A grinding process optimization method and device
By constructing a grinding process evaluation model and a gray correlation model, and optimizing the grinding process parameters, the consistency and efficiency of grinding process in aircraft engine blade manufacturing are solved, and higher quality surface treatment effect and processing efficiency are achieved.
Patent Information
- Application Number
- CN202411254311.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-09
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-09-09
AI Technical Summary
The existing grinding process has problems such as poor surface consistency, low processing efficiency and difficulty in achieving closed-loop control of processing and measurement in the manufacturing of aircraft engine blades, which is difficult to meet the needs of modern automation development.
By constructing a grinding process evaluation model and a gray correlation model, the weight matrix and limit matrix of the process target are determined, and combined with expert experience and experimental data, the grinding process parameters are optimized to improve surface quality and processing efficiency.
It achieves a higher quality surface treatment effect, improves the efficiency and stability of the grinding process, reduces the influence of artificial variables, and meets high-standard production requirements.
Smart Images

Figure CN118761338B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of surface treatment technology, and in particular to a grinding process optimization method and device. Background Art
[0002] Aircraft engine blades are core components of the engine. They are typical complex curved surfaces with complex profiles, weak structural rigidity, thin blade edges, and strict surface roughness requirements. Their contour accuracy and surface quality directly affect the aerodynamic performance and service life of aircraft engines. Grinding, as the final key process in the manufacture of aircraft engine blades, greatly determines the surface quality of the blades. Therefore, ensuring the grinding quality of aircraft engine blades is a key link in the development of the aviation industry.
[0003] The current manual grinding process has problems such as poor surface consistency and low processing efficiency, which makes it difficult to accurately control the removal of excess and contact force. CNC machine tool grinding can improve processing accuracy, but its versatility is poor and it is difficult to achieve closed-loop control of processing and measurement. Both methods cannot meet the needs of modern automation development.
[0004] Robots have the advantages of high processing efficiency, good flexibility, and easy combination with a variety of external sensors to improve processing quality. Combined with the grinding process, they can achieve high-precision grinding processing. They have been gradually applied to the precision grinding of aircraft engine blades. This breakthrough effectively reduces the influence of human variables in the grinding process and significantly improves the stability and efficiency of the processing. Since the robot grinding of aircraft engine blades is a material removal process to reduce the surface roughness, there are multiple mutually coupled weak rigid systems in the grinding process, and there are many variables affecting the grinding process. Therefore, how to select the parameters in the grinding process and make quantitative recommendations to ensure that the surface quality after grinding meets the requirements has become the focus of current research. Summary of the invention
[0005] In view of this, the present application provides a grinding process optimization method and device to effectively improve the grinding effect and grinding quality.
[0006] Specifically, the present application is implemented through the following technical solutions:
[0007] A first aspect of the present application provides a grinding process optimization method, the method comprising:
[0008] Constructing a polishing process evaluation model including the mutual influence between subjective factors and polishing process parameters; the polishing process evaluation model includes a network layer and a control layer, the network layer includes a number of the polishing process parameters, and the control layer includes a number of process targets; the polishing process parameters in the network layer influence each other; the process targets in the control layer are independent of each other, and any process target is associated with at least one process parameter;
[0009] Constructing a grey correlation model including the grinding process parameters and the response variables; the grey correlation model is used to determine the grey correlation degree between the grinding process parameters and the response variables;
[0010] Determine a weight matrix and a limit matrix of the process target based on the polishing process evaluation model;
[0011] Introducing the weight matrix and the limit matrix into the grey correlation degree determined by the grey correlation model to determine the correlation mapping relationship;
[0012] Based on the association mapping relationship, a parameter combination of the polishing process is determined, and the polishing process is optimized based on the parameter combination of the polishing process.
