A stator core cutting process optimization management method

Abstract

The present invention discloses a method for optimizing the cutting process management of a stator core, and relates to the technical field related to laser cutting. The method comprises: obtaining a laser cutting record set based on the attribute information of the silicon steel sheet to be cut as a constraint; identifying cutting defect indicators; determining the target cutting constraint and evaluating and assigning the impact; screening the cutting control parameters through the laser cutting record set; building a cutting quality evaluation function, using a cutting simulation model for optimization analysis, and outputting the optimized cutting parameter threshold; establishing a mapping association to achieve the optimization guidance of subsequent cutting parameters. The method solves the technical problem that the unreasonable setting of cutting parameters leads to a large amount of heat-affected zone and burr accumulation on the cutting surface, causing adhesion between the stator sheet and the lower support, thereby affecting the cutting accuracy and quality of the stator sheet surface, and achieves the technical effect of reasonably setting parameters to improve the cutting accuracy and cutting quality of the stator core.

Description

Technical Field

[0001] The present application relates to the field related to laser cutting technology, and specifically to a method for optimizing the cutting process of a stator core. Background Art

[0002] As a key component in motor manufacturing, the quality and production efficiency of the stator core directly affect the overall performance and production cost of the motor. With the continuous development of industrial manufacturing technology, the requirements for stator core cutting technology are becoming higher and higher. In order to achieve precise cutting of the stator core and improve material utilization and production efficiency, major motor manufacturers have begun to seek solutions for process optimization management, upgrade cutting equipment, adjust cutting parameters, and optimize production processes. However, traditional stator core cutting processes often have unreasonable parameter settings, insufficient precision, and serious material waste, resulting in large amounts of surface heat-affected zones and burr accumulation, as well as low production efficiency. This not only increases manufacturing costs, but may also lead to a decline in motor performance.

[0003] Therefore, in the current technologies related to stator core cutting process optimization, there is a technical problem that due to unreasonable cutting parameter settings, a large amount of heat-affected zone and burrs accumulate on the cutting surface, causing adhesion between the stator sheets and the underlying support parts, thereby affecting the cutting accuracy and quality of the stator sheet surface. Summary of the invention

[0004] The present application provides a method for optimizing the cutting process management of a stator core, and adopts technical means such as microscope image acquisition, construction of simulation models, establishment of quality evaluation functions, and mapping association, to solve the technical problem existing in the existing cutting process optimization management that unreasonable setting of cutting parameters leads to large accumulation of heat-affected zones and burrs on the cutting surface, causing adhesion between the stator sheets and the underlying support members, thereby affecting the cutting accuracy and quality of the stator sheet surface, thereby achieving the technical effect of improving the cutting accuracy and cutting quality of the stator core by reasonably setting parameters.

[0005] The present application provides a method for optimizing the cutting process of a stator core, the method comprising: reading the attribute information of a silicon steel sheet to be cut, using the attribute information as a constraint, retrieving a laser cutting log, and obtaining a laser cutting record set, wherein the attribute information includes material thickness, surface quality, composition, and proportion; identifying cutting defect indicators, wherein the cutting defect indicators include section shape, section angle, section smoothness, burr thickness, and heat-affected zone width; determining a target cutting constraint, and performing an influence evaluation on the cutting defect indicator based on the target cutting constraint, assigning a value to the cutting defect indicator according to the influence evaluation result, and obtaining an assigned cutting defect indicator; based on the single factor analysis principle and the assigned cutting defect indicator, The laser cutting record set performs cutting control parameter screening to determine key cutting parameters and adaptive cutting thresholds, wherein the key cutting parameters include laser power, cutting speed, gas pressure and defocus; a cutting quality evaluation function is constructed based on the assigned cutting defect index, and a laser cutting simulation model is used to optimize and analyze the key cutting parameters, output an optimized cutting parameter threshold, and select the optimal cutting parameter of the optimized cutting parameter threshold to execute the cutting process of the silicon steel sheet to be cut; a mapping association is established among the attribute information, the target cutting constraint and the optimized cutting parameter threshold, and a cutting parameter matching library is established based on the mapping association, and subsequent cutting parameter optimization guidance is realized through the cutting parameter matching library.

[0006] In a possible implementation, based on the single factor analysis principle and the assigned cutting defect index, the cutting control parameters are screened through the laser cutting record set to determine the key cutting parameters and the adaptive cutting threshold, and the following processing is performed: the parameter tolerance threshold of the cutting control parameter is configured, wherein each cutting control parameter has a unique parameter tolerance threshold; a first cutting control parameter is selected, and the first cutting control parameter is used as a variable, and other cutting control parameters except the first cutting control parameter are used as quantities, and the laser cutting record set is screened using the parameter tolerance threshold to determine the first cutting record set; based on the single factor analysis principle and the first cutting record set, the first cutting control parameter and the assigned cutting defect index are evaluated for correlation, and the key cutting parameters and the adaptive cutting threshold are determined according to the correlation evaluation result.

[0007] In a possible implementation, the key cutting parameters and the adaptive cutting threshold are determined according to the association evaluation result, and the following processing is performed: the indicator tolerance threshold of the assigned cutting defect indicator is configured, wherein each assigned cutting defect indicator has a unique indicator tolerance threshold; based on the principle of single factor analysis, a first assigned cutting defect indicator is selected, the first assigned cutting defect indicator is used as a variable, and other assigned cutting defect indicators except the first assigned cutting defect indicator are used as quantitative factors, and the first cutting record set is screened based on the indicator tolerance threshold to obtain a first associated cutting record set; a first trend influence curve is constructed based on the first associated cutting record set, and a trend influence analysis is performed on the first cutting control parameter and the first assigned cutting defect indicator according to the first trend influence curve to generate a first associated scalar; the first trend influence curve is identified according to a predetermined defect indicator constraint, and a first initial adaptive cutting threshold is determined; the cutting control parameters are arranged from large to small according to the associated scalar to generate a first cutting parameter sequence, and a first initial adaptive cutting threshold sequence of the first cutting parameter sequence is obtained, and the first cutting parameter sequence and the first cutting defect indicator correspond one to one; the key cutting parameters and the adaptive cutting threshold are determined based on the first cutting parameter sequence and the first initial adaptive cutting threshold sequence.

