducing the non-dominated sorting evolutionary algorithm for updating infeasible solutions and B-spline technology, the present invention shows advantages in the following aspects:

[0223] Solve the interpretability problem of NURBS algorithm: The non-dominated sorting evolutionary algorithm for updating infeasible solutions has good global search capabilities, can effectively optimize the distribution of control points, and improve the interpretability of NURBS algorithm.

[0224] Improve the accuracy after multiple iterations: B-spline has good local control characteristics, combined with the optimized NURBS curve interpolation algorithm, it can maintain a high-precision cutting path after multiple iterations.

[0225] The specific manifestations are:

[0226] 1. Research on B-spline combination.

[0227] Control point optimization: By reasonably distributing control points, the B-spline curve can better fit the complex cutting trajectory, improving the smoothness and accuracy of the overall cutting path.

[0228] Reduce error accumulation: The local nature of B-splines reduces the error accumulation problem and improves the overall accuracy of the curve.

[0229] 2. Non-dominated sorting evolutionary algorithm for updating infeasible solutions.

[0230] Multi-objective optimization: NSGA-II performs well in solving multi-objective optimization problems. It can optimize the smoothness and speed of the cutting path at the same time, achieving a balance between high efficiency and high precision.

[0231] Avoid local optimum: By introducing an update mechanism for infeasible solutions, the algorithm effectively avoids falling into local optimum, making the optimization of the cutting path more global and reasonable.

[0232] According to the laser cutting trajectory interpolation method provided in this embodiment, scientific analysis is performed and corresponding test results are obtained, and the following can be obtained:

[0233] 1. This method can improve cutting quality and efficiency.

[0234] Test results: In the actual cutting experiment, the laser cutting trajectory optimization using the method of the present invention increased the cutting efficiency by about 15%, decreased the cutting quality scrap rate by about 10%, and significantly reduced the workpiece loss and subsequent processing requirements.

[0235] 2. This method has certain economic and technical benefits.

[0236] Economic benefits: The improved cutting efficiency and quality directly reduce the consumption of materials and energy, and reduce production costs. Taking the application of a shipyard in Shanghai as an example, it can save about 15% of production costs each year.

[0237] Technical benefits: The optimized cutting trajectory reduces the wear of laser cutting equipment, increases the service life of the equipment, and enhances the stability and reliability of the production line.

[0238] Example 2

[0239] Embodiment 2 provides a laser cutting trajectory interpolation device, such as Figure 3 As shown, including:

[0240] Laser head data point acquisition module, used to import discrete laser head data points in the laser cutting trajectory;

[0241] A first laser cutting target trajectory acquisition module is used to perform B-spline curve fitting on the imported laser head data points to obtain the first laser cutting target trajectory;

[0242] A second laser cutting target trajectory acquisition module is used to set speed constraint conditions, interpolate the first laser cutting target trajectory, and obtain a second laser cutting target trajectory;

[0243] A third laser cutting target trajectory acquisition module is used to construct a multi-objective optimization model, and optimize the second laser cutting target trajectory by adopting a non-dominated sorting evolution mechanism of updating infeasible solutions based on the multi-objective optimization model to obtain a third laser cutting target trajectory;

[0244] The final trajectory path acquisition module is used to adjust the interpolation position of the third laser cutting target trajectory, output the fourth laser cutting target trajectory, use the fourth laser cutting target trajectory as the final trajectory path of the laser head and output the laser head position coordinates.

[0245] Example 3

[0246] Embodiment 3 provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned laser cutting trajectory interpolation method when executing the program.

[0247] Example 4

[0248] Embodiment 4 provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the above-mentioned laser cutting trajectory interpolation method is implemented.

[0249] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0250] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0251] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A laser cutting trajectory interpolation method, characterized in that: include: Import discrete laser head data points in the laser cutting trajectory; Perform B-spline curve fitting on the imported laser head data points to obtain the first laser cutting target trajectory; Set a speed constraint condition, interpolate the first laser cutting target trajectory, and obtain a second laser cutting target trajectory; A multi-objective optimization model is constructed, and based on the multi-objective optimization model, a non-dominated sorting evolution mechanism of updating infeasible solutions is adopted to optimize the second laser cutting target trajectory to obtain the third laser cutting target trajectory; The interpolation position of the third laser cutting target trajectory is adjusted, and the fourth laser cutting target trajectory is output. The fourth laser cutting target trajectory is used as the final trajectory path of the laser head and the position coordinates of the laser head are output.

2. The laser cutting trajectory interpolation method according to claim 1, characterized in that: The method to import discrete laser head data points in the laser cutting trajectory is: Obtain the cutting line of the marine sample box, discretize the cutting line through computer-aided manufacturing, and obtain discrete laser head data points; Set process constraints, including maximum bow height error, interpolation cycle, maximum cutting speed, centripetal acceleration, and maximum centripetal jerk.

3. The laser cutting trajectory interpolation method according to claim 1, characterized in that: The method of setting the speed constraint condition and interpolating the first laser cutting target trajectory to obtain the second laser cutting target trajectory is as follows: The pre-interpolation feed speed is obtained according to the maximum bow height error and the maximum centripetal acceleration limit conditions; The pre-interpolation feed speed is used as a speed constraint condition to obtain the ideal step length in each interpolation cycle, and the corresponding interpolation point coordinates are obtained based on the ideal step length in each interpolation cycle.