[0013] The second aspect of the present application provides a polishing process optimization device, which includes a construction module, a determination module and an optimization module; wherein:
[0014] The construction module is used to construct a polishing process evaluation model including the mutual influence between subjective factors and polishing process parameters; the polishing process evaluation model includes a network layer and a control layer, the network layer includes a number of the polishing process parameters, and the control layer includes a number of process targets; the polishing process parameters in the network layer influence each other; the process targets in the control layer are independent of each other, and any process target is associated with at least one process parameter;
[0015] The construction module is further used to construct a grey correlation model including the polishing process parameters and the response variables; the grey correlation model is used to determine the grey correlation degree between the polishing process parameters and the response variables;
[0016] The determination module is used to determine the weight matrix and limit matrix of the process target based on the polishing process evaluation model;
[0017] The determination module is further used to introduce the weight matrix and the limit matrix into the grey correlation degree determined by the grey correlation model to determine the correlation mapping relationship;
[0018] The optimization module is used to determine a parameter combination of the polishing process based on the association mapping relationship, and optimize the polishing process based on the parameter combination of the polishing process.
[0019] The third aspect of the present application provides a polishing process optimization device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of any one of the methods provided in the first aspect of the present application are implemented.
[0020] The fourth aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any one of the methods provided in the first aspect of the present application.
[0021] The polishing process optimization method and device provided in the present application first construct a polishing process evaluation model including the mutual influence between subjective factors and polishing process parameters, then construct a gray correlation model including polishing process parameters and response variables, and then determine the weight matrix and limit matrix of the process target based on the polishing process evaluation model, and then introduce the weight matrix and limit matrix into the gray correlation degree determined by the gray correlation model to determine the correlation mapping relationship, and finally determine the parameter combination of the polishing process based on the correlation mapping relationship, and optimize the polishing process based on the parameter combination of the polishing process. In this way, the advice and experience of experts are integrated through the polishing process evaluation model to deal with the complex problems of subjective judgment, and the influence of several polishing process parameters on the response variables is analyzed through the gray correlation model. Furthermore, through the combination of the polishing process evaluation model and the gray correlation model, the optimization of process parameters is not only based on the experience and judgment of experts, but also based on actual experimental data, ensuring the comprehensiveness and scientificity of the optimization process. Furthermore, by calculating the correlation mapping relationship through the grinding process evaluation model and the grey correlation model, it can be ensured that the optimized grinding process parameters can effectively meet the requirements of the process objectives, thereby providing a higher quality surface treatment effect and improving the grinding effect of the grinding process. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] Figure 1 A schematic diagram of the first embodiment of the application scenario of the polishing process optimization method provided in this application;
[0023] Figure 2 A flow chart of Embodiment 1 of the polishing process optimization method provided in this application;
[0024] Figure 3 A schematic diagram of a first embodiment of a polishing process evaluation model provided in the present application;
[0025] Figure 4 A flow chart of Embodiment 2 of the polishing process optimization method provided in this application;
[0026] Figure 5 A flowchart of Embodiment 3 of the polishing process optimization method provided in this application;
[0027] Figure 6 A schematic diagram of a first embodiment of the polishing process provided in this application;
[0028] Figure 7 A schematic diagram of the calculation results of the polishing process optimization method provided in this application in Example 1;
[0029] Figure 8 A hardware structure diagram of the grinding process optimization equipment where the grinding process optimization device of this application is located;
[0030] Fig. 9 This is a structural schematic diagram of Example 1 of the polishing process optimization device provided in this application. DETAILED DESCRIPTION
[0031] Here, exemplary embodiments are described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application.
[0032] The terms used in this application are only for the purpose of describing specific embodiments and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in this article refers to and includes any or all possible combinations of one or more associated listed items.
[0033] It should be understood that although the terms first, second, third, etc. may be used in the present application to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".
[0034] The present application provides a grinding process optimization method and device for effectively improving the grinding effect and grinding quality.
[0035] Specific embodiments are given below to introduce the technical solution of the present application in detail.
[0036] The grinding process optimization method provided in this application can be applied to special grinding fields with characteristics of complex profiles, weak structural rigidity, thin blade edges, and strict surface roughness requirements. For example, it can be applied to the grinding field of aircraft engine blades. Figure 1 This is a schematic diagram of the first embodiment of the application scenario of the polishing process optimization method provided in this application. Figure 1 ,In the blade grinding process, material removal is achieved through the flexible contact between the grinding wheel and the workpiece surface. From a macroscopic point of view, grinding with a grinding wheel is a high-speed micro-cutting process with cumulative grinding results.