[0008] In a possible implementation, the key cutting parameters and the adaptive cutting threshold are determined based on the first cutting parameter sequence and the first initial adaptive cutting threshold sequence, and the following processing is also performed: a plurality of cutting parameter sequences are analyzed and acquired in sequence, and cutting parameters are extracted from the plurality of cutting parameter sequences according to preset rankings to generate a plurality of initial key cutting parameter sets; the plurality of initial key cutting parameter sets are weighted with influence as weight, and weighted calculation is performed on the same initial key cutting parameters, and they are arranged from large to small according to the weighted calculation results to generate an initial key cutting parameter sequence; the initial key cutting parameters of the preset ranking in the initial key cutting parameter sequence are selected and set as the key cutting parameters; a plurality of initial adaptive cutting threshold sets corresponding to the key cutting parameters are acquired, and the intersection of the thresholds in the initial adaptive cutting threshold set is extracted and set as the adaptive cutting threshold.

[0009] In a possible implementation, a cutting quality evaluation function is built based on the assigned cutting defect index, and the key cutting parameters are optimized and analyzed using a laser cutting simulation model to output an optimized cutting parameter threshold. The following processing is also performed: a cutting quality evaluation function is built based on the assigned cutting defect index, wherein the cutting quality evaluation result is the inverse of the sum of the assigned cutting defect evaluation results; a parameter optimization space is built using the adaptive cutting threshold, and a plurality of cutting parameter combinations are randomly generated based on the parameter optimization space; based on digital twin technology, the laser cutting simulation model is built in combination with the attribute information of the silicon steel sheet to be cut and the laser cutting equipment information; based on the cutting quality evaluation function, the adaptive cutting threshold, the plurality of cutting parameter combinations and the laser cutting simulation model, the key cutting parameters are optimized and analyzed to obtain the optimized cutting parameter threshold.

[0010] In a possible implementation, the key cutting parameters are optimized and analyzed based on the cutting quality evaluation function, the adaptive cutting threshold, the multiple cutting parameter combinations and the laser cutting simulation model, and the following processing is also performed: based on the target cutting path, the multiple cutting parameter combinations are simulated cutting processing performed through the laser cutting simulation model to generate multiple simulated processing results; the multiple simulated processing results are evaluated using the cutting quality evaluation function to generate a cutting quality evaluation coefficient sequence; the multiple cutting parameter combinations are set as multiple initial solutions, and the multiple initial solutions are mapped and arranged based on the cutting quality evaluation coefficient sequence to obtain an initial solution sequence; with the adaptive cutting threshold as the constraint space, the key cutting parameters are optimized and analyzed based on the initial solution sequence to obtain the optimized cutting parameter threshold.

[0011] In a possible implementation, the adaptive cutting threshold is used as a constraint space, the key cutting parameter is optimized and analyzed based on the initial solution sequence to obtain the optimized cutting parameter threshold, and the following processing is performed: the first N initial solutions of the initial solution sequence are extracted and set as the first solution, and the other initial solutions except the first solution are set as the tail solutions, where N is an integer greater than 10; the tail solutions are clustered with the first solution as the leader to obtain N cluster neighborhoods, where the clustering principle is random clustering, and the number of tail solutions in each cluster neighborhood is the same; the optimization adjustment step of the key cutting parameter is configured, where each key cutting parameter has a unique optimization adjustment step; the N The tail solutions in each clustering neighborhood are adjusted once, and the adjusted tail solutions are evaluated by the cutting quality evaluation function to obtain multiple tail solution cutting quality evaluation coefficients; whether the tail solution cutting quality evaluation coefficient is greater than the cutting quality evaluation coefficient of the first solution in the same clustering neighborhood is determined, and if so, the first solution is updated with the adjusted tail solution to obtain multiple updated neighborhoods; iterative optimization is continued until a predetermined number of constraints are met, multiple optimized neighborhoods are output, and the mean values ​​of the multiple cutting quality evaluation coefficients of the multiple optimized neighborhoods are calculated; the optimized neighborhood corresponding to the maximum mean value of the cutting quality evaluation coefficient is selected as the optimal neighborhood, and the optimized cutting parameter threshold is set based on the optimal neighborhood.

[0012] In a possible implementation, a cutting parameter matching library is established based on the mapping association, and the following processing is performed: configuring the first error interval and the second error interval of the attribute information and the target cutting constraint; expanding the attribute information and the target cutting constraint based on the first error interval and the second error interval to generate expanded attribute information and expanded cutting constraints; based on the decision tree principle, the cutting parameter matching library is constructed with the expanded attribute information as the first decision node, the expanded cutting constraint as the second decision node, and the optimized cutting parameter threshold as the associated child node of the first decision node and the second decision node.