4. The laser cutting trajectory interpolation method according to claim 1, characterized in that: A multi-objective optimization model is constructed. Based on the multi-objective optimization model, the non-dominated sorting evolution mechanism of infeasible solution update is adopted to optimize the second laser cutting target trajectory. The method for obtaining the third laser cutting target trajectory is as follows: Obtaining a first multi-objective optimization function based on process constraints, and constructing a multi-objective optimization model for the second laser cutting target trajectory; Converting the multi-objective optimization model into a second multi-objective optimization function based on control points; For the second multi-objective optimization function based on control points, the non-dominated solution set is updated by using the infeasible solution, and the Pareto optimal solution set is found through the non-dominated sorting evolution mechanism of the non-dominated solution set. The third laser cutting target trajectory is obtained based on the Pareto optimal solution set.

5. The laser cutting trajectory interpolation method according to claim 4, characterized in that: The method of converting the multi-objective optimization model into a second multi-objective optimization function based on control points is: Parametric control points: express process constraints as constraints on control points; Adaptive pre-interpolation: dynamically adjust the speed and acceleration of pre-interpolation according to the specific position of each control point and the geometric characteristics of the curve; Dynamic solution and optimization: Combined with the adaptive Simpson numerical integration method, the tool position data points of the second laser cutting target trajectory are fitted, and based on the real-time calculated curve length, the acceleration and deceleration control parameters are dynamically solved and optimized.

6. The laser cutting trajectory interpolation method according to claim 4, characterized in that: For the second multi-objective optimization function based on control points, the method of updating the non-dominated solution set by using the infeasible solution and finding the Pareto optimal solution set through the non-dominated sorting evolution mechanism of the non-dominated solution set is: Population generation: randomly generate an initial population with a population size of N; Non-dominated sorting: Divide the initial population into different levels according to non-dominated sorting; Fitness assignment: Calculate the crowding degree of individuals in each non-dominated level and assign a fitness value; Selection operation: Use the binary tournament selection method to select individuals according to their fitness values ​​to enter the crossover and mutation operations, repeating population generation and non-dominated sorting until the required number of individuals are selected as the parent population; Crossover and mutation: Perform simulated binary crossover and polynomial mutation operations on the parent population to generate a mutant population; Merge population: merge the parent population and the mutant population to form a merged population with a population size of 2N; Infeasible solution processing: Select infeasible solutions based on the merged population; Non-dominated sorting update: perform a new non-dominated sorting on the infeasible solutions and divide them into different levels, and select the top N individuals into the next generation population; Get the number of iterations n. If the number of iterations n is less than the maximum number of iterations N, max , repeat the selection operation to update the non-dominated sort until the number of iterations n reaches the preset maximum number of iterations N max .

7. The laser cutting trajectory interpolation method according to claim 6, characterized in that: Infeasible solution processing: The method of selecting infeasible solutions based on the merged population is: Calculate the value of the second multi-objective optimization function for each individual in the merged population; Setting a first threshold of the second multi-objective optimization function, obtaining all first entities whose second multi-objective optimization function values ​​satisfy the first threshold of the second multi-objective optimization function, treating each first entity as an infeasible solution and calculating the constraint violation degree of the first entity, and imposing a certain penalty on the first entity based on the constraint violation degree; Setting a second threshold of the second multi-objective optimization function, the second threshold of the second multi-objective optimization function is better than the first threshold of the second multi-objective optimization function, and obtaining all second individuals whose second multi-objective optimization function values ​​satisfy the second threshold of the second multi-objective optimization function; Take all first bodies and all second bodies as updated infeasible solutions.

8. A laser cutting trajectory interpolation device, characterized in that: include: Laser head data point acquisition module, used to import discrete laser head data points in the laser cutting trajectory; A first laser cutting target trajectory acquisition module is used to perform B-spline curve fitting on the imported laser head data points to obtain the first laser cutting target trajectory; A second laser cutting target trajectory acquisition module is used to set speed constraint conditions, interpolate the first laser cutting target trajectory, and obtain a second laser cutting target trajectory; A third laser cutting target trajectory acquisition module is used to construct a multi-objective optimization model, and optimize the second laser cutting target trajectory by adopting a non-dominated sorting evolution mechanism of updating infeasible solutions based on the multi-objective optimization model to obtain a third laser cutting target trajectory; The fourth laser cutting target trajectory acquisition module is used to adjust the interpolation position of the third laser cutting target trajectory, output the fourth laser cutting target trajectory, use the fourth laser cutting target trajectory as the final trajectory path of the laser head and output the laser head position coordinates.

9. An electronic device, characterized in that: The invention comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the program, the laser cutting trajectory interpolation method is realized.

10. A non-transitory computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the above-mentioned laser cutting trajectory interpolation method is implemented.

Citation Information

Cited By

  • Sawing method and system of high-speed circular sawing machine based on least square method

    CN120901759A

  • Method for automatically generating and optimizing crushed material line of cutting piece

    CN121745646A