[0037] Furthermore, in the grinding process of the grinding wheel, the grinding process parameters refer to the key operating variables that can affect the grinding effect and efficiency during the grinding process. The setting of the grinding process parameters directly determines the grinding efficiency and the final grinding effect. It should be noted that the specific parameters included in the grinding process parameters are set according to actual needs, and this is not limited in this embodiment. For example, in one embodiment, the grinding process parameters include feed speed, grinding positive pressure and grinding wheel speed.
[0038] For further information, please refer to Figure 1 The feed speed Vf determines the speed at which the grinding wheel moves on the blade surface. The feed speed directly affects the speed at which the abrasive removes material from the workpiece surface, and thus affects the depth of material removal. The grinding positive pressure Fn determines the contact strength between the grinding wheel and the blade surface, which directly affects the grinding effect and the wear of the tool head. Appropriate positive pressure can make the grinding wheel fit tightly against the blade surface and improve the grinding effect. The grinding wheel speed Vg determines the contact frequency and friction heat between the grinding wheel and the workpiece surface. A reasonable grinding wheel speed can improve the grinding efficiency, while reducing heat accumulation and preventing deformation of the grinding wheel or burns on the blade surface due to overheating. Therefore, the setting of the grinding process parameter values directly determines the grinding efficiency and the final grinding effect.
[0039] Figure 2 This is a flow chart of Example 1 of the polishing process optimization method provided in this application. Please refer to Figure 2 , the method provided in this embodiment may include:
[0040] S201. Construct a polishing process evaluation model that includes the mutual influence between subjective factors and polishing process parameters; the polishing process evaluation model includes a network layer and a control layer, the network layer includes a number of the polishing process parameters, and the control layer includes a number of process targets; the several polishing process parameters in the network layer influence each other; the several process targets in the control layer are independent of each other, and any process target is associated with at least one process parameter.
[0041] Specifically, the subjective factor is the expert's evaluation of the grinding process parameters based on knowledge and experience during the grinding process. The subjective factor can quickly and flexibly give accurate judgments during the grinding process.
[0042] Furthermore, the grinding process evaluation model fully considers the mutual influence between and within the grinding process parameters. Through the grinding process evaluation model, the influence of the grinding process parameters on the process objectives can be quantitatively analyzed.
[0043] It should be noted that the process target is an indicator for evaluating the grinding effect. The specific content of the process target is determined according to actual needs, and is not limited in this embodiment. For example, in one embodiment, the process target is to obtain the optimal grinding process parameter combination.
[0044] Optionally, the process goal is to optimize the process combination; the control layer also includes a number of process decision criteria, and the process decision criteria include at least minimum surface roughness and maximum material removal rate; and the grinding process parameters include at least feed speed, grinding pressure and grinding wheel speed.
[0045] Furthermore, Figure 3 This is a schematic diagram of the first embodiment of the polishing process evaluation model provided in this application. Figure 3 The polishing process evaluation model provided in this embodiment has a control layer including a target layer and a decision layer. The target layer includes several process targets. These process targets are located in the target layer and are independent of each other.
[0046] For further information, please refer to Figure 3 , the decision layer contains the polishing decision criteria, all decision criteria are independent of each other and are only controlled by the goal. It should be noted that there may be no decision criteria in the control layer, but it contains at least one process goal.
[0047] In a specific implementation, for example, in one embodiment, continue to refer to Figure 3 The decision criteria include surface roughness (SR) and material removal rate (MRR). The material removal rate indicates the volume of material removed from the workpiece surface per unit time. A higher material removal rate indicates higher grinding efficiency. Improving the material removal rate can reduce processing time and production costs. Surface roughness characterizes the characteristics of the micro-geometric shape of the workpiece surface after grinding, reflecting the degree of surface waviness and unevenness. A smaller surface roughness indicates a higher surface finish.