[0013] The present application proposes a method for optimizing the cutting process of the stator core, which uses the attribute information of the silicon steel sheet to be cut as a constraint to obtain a laser cutting record set; identify cutting defect indicators; determine the target cutting constraints and evaluate and assign impacts; screen cutting control parameters through the laser cutting record set; build a cutting quality evaluation function, use the cutting simulation model to perform optimization analysis, and output the optimized cutting parameter threshold; establish a mapping association to achieve optimization guidance for subsequent cutting parameters. This solves the technical problem of the existing cutting process optimization management, which is that the unreasonable setting of cutting parameters leads to a large amount of heat-affected zone and burr accumulation on the cutting surface, causing adhesion between the stator sheet and the lower support, thereby affecting the surface cutting accuracy and quality of the stator sheet, and achieves the technical effect of reasonably setting parameters to improve the cutting accuracy and cutting quality of the stator core. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0015] Figure 1 A schematic diagram of a process flow of a stator core cutting process optimization management method provided in an embodiment of the present application;

[0016] Figure 2 A schematic diagram of a flow chart for screening cutting control parameters in a stator core cutting process optimization management method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0018] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0019] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0020] The present application embodiment provides a method for optimizing the cutting process of a stator core. Figure 1 As shown, the method includes:

[0021] Step S100, read the attribute information of the silicon steel sheet to be cut, and retrieve the laser cutting log with the attribute information as a constraint to obtain a laser cutting record set, wherein the attribute information includes material thickness, surface quality, composition and proportion. By reading the attribute information such as material thickness, surface quality, composition and proportion of the silicon steel sheet to be cut, and taking these attribute information as constraints, intelligently retrieve the previous laser cutting log, obtain the cutting record matching the current silicon steel sheet attribute information, and form a laser cutting record set. Specifically, the material thickness of the silicon steel sheet directly affects the power, speed and focus position required for laser cutting. Silicon steel sheets of different thicknesses require different cutting strategies to ensure the smoothness and verticality of the cutting surface; the surface quality of the silicon steel sheet, such as smoothness, oxidation degree, etc., affects the effect of laser cutting. For example, a silicon steel sheet with more oil or oxide on the surface may require a higher laser power or a special pretreatment step to ensure the cutting quality; the composition and proportion of the silicon steel sheet mainly refer to the chemical composition, especially the proportion of the main metal elements such as silicon and iron and the carbon content, which determine its physical properties and processing characteristics. The proportion of these components will affect the thermal conductivity and melting point of silicon steel sheets, thus affecting the thermal impact and cutting speed during laser cutting. The laser cutting record set may contain key parameters such as laser power, cutting speed, and focus position used in previous successful cutting of similar silicon steel sheets. By referring to these cutting records, the system can automatically adjust the parameters of the cutting equipment to achieve accurate cutting of silicon steel sheets, improve cutting efficiency and quality, and reduce material waste.

[0022] Step S200, identifying cutting defect indicators, wherein the cutting defect indicators include section shape, section angle, section smoothness, burr thickness and heat-affected zone width. Among them, the cutting defect indicators are used to measure the cutting quality and effect of the stator core, including section shape, section angle, section smoothness, burr thickness and heat-affected zone width. Specifically, the section shape refers to the overall outline of the cross section of the silicon steel sheet after cutting. The ideal section shape should be flat and uniform, without obvious unevenness or distortion; the section angle refers to the angle between the cut section and the original plane of the silicon steel sheet. The accuracy of the section angle ensures the geometric size and performance of the stator core. If the section angle is inaccurate, it may cause gaps or misalignments in the stator core during assembly; the section smoothness describes the roughness of the cut surface. A smooth section can reduce mechanical stress and electromagnetic loss, and improve the cutting of the stator core. The working efficiency of the laser cutting machine is affected by the poor smoothness of the cut surface, which may cause stress concentration when the silicon steel sheet is subjected to force, thus affecting its service life. During the laser cutting process, burrs may be formed on the cutting edge due to material melting, gasification and oxide generation. The burr thickness refers to the height or thickness of these protrusions, which not only affects the appearance quality of the silicon steel sheet, but also may affect its assembly and performance in the stator core. The heat affected zone (HAZ) refers to the area where the material properties of the silicon steel sheet change due to high temperature during the laser cutting process. The width of the heat affected zone refers to the size of this area in the direction perpendicular to the cutting edge. An overly wide heat affected zone may cause the mechanical and electromagnetic properties of the silicon steel sheet to decrease, thereby affecting the overall performance of the stator core. By identifying these cutting defect indicators, problems in the stator core cutting process can be discovered in a timely manner and adjusted in a timely manner to ensure the cutting accuracy and quality of the silicon steel sheet.

[0023] Step S300, determine the target cutting constraint, and evaluate the impact of the cutting defect index based on the target cutting constraint, assign a value to the cutting defect index according to the impact evaluation result, and obtain the assigned cutting defect index. The target cutting constraint refers to the requirements for cutting quality in a specific application scenario, which may include the regularity of the cut surface shape, the accuracy of the cut surface angle, the requirements for the smoothness of the cut surface, the limitation of the burr thickness, and the minimization of the width of the heat-affected zone. For example, the cutting speed has a significant impact on the roughness of the cutting surface, the width of the heat-affected zone, and the macroscopic defects of the cut surface. As the cutting speed decreases, the width of the heat-affected zone increases. If the surface quality requirements for the cutting surface are not high and the cutting surface is no longer reprocessed, it is recommended to use the maximum cutting speed. If the surface quality requirements are higher than the requirements for the width of the heat-affected zone, it is recommended to use a cutting speed lower than the maximum cutting speed to cut the silicon steel sheet. Impact assessment refers to analyzing the influence of various cutting defect indicators on silicon steel sheet cutting based on the target cutting constraint. For example, if the target cutting constraint requires the accuracy of the cutting angle, then it is necessary to evaluate the influence of the deviation of the cutting angle on the assembly accuracy of the stator core. Similarly, it is necessary to evaluate the influence of the cutting shape, cutting smoothness, burr thickness and heat-affected zone width defect indicators such as cutting smoothness and burr thickness on the target constraint. After completing the impact assessment, it is necessary to assign values ​​to the cutting defect indicators based on the evaluation results, which means quantifying the influence of each indicator to facilitate subsequent optimization analysis. For example, for the indicator of cutting smoothness, a corresponding value can be assigned according to its influence on the performance of the stator core. The larger the value, the more serious the influence, and the assigned cutting defect indicator is obtained.