[0048] For further information, please refer to Figure 3,The grinding process parameters in the network layer can include feed speed, grinding pressure and grinding wheel speed, and the process objectives of the control layer are achieved through the grinding process parameters. ,It should be explained that the grinding process parameters in the network layer are interdependent and mutually ,dominated, and the grinding process parameters and the layers are not independent, and the network layer is an interdependent ,feedback network structure.
[0049] The polishing process evaluation model provided in this embodiment provides scientific decision support by quantitatively analyzing the polishing process parameters and combining expert opinions, and reduces the trial and error cost by considering the subjective aspects of the polishing process. Through the polishing process evaluation model, the polishing process can be effectively optimized to meet high-standard production requirements and achieve quality polishing.
[0050] S202, constructing a grey correlation model including the polishing process parameters and the response variables; the grey correlation model is used to determine the grey correlation degree between the polishing process parameters and the response variables.
[0051] Specifically, the response variable is the result after grinding, and the response variable is directly affected by the grinding process parameters. The process target can be reflected through the response variable. It should be noted that the specific elements contained in the response variable are determined according to actual needs, and this is not limited in this embodiment. For example, in one embodiment, in combination with the above example, the material removal rate and surface roughness can be determined as the response variables.
[0052] Furthermore, the grey correlation degree is used to measure the closeness between the grinding process parameters and the response variables, and the degree of mutual influence is determined by analyzing the change trend between the two.
[0053] S203: Determine a weight matrix and a limit matrix of the process target based on the polishing process evaluation model.
[0054] Specifically, the importance of each grinding process parameter can be determined by the corresponding values of the grinding process parameters in the weight matrix. For example, in one embodiment, the weight matrix includes the weight of the feed speed and the weight of the grinding pressure, and the importance or priority of the feed speed and the grinding pressure can be determined by this weight matrix.
[0055] Furthermore, the limit matrix is used to determine the impact of each grinding process parameter on the grinding target under extreme conditions. The limit matrix can be used to determine the performance of each grinding process parameter under the most favorable or most unfavorable conditions. It should be noted that the limit matrix can be used to formulate a more reasonable optimization strategy based on the performance under extreme conditions to ensure that the process can operate stably under various conditions.
[0056] S204: Introduce the weight matrix and the limit matrix into the grey correlation degree determined by the grey correlation model to determine a correlation mapping relationship.
[0057] Specifically, the association mapping relationship is used to characterize the influence of each grinding process parameter on the response variable. In other words, the association mapping relationship is used to evaluate the degree of change of each grinding process parameter on the response variable to determine the value of each grinding process parameter.
[0058] When this step is implemented specifically, the weight matrix and the limit matrix are input into the grey correlation model, the grey correlation degree is output, and the correlation mapping relationship corresponding to each grinding process parameter is determined through the grey correlation degree.
[0059] The following is a specific implementation method to introduce in detail the process of determining the association mapping relationship:
[0060] (1) Calculating a plurality of parameter weights corresponding to a plurality of the grinding process parameters based on the weight matrix and the limit matrix.
[0061] Specifically, the weight matrix and the limit matrix contain weights corresponding to various grinding process parameters, and these parameters are extracted as parameter weights.
[0062] (2) Introducing the parameter weight into the grey correlation model, and screening out the target parameter combination with the smallest grinding process distance from the response variable.
[0063] In specific implementation, in this step, the grey correlation degree corresponding to the parameter weight processing is performed to obtain the optimal correlation degree of each polishing process parameter.
[0064] The parameter combination of the grinding process with the optimal correlation is determined as the target parameter combination.
[0065] It should be noted that the calculated parameter weights are introduced into the grey correlation model, and the influence of each grinding process parameter combination on the response variable is evaluated by calculating the grey correlation degree. After sorting from large to small, the grinding process parameter combination with the largest value is screened out and determined as the optimal correlation degree, and then the corresponding target parameter combination is obtained.
[0066] (3) Determine the target parameter combination as the association mapping relationship.
[0067] In specific implementation, the closest polishing process parameter combination is determined as the polishing process parameter combination corresponding to the response variable, and an associated mapping relationship is constructed for it.
[0068] The polishing process optimization method provided in this embodiment first calculates several parameter weights corresponding to several polishing process parameters based on the weight matrix and the limit matrix, then introduces the parameter weights into the grey correlation model, selects the parameter combination of the polishing process closest to the response variable, and finally determines the parameter combination of the polishing process closest to the response variable as a correlation mapping relationship. In this way, the optimal parameter combination of the polishing process is determined based on the correlation degree of the polishing process parameters, which can ensure that the response variable corresponding to the polishing process processed by this group of polishing process parameter combinations is optimal.
[0069] S205 . Determine a parameter combination of the polishing process based on the association mapping relationship, and optimize the polishing process based on the parameter combination of the polishing process.
[0070] Specifically, the grinding process parameter combination corresponds to the grinding process parameters of the input model. For example, in one embodiment, in combination with the above embodiment, the grinding process parameter combination may be feed speed, grinding pressure and grinding wheel speed.
[0071] Furthermore, the specific method for optimizing the grinding process based on the grinding process parameter combination can be selected according to actual needs. For example, in one embodiment, grinding can be performed directly according to the grinding process parameter combination. For another example, in another embodiment, the grinding process parameter combination and the current grinding data can be averaged before grinding.
[0072] The polishing process optimization method provided in this embodiment first constructs a polishing process evaluation model including the mutual influence between subjective factors and polishing process parameters, then constructs a gray correlation model including polishing process parameters and response variables, and then determines the weight matrix and limit matrix of the process target based on the polishing process evaluation model, and then introduces the weight matrix and limit matrix into the gray correlation degree determined by the gray correlation model to determine the correlation mapping relationship, and finally determines the parameter combination of the polishing process based on the correlation mapping relationship, and optimizes the polishing process based on the parameter combination of the polishing process. In this way, the advice and experience of experts are integrated through the polishing process evaluation model to deal with the complex problems of subjective judgment, and the influence of several polishing process parameters on the response variables is analyzed through the gray correlation model. Further, through the combination of the polishing process evaluation model and the gray correlation model, the optimization of process parameters is not only based on the expert's experience and judgment, but also based on actual experimental data, ensuring the comprehensiveness and scientificity of the optimization process. Further, through the subjective and objective fusion quantitative optimization, the problems of single distribution influencing factors, too subjective or too objective, and insufficient flexibility are effectively solved. Furthermore, by calculating the correlation mapping relationship through the grinding process evaluation model and the grey correlation model, it can be ensured that the optimized grinding process parameters can effectively meet the requirements of the process objectives, thereby providing a higher quality surface treatment effect and improving the grinding effect of the grinding process.
[0073] Figure 4 This is a flow chart of Example 2 of the polishing process optimization method provided in this application. Please refer to Figure 4 Based on the above embodiment, the step of determining the weight matrix and limit matrix of the process target based on the polishing process evaluation model includes:
[0074] S401, constructing a decision matrix through the polishing process parameters and the process objectives; the decision matrix includes expert evaluation results represented by intuitive trapezoidal fuzzy numbers.
[0075] Specifically, the decision matrix is a tool used to integrate multiple experts' evaluation of grinding process parameters and response variables. The decision matrix can be used to make multi-attribute decisions in the presence of uncertainty and ambiguity.
[0076] Furthermore, the intuitionistic trapezoidal fuzzy number combines the trapezoidal fuzzy number and the intuitionistic fuzzy set. In specific implementation, for example, in one embodiment, the intuitionistic trapezoidal fuzzy number is represented as a quaternary group, and the quaternary group is used to represent the intuitionistic fuzzy part and the reminder fuzzy number part.
[0077] A specific implementation method is given below to introduce the construction process of the decision matrix in detail.
[0078] (1) Obtaining the expert evaluation result, and converting the expert evaluation result into the intuitive trapezoidal fuzzy number.
[0079] Specifically, the expert evaluation result is the evaluation of the grinding process parameters and response variables by multiple experts, and the subjective factors in the grinding process can be characterized by the expert evaluation result.
[0080] Furthermore, the expert evaluation results are converted into the form of intuitive trapezoidal fuzzy numbers. The specific conversion method can be found in the relevant description and will not be repeated here.