[0024] Step S400, based on the single factor analysis principle and the assigned cutting defect index, the cutting control parameters are screened through the laser cutting record set to determine the key cutting parameters and the adaptive cutting threshold, wherein the key cutting parameters include laser power, cutting speed, gas pressure and defocus. Based on the assigned cutting defect index and the single factor analysis principle, the cutting control parameters of the laser cutting record set are screened to determine the key cutting parameters and the adaptive cutting threshold, wherein the single factor analysis principle is a statistical method used to study the influence of a single factor on a certain variable (in this case, the cutting quality). During the laser cutting process, by controlling a single variable (such as laser power, cutting speed, gas pressure, defocus, etc.) and observing its influence on the cutting defect index, the influence of each factor on the cutting quality can be evaluated. According to the assigned cutting defect index, the influence of different cutting parameter combinations on the cutting quality can be quantitatively evaluated. Through the laser cutting record The laser cutting record set is used for parameter screening. The laser cutting record set contains various parameter combinations and their corresponding cutting quality data used in the previous successful cutting of similar silicon steel sheets. The key cutting parameters that have a significant impact on the cutting quality are screened out, including laser power, cutting speed, gas pressure and defocus. Specifically, laser power determines the energy input during the cutting process; cutting speed determines the movement speed of the tool or laser beam during the cutting process; gas pressure mainly affects the gas flow and cooling effect during the cutting process; defocus is the distance between the focus of the laser beam and the surface of the workpiece; key cutting parameters play an important role in the silicon steel sheet cutting process and directly affect the cutting accuracy, quality and efficiency.

[0025] In one possible implementation, Figure 2As shown, step S400 further includes step S410, configuring the parameter tolerance threshold of the cutting control parameter, wherein each cutting control parameter has a unique parameter tolerance threshold. Configuring the parameter tolerance threshold of the cutting control parameter means setting a range of allowable changes for the key control parameters in the laser cutting process to ensure that the cutting quality can still be maintained within an acceptable range when the parameters fluctuate to a certain extent, and each cutting control parameter has its own unique parameter tolerance threshold due to its different effects on the cutting quality. For example, for the parameter of laser power, its tolerance threshold may vary according to the thickness and material of the silicon steel sheet. Thicker silicon steel sheets may require higher laser power to ensure cutting depth and quality, so its power tolerance threshold may be relatively large, and for the cutting speed, its tolerance threshold may be limited by equipment performance and cutting accuracy, and needs to be adjusted within a relatively small range. It also includes step S420, selecting a first cutting control parameter, and taking the first cutting control parameter as a variable, taking other cutting control parameters other than the first cutting control parameter as a quantitative, and using the parameter tolerance threshold to screen the laser cutting record set to determine the first cutting record set. The first cutting control parameter is selected, and the parameter is used as a variable, and other cutting control parameters except the parameter are used as quantitative, and the parameter tolerance threshold is used to screen the laser cutting record set to determine the first cutting record set, which includes the cutting quality data when other cutting control parameters are kept quantitative under different values ​​of the first cutting control parameter, wherein the first cutting control parameter is any one of the cutting control parameters. For example, if the laser power is the first cutting control parameter, it is used as a variable, and other cutting control parameters (such as cutting speed, gas pressure, defocus, etc.) are kept quantitative, and only the influence of the change of the first cutting control parameter on the cutting quality is concerned. It also includes step S430, based on the single factor analysis principle and the first cutting record set, the first cutting control parameter and the assigned cutting defect index are evaluated for correlation, and the key cutting parameters and the adaptive cutting threshold are determined according to the correlation evaluation result. Analyze and evaluate the correlation between the first cutting control parameter and each assigned cutting defect index. According to the correlation evaluation result, we can determine whether the first cutting control parameter is a key cutting parameter. For example, if it has a high correlation with one or some cutting defect indicators and has a significant impact on the cutting quality, then it can be regarded as a key cutting parameter. Based on the correlation evaluation result and the first cutting record set, we can determine the adaptive cutting threshold of the key cutting parameter. The adaptive cutting threshold refers to the maximum or minimum range within which the key cutting parameter can be adjusted while ensuring the cutting quality.

[0026] In a possible implementation, step S430 further includes step S431, configuring the indicator tolerance threshold of the assigned cutting defect indicator, wherein each assigned cutting defect indicator has a unique indicator tolerance threshold. Configuring the indicator tolerance threshold of the assigned cutting defect indicator refers to setting a range of allowable variation for each assigned cutting defect indicator to ensure that the cutting quality can still be maintained within an acceptable range even if certain fluctuations occur during the cutting process. Since different cutting defect indicators may have different degrees of influence and importance on the cutting quality, each assigned cutting defect indicator needs to have its own unique indicator tolerance threshold. It also includes step S432, based on the principle of single factor analysis, selecting the first assigned cutting defect indicator, taking the first assigned cutting defect indicator as a variable, taking other assigned cutting defect indicators other than the first assigned cutting defect indicator as quantitative, and screening the first cutting record set based on the indicator tolerance threshold to obtain a first associated cutting record set. Among them, the first assigned cutting defect indicator is any one of a plurality of assigned cutting defect indicators, and the first associated cutting record set includes other cutting defect indicators quantitatively and related cutting records of silicon steel sheets under different conditions of the first assigned cutting defect indicator. The method further includes step S433, constructing a first trend influence curve based on the first associated cutting record set, performing trend influence analysis on the first cutting control parameter and the first assigned cutting defect index according to the first trend influence curve, and generating a first associated scalar. The first trend influence curve is constructed according to the first associated cutting record set, and the first trend influence curve is the influence curve of the first assigned cutting defect index on the cutting quality, and the trend influence analysis is performed on it to generate a first associated scalar, and the first associated scalar refers to the influence amplitude of the first assigned cutting defect index on the cutting quality. The method further includes step S434, identifying the first trend influence curve according to the predetermined defect index constraint, and determining the first initial adaptive cutting threshold. The obtained first trend influence curve is identified using the predetermined defect index constraint to determine the first initial adaptive cutting threshold. The method further includes step S435, arranging the cutting control parameters from large to small according to the associated scalar, generating a first cutting parameter sequence, and obtaining a first initial adaptive cutting threshold sequence of the first cutting parameter sequence, and the first cutting parameter sequence and the first cutting defect index correspond one to one. The method further includes step S436, determining the key cutting parameters and the adaptive cutting threshold based on the first cutting parameter sequence and the first initial adaptive cutting threshold sequence. In the process of laser cutting process optimization, by analyzing the first cutting parameter sequence (i.e. a series of values ​​for the first cutting control parameter) and the first initial adaptive cutting threshold sequence (i.e. the acceptable range initially set for the parameter), it is finally determined which cutting parameters are critical and the adaptive cutting thresholds that should be used for these critical parameters in actual operations.

[0027] In a possible implementation, step S436 further includes step S436-1, analyzing and acquiring multiple cutting parameter sequences in sequence, and extracting cutting parameters from the multiple cutting parameter sequences according to a preset ranking to generate multiple initial key cutting parameter sets. Analyzing multiple cutting parameter sequences in the laser cutting process in sequence, extracting cutting parameters from each cutting parameter sequence according to a preset ranking to generate multiple initial key cutting parameter sets. It also includes step S436-2, weighting the multiple initial key cutting parameter sets with influence as weight, performing weighted calculation on the same initial key cutting parameters, arranging them from large to small according to the weighted calculation results, and generating an initial key cutting parameter sequence. Weights are assigned to multiple initial key cutting parameter sets according to the influence of each key cutting parameter, and weighted calculation is performed on the same initial key cutting parameters, and sorting is performed according to the weighted calculation results to generate an initial key cutting parameter sequence. It also includes step S436-3, selecting the initial key cutting parameters of the preset ranking in the initial key cutting parameter sequence and setting them as the key cutting parameters. The initial key cutting parameter sequence is analyzed to select the cutting parameter with the highest ranking as the initial key cutting parameter. The method also includes step S436-4, obtaining multiple initial adaptive cutting threshold sets corresponding to the key cutting parameters, and extracting the threshold intersection in the initial adaptive cutting threshold set as the adaptive cutting threshold. Multiple initial adaptive cutting threshold sets corresponding to the key cutting parameters are obtained, and the intersection of all initial adaptive cutting threshold sets is set as the adaptive cutting threshold, that is, the value range of the key cutting parameters that can meet the cutting quality requirements under all conditions.

[0028] Step S500, building a cutting quality evaluation function based on the assigned cutting defect index, using a laser cutting simulation model to optimize and analyze the key cutting parameters, outputting an optimized cutting parameter threshold, and selecting the optimal cutting parameter of the optimized cutting parameter threshold to perform the cutting process of the silicon steel sheet to be cut. By quantifying and assigning values ​​to cutting defect indicators (section shape, section angle, section smoothness, burr thickness, and heat-affected zone width, etc.), and constructing corresponding evaluation functions, in order to objectively and accurately evaluate the cutting quality, the laser cutting simulation model is used to optimize and analyze the key cutting parameters. According to the results of the simulation analysis, the optimized cutting parameter threshold is output. The optimized cutting parameter threshold represents the reasonable range of key cutting parameters under different conditions to ensure the stability and efficiency of the cutting quality. Specifically, the laser cutting simulation model can simulate the real cutting process and predict the cutting effect under different parameter combinations. By simulating and analyzing the key cutting parameters (such as laser power, cutting speed, focal length, etc.), we can find the influence of these parameters on the cutting quality, so as to determine the optimal cutting parameter value range. The simulation results are applied to actual cutting, and the optimal cutting parameters in the optimized cutting parameter threshold are selected to perform the cutting process of the silicon steel sheet to be cut. By precisely controlling the key cutting parameters, efficient and high-quality cutting of silicon steel sheets can be achieved.

[0029] In a possible implementation, step S500 further includes step S510, building a cutting quality evaluation function based on the assigned cutting defect index, wherein the cutting quality evaluation result is the reciprocal of the sum of the assigned cutting defect evaluation results. The cutting quality evaluation function is built based on the assigned cutting defect index, and the cutting quality evaluation function combines multiple cutting defect indicators to obtain a comprehensive cutting quality evaluation result, and uses a single value to quantify the quality of the cutting. The cutting quality evaluation result is the reciprocal of the sum of the assigned cutting defect evaluation results, that is, the assigned values ​​of each cutting defect indicator are added, and then the reciprocal of this sum is taken as the final cutting quality evaluation result. When the sum of the assigned cutting defect evaluation results increases (that is, the cutting quality deteriorates), the value of the cutting quality evaluation result will decrease rapidly, so that it can more sensitively reflect the change in cutting quality. It also includes step S520, using the adaptive cutting threshold to build a parameter optimization space, and randomly generating multiple cutting parameter combinations based on the parameter optimization space. According to the determined adaptive cutting threshold, a parameter optimization space is built. This parameter optimization space is a multi-dimensional range, and its dimension corresponds to the number of key cutting parameters. The range on each dimension is determined by the adaptive cutting threshold, that is, the value of the key cutting parameter on this dimension is within the reasonable range of the adaptive cutting threshold. Based on the constructed parameter optimization space, multiple cutting parameter combinations are randomly generated, and each combination represents a set of possible cutting parameter values, ensuring that different areas in the parameter optimization space are covered, thereby increasing the possibility of finding the optimal parameter combination. It also includes step S530, based on the digital twin technology, combining the attribute information of the silicon steel sheet to be cut and the information combination of the laser cutting equipment to build the laser cutting simulation model. Using the new technology of digital twin, the physical properties of the silicon steel sheet to be cut and the performance parameters of the laser cutting equipment are mapped and integrated in the virtual space, so as to build a highly realistic simulation model, simulate and predict the cutting process, and quickly iterate and optimize various parameters in the cutting process to improve cutting quality and efficiency. The method further includes step S540, wherein the key cutting parameters are optimized and analyzed based on the cutting quality evaluation function, the adaptive cutting threshold, the multiple cutting parameter combinations and the laser cutting simulation model to obtain the optimized cutting parameter threshold. Using the laser cutting simulation model, we can quickly evaluate the cutting effect under different cutting parameter combinations, and find the cutting parameter combination that can make the cutting quality evaluation function optimal (i.e., the best cutting quality) by comparing and analyzing the simulation results and cutting quality evaluation results under different cutting parameter combinations, and use the key cutting parameter value range in this cutting parameter combination as the optimized cutting parameter threshold.

[0030] In a possible implementation, step S540 further includes step S541, based on the target cutting path, through the laser cutting simulation model, performing simulated cutting processing of the multiple cutting parameter combinations, and generating multiple simulated processing results. The target cutting path is set based on the shape, size and processing requirements of the silicon steel sheet to be cut, including taking into account the starting point, end point and possible changes in the cutting process, using the laser cutting simulation model to simulate the cutting process, and inputting the previously randomly generated multiple cutting parameter combinations into the simulation model in turn to generate corresponding simulated processing results, wherein each cutting parameter combination represents a possible set of laser power, cutting speed, gas pressure and defocus. It also includes step S542, using the cutting quality evaluation function to evaluate the multiple simulated processing results and generate a cutting quality evaluation coefficient sequence. After completing the simulated cutting process and obtaining the corresponding simulated processing results, the previously constructed cutting quality evaluation function is further used to perform a quantitative quality evaluation on these simulated processing results. Specifically, according to the cutting quality evaluation function, the cutting quality evaluation coefficient corresponding to each simulated processing result is calculated. This coefficient is a single value, which integrates the influence of multiple cutting defect indicators and can reflect the overall quality level of the simulated processing results. The above evaluation process is performed on all simulated processing results to obtain a cutting quality evaluation coefficient sequence. Each element in this sequence corresponds to a simulated processing result and its corresponding cutting quality evaluation coefficient, which are arranged in the order of simulated processing or the order of parameter combination. It also includes step S543, setting the multiple cutting parameter combinations as multiple initial solutions, mapping and arranging the multiple initial solutions based on the cutting quality evaluation coefficient sequence, and obtaining an initial solution sequence. Using the generated cutting quality evaluation coefficient sequence, we can evaluate the quality of each initial solution and sort the initial solutions according to the cutting quality evaluation coefficient. The initial solutions with higher cutting quality evaluation coefficients are positioned higher in the sequence, which means that these cutting parameter combinations are more likely to produce high-quality cutting results. Finally, the initial solution sequence is obtained, that is, a list of cutting parameter combinations arranged according to the cutting quality evaluation coefficient. It also includes step S544, taking the adaptive cutting threshold as the constraint space, optimizing and analyzing the key cutting parameters based on the initial solution sequence, and obtaining the optimized cutting parameter threshold. Based on the obtained initial solution sequence, with the adaptive cutting threshold as the constraint, optimizing and analyzing each key cutting parameter, and obtaining the optimized cutting parameter threshold according to the optimization analysis result.

[0031] In a possible implementation, step S544 further includes step S544-1, extracting the first N initial solutions of the initial solution sequence and setting them as the first solutions, and setting the other initial solutions except the first solution as the tail solutions, where N is an integer greater than 10. Among them, the first solution is the N solutions with the highest cutting quality evaluation coefficients in the initial solution sequence, which represent the cutting parameter combination that is most likely to produce high-quality cutting results; the tail solutions are all other solutions in the initial solution sequence except the first solution. Although they are slightly inferior to the first solution in cutting quality, they still have a certain reference value. For example, in the subsequent optimization process, the search strategy can be adjusted or new optimization directions can be explored according to the performance of the tail solutions. It also includes step S544-2, taking the first solution as the leader, clustering the tail solutions to obtain N cluster neighborhoods, where the clustering principle is random clustering, and the number of tail solutions in each cluster neighborhood is the same. Taking the first solution as the cluster center, cluster analysis is performed on the tail solutions to generate N cluster groups, which are cluster neighborhoods. It also includes step S544-3, configuring the optimization adjustment step of the key cutting parameters, wherein each key cutting parameter has a unique optimization adjustment step. During the laser cutting process, for each key cutting parameter (such as laser power, cutting speed, focal length, etc.), a unique optimization adjustment step is set for it, and the optimization adjustment step determines the size or increment of each adjustment of the parameter during the optimization process. It also includes step S544-4, adjusting the tail solutions in the N clustering neighborhoods based on the optimization adjustment step, evaluating the adjusted tail solutions through the cutting quality evaluation function, and obtaining multiple tail solution cutting quality evaluation coefficients. The cutting quality evaluation function is used to evaluate the cutting quality according to various indicators in the cutting process, and a cutting quality evaluation coefficient is output, which is a quantitative representation of the effect of the adjusted tail solution on the cutting quality. It also includes step S544-5, judging whether the tail solution cutting quality evaluation coefficient is greater than the cutting quality evaluation coefficient of the first solution in the same clustering neighborhood. If so, the first solution is updated using the adjusted tail solution to obtain multiple updated neighborhoods. Compare the cutting quality evaluation coefficient of the tail solution with the cutting quality evaluation coefficient of the first solution in the unified clustering neighborhood. If it is greater than, use the adjusted tail solution to update the first solution to obtain an updated neighborhood. It also includes step S544-6, continuing to iterate and optimize until the predetermined number of constraints are met, outputting multiple optimized neighborhoods, and calculating the average of multiple cutting quality evaluation coefficients of the multiple optimized neighborhoods. Among them, the multiple optimized neighborhoods are the updated neighborhoods obtained by updating the first solution with the adjusted tail solution, and calculating the average of the cutting quality evaluation coefficients of the multiple optimized neighborhoods. It also includes step S544-7, selecting the optimized neighborhood corresponding to the maximum cutting quality evaluation coefficient average as the optimal neighborhood, and setting the optimized cutting parameter threshold based on the optimal neighborhood.By comparing the mean values ​​of the evaluation coefficients of different neighborhoods, we can preliminarily determine which neighborhood's solution is more likely to produce high-quality cutting results, and select the neighborhood with the largest mean value of the evaluation coefficient as the optimal neighborhood. According to the characteristics and distribution of the solutions in the optimal neighborhood, we can determine the appropriate value range of the key cutting parameters and set it as the optimized cutting parameter threshold.

[0032] Step S600, establish a mapping association between the attribute information, the target cutting constraint and the optimized cutting parameter threshold, build a cutting parameter matching library based on the mapping association, and implement subsequent cutting parameter optimization guidance through the cutting parameter matching library. Establish a mapping association between the attribute information, the target cutting constraint and the optimized cutting parameter threshold, and build a cutting parameter matching library based on these mapping associations. Specifically, these information are mapped and associated with each other to form a mapping relationship, that is, different attribute information and target cutting constraints correspond to different optimized cutting parameter thresholds. Using the established mapping association, various possible attribute information combinations, target cutting constraints and corresponding optimized cutting parameter thresholds are matched and stored to form a cutting parameter matching library. When facing a new cutting task, you can query this matching library to quickly find the cutting parameter threshold that is suitable for the current task attribute information and target cutting constraints, and implement subsequent cutting parameter optimization guidance to ensure cutting quality and cutting accuracy.

[0033] In a possible implementation, step S600 further includes step S610, configuring the first error interval and the second error interval of the attribute information and the target cutting constraint. In the laser cutting process, in order to cope with various uncertainties and changes that may occur in actual operation, configuring a certain allowable error range for the attribute information and the target cutting constraint helps to improve the flexibility and adaptability of the cutting parameters, so that the cutting process can be automatically adjusted and optimized within a certain range, wherein the error interval of the attribute information is called the first error interval, and the error interval of the target cutting constraint is called the second error interval. It also includes step S620, based on the first error interval and the second error interval, the attribute information and the target cutting constraint are expanded to generate expanded attribute information and expanded cutting constraints. The attribute information and the target cutting constraint are expanded according to the first error interval and the second error interval, for example, the thickness of the material in the original attribute is 5mm, and the first error interval is ±0.1mm, then the expanded attribute information includes 4.9mm~5.1mm. When facing a specific cutting task, we can find a suitable combination of cutting parameters in a larger range according to the expanded attribute information and cutting constraints, thereby improving the flexibility and adaptability of the cutting process. It also includes step S630, based on the principle of decision tree, taking the extended attribute information as the first decision node, taking the extended cutting constraint as the second decision node, taking the optimized cutting parameter threshold as the associated child node of the first decision node and the second decision node, to construct the cutting parameter matching library. Based on the principle of decision tree, taking the extended attribute information as the first decision node, taking the extended cutting constraint as the second decision node, taking the optimized cutting parameter threshold as the associated child node of the two decision nodes, to construct the cutting parameter matching library, specifically, the first decision node: extended attribute information, the second decision node: extended cutting constraint, for each combination determined by the first decision node and the second decision node, the corresponding optimized cutting parameter threshold is associated as a child node, according to the relationship between the above decision nodes and child nodes, a complete decision tree structure is constructed, reflecting the mapping relationship from attribute information and cutting constraints to cutting parameter thresholds, and finally forming a cutting parameter matching library, when facing a new cutting task, this matching library can be queried, that is, traversing the decision tree, according to the specific attribute information and cutting constraints, to quickly find a suitable cutting parameter threshold.

[0034] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A method for optimizing the cutting process of a stator core, characterized in that: The method comprises: Reading the attribute information of the silicon steel sheet to be cut, and using the attribute information as a constraint, searching the laser cutting log to obtain a laser cutting record set, wherein the attribute information includes material thickness, surface quality, composition and proportion; Identify cutting defect indicators, including cut surface shape, cut surface angle, cut surface smoothness, burr thickness, and heat-affected zone width; Determine a target cutting constraint, and perform an impact assessment on the cutting defect index based on the target cutting constraint, and assign a value to the cutting defect index according to the impact assessment result to obtain an assigned cutting defect index; Based on the single factor analysis principle and the assigned cutting defect index, the cutting control parameters are screened through the laser cutting record set to determine the key cutting parameters and the adaptive cutting threshold, wherein the key cutting parameters include laser power, cutting speed, gas pressure and defocusing amount; A cutting quality evaluation function is constructed based on the assigned cutting defect index, and the key cutting parameters are optimized and analyzed using a laser cutting simulation model, an optimized cutting parameter threshold is output, and the optimal cutting parameter of the optimized cutting parameter threshold is selected to perform the cutting process of the silicon steel sheet to be cut; Establishing a mapping association among the attribute information, the target cutting constraint, and the optimized cutting parameter threshold, building a cutting parameter matching library based on the mapping association, and implementing subsequent cutting parameter optimization guidance through the cutting parameter matching library; Among them, based on the single factor analysis principle and the assigned cutting defect index, the cutting control parameters are screened through the laser cutting record set to determine the key cutting parameters and the adaptive cutting threshold, including: configuring a parameter tolerance threshold of a cutting control parameter, wherein each cutting control parameter has a unique parameter tolerance threshold; Selecting a first cutting control parameter, taking the first cutting control parameter as a variable, taking other cutting control parameters except the first cutting control parameter as a quantity, and using the parameter tolerance threshold to screen the laser cutting record set to determine a first cutting record set; Based on the single factor analysis principle and the first cutting record set, a correlation evaluation is performed on the first cutting control parameter and the assigned cutting defect index, and the key cutting parameter and the adaptive cutting threshold are determined according to the correlation evaluation result; Among them, a cutting quality evaluation function is built based on the assigned cutting defect index, and the key cutting parameters are optimized and analyzed using a laser cutting simulation model to output the optimized cutting parameter thresholds, including: Building a cutting quality evaluation function based on the assigned cutting defect index, wherein the cutting quality evaluation result is the reciprocal of the sum of the assigned cutting defect evaluation results; Using the adaptive cutting threshold to build a parameter optimization space, and randomly generating multiple cutting parameter combinations based on the parameter optimization space; Based on the digital twin technology, the laser cutting simulation model is constructed by combining the property information of the silicon steel sheet to be cut and the information of the laser cutting equipment; Based on the cutting quality evaluation function, the adaptive cutting threshold, the multiple cutting parameter combinations and the laser cutting simulation model, the key cutting parameters are optimized and analyzed to obtain the optimized cutting parameter threshold; The key cutting parameters are optimized and analyzed based on the cutting quality evaluation function, the adaptive cutting threshold, the multiple cutting parameter combinations and the laser cutting simulation model, including: Based on the target cutting path, the laser cutting simulation model is used to perform simulated cutting processing of the plurality of cutting parameter combinations to generate a plurality of simulated processing results; Using the cutting quality evaluation function to evaluate the plurality of simulation processing results, and generating a cutting quality evaluation coefficient sequence; The plurality of cutting parameter combinations are set as a plurality of initial solutions, and the plurality of initial solutions are mapped and arranged based on the cutting quality evaluation coefficient sequence to obtain an initial solution sequence; Taking the adapted cutting threshold as the constraint space, optimizing and analyzing the key cutting parameters based on the initial solution sequence to obtain the optimized cutting parameter threshold; The adaptive cutting threshold is used as the constraint space, and the key cutting parameters are optimized and analyzed based on the initial solution sequence to obtain the optimized cutting parameter threshold, including: Extract the first N initial solutions of the initial solution sequence and set them as the first solution, and set the other initial solutions except the first solution as the last solution, where N is an integer greater than 10; Taking the first solution as the leader, cluster the tail solutions to obtain N cluster neighborhoods, where the clustering principle is random clustering, and the number of tail solutions in each cluster neighborhood is the same; Configuring an optimization adjustment step of the key cutting parameters, wherein each key cutting parameter has a unique optimization adjustment step; Based on the optimization adjustment step, the tail solutions in the N clustering neighborhoods are adjusted once, and the adjusted tail solutions are evaluated by the cutting quality evaluation function to obtain a plurality of tail solution cutting quality evaluation coefficients; Determine whether the cutting quality evaluation coefficient of the tail solution is greater than the cutting quality evaluation coefficient of the first solution in the same clustering neighborhood. If so, use the adjusted tail solution to update the first solution to obtain multiple updated neighborhoods. Continue iterating and optimizing until a predetermined number of constraints are met, output a plurality of optimized neighborhoods, and calculate and obtain the average values ​​of a plurality of cutting quality evaluation coefficients of the plurality of optimized neighborhoods; The optimized neighborhood corresponding to the maximum cutting quality evaluation coefficient mean is selected as the optimal neighborhood, and the optimized cutting parameter threshold is set based on the optimal neighborhood.

2. The method for optimizing the cutting process of a stator core according to claim 1, characterized in that: Determining the key cutting parameters and the adaptive cutting threshold according to the correlation evaluation result includes: Configuring an indicator tolerance threshold of the valued cutting defect indicator, wherein each valued cutting defect indicator has a unique indicator tolerance threshold; Based on the single factor analysis principle, a first valued cutting defect indicator is selected, the first valued cutting defect indicator is used as a variable, other valued cutting defect indicators except the first valued cutting defect indicator are used as quantitative, and the first cutting record set is screened based on the indicator tolerance threshold to obtain a first associated cutting record set; constructing a first trend influence curve based on the first associated cutting record set, performing trend influence analysis on the first cutting control parameter and the first assigned cutting defect index according to the first trend influence curve, and generating a first associated scalar; Identify the first trend influence curve according to a predetermined defect index constraint, and determine a first initial adaptation cutting threshold; Arrange the cutting control parameters from large to small according to the associated scalar to generate a first cutting parameter sequence, and obtain a first initial adaptive cutting threshold sequence of the first cutting parameter sequence, wherein the first cutting parameter sequence corresponds to the first cutting defect index one by one; The key cutting parameters and the adapted cutting threshold are determined based on the first cutting parameter sequence and the first initial adapted cutting threshold sequence.

3. The method for optimizing the cutting process of a stator core according to claim 2, characterized in that: Determining the key cutting parameter and the adaptive cutting threshold based on the first cutting parameter sequence and the first initial adaptive cutting threshold sequence includes: Analyze and obtain multiple cutting parameter sequences in sequence, and extract cutting parameters from the multiple cutting parameter sequences according to a preset ranking to generate multiple initial key cutting parameter sets; The plurality of initial key cutting parameter sets are weighted by taking the influence as the weight, and the same initial key cutting parameters are weightedly calculated, and the initial key cutting parameter sequences are generated by arranging the parameters from large to small according to the weighted calculation results; Selecting the initial key cutting parameter of the preset ranking in the initial key cutting parameter sequence and setting it as the key cutting parameter; A plurality of initial adaptive cutting threshold sets corresponding to the key cutting parameters are obtained, and a threshold intersection in the initial adaptive cutting threshold sets is extracted and set as the adaptive cutting threshold.

4. The method for optimizing the cutting process of a stator core according to claim 1, characterized in that: A cutting parameter matching library is formed based on the mapping association, including: configuring the attribute information and a first error interval and a second error interval of the target cutting constraint; Expanding the attribute information and the target cutting constraint based on the first error interval and the second error interval to generate expanded attribute information and expanded cutting constraint; Based on the decision tree principle, the cutting parameter matching library is constructed by taking the extended attribute information as the first decision node, the extended cutting constraint as the second decision node, and the optimized cutting parameter threshold as the associated sub-node of the first decision node and the second decision node.

Images