[0081] (2) Construct a decision information matrix based on all the intuitive trapezoidal fuzzy numbers.
[0082] Specifically, the decision information matrix contains all the expert evaluation results. In the specific implementation, the expert evaluation results are converted into the form of intuitive trapezoidal fuzzy numbers to ensure that all evaluation results have a unified representation method, and then all intuitive trapezoidal fuzzy numbers are formed into a decision information matrix in the form of a matrix.
[0083] (3) The decision information matrix is processed by intuitive trapezoidal fuzzy weighted averaging to obtain a comprehensive fuzzy number evaluation.
[0084] Specifically, the intuitive trapezoidal fuzzy weighted average method is used to process the trapezoidal fuzzy number part and the intuitive fuzzy number part respectively to obtain the comprehensive fuzzy number evaluation.
[0085] For example, in one embodiment, the comprehensive fuzzy number evaluation can be calculated by the following formula:
[0086] ,
[0087] Among them, r ij is the comprehensive fuzzy number evaluation, is the intuitionistic trapezoidal fuzzy number on the real number set R, where .
[0088] Furthermore, its membership and non-membership are defined as:
[0089] ,
[0090] ,
[0091] ,
[0092] Among them, A1, R1, R2 are decision matrices, v A , μ A is a fuzzy number.
[0093] (4) Constructing the decision matrix based on the comprehensive fuzzy number evaluation.
[0094] Specifically, the comprehensive fuzzy number evaluation combines all the expert evaluation results and summarizes the comprehensive fuzzy number evaluation into a decision matrix, through which all the expert evaluation results can be represented.
[0095] The polishing process optimization method provided in this embodiment first obtains the expert evaluation results, converts the expert evaluation results into intuitive trapezoidal fuzzy numbers, then constructs a decision information matrix based on all intuitive trapezoidal fuzzy numbers, and then uses intuitive trapezoidal fuzzy weighted average to process the decision information matrix to obtain a comprehensive fuzzy number evaluation, and finally constructs a decision matrix based on the comprehensive fuzzy number evaluation. In this way, by constructing an accurate decision matrix through intuitive trapezoidal fuzzy numbers and intuitive trapezoidal fuzzy weighted average methods, multi-attribute decision-making can be performed in the presence of uncertainty and ambiguity.
[0096] S402: Calculate the expected value corresponding to the decision matrix, and construct the weight matrix according to the expected value.
[0097] Specifically, the expected value is a single-valued index extracted from the intuitive trapezoidal fuzzy number, which is used to represent the central tendency of the intuitive trapezoidal fuzzy number. In specific implementation, for example, in one embodiment, the expected value can be calculated based on the four boundary points of the intuitive trapezoidal fuzzy number and the membership and non-membership of the intuitive trapezoidal fuzzy number.
[0098] When this step is implemented specifically, the weight matrix can be constructed based on the following formula:
[0099] ,
[0100] ,
[0101] ,
[0102] ,
[0103] Among them, ξ ij is the expected value, is the weight vector, S is the decision matrix, and U is the weight matrix.
[0104] S403: Combine the polishing process parameters in the network layer and obtain a weighted super matrix in the control layer after weighted processing.
[0105] Specifically, the weighted supermatrix is used to integrate the mutual influences between elements of each network layer and provide quantitative analysis for multi-criteria decision-making problems. In other words, the weighted supermatrix reflects the complex relationship of the entire system by combining the weight vectors of the mutual influences into a matrix.
[0106] In a specific implementation, for example, in one embodiment, the sorting vectors of the network layer elements that influence each other are combined to obtain a weighted supermatrix based on the following formula:
[0107] ,
[0108] Where P is the weighted super matrix.
[0109] S404: Construct the limit matrix based on the weighted supermatrix.
[0110] In specific implementation, for example, in one embodiment, the limit matrix can be obtained based on the following formula:
[0111] ,
[0112] Among them, β is the weighted super matrix and A is the limit matrix.
[0113] The polishing process optimization method provided in this embodiment first constructs a decision matrix through polishing process parameters and process objectives; the decision matrix contains the expert evaluation results represented by intuitive trapezoidal fuzzy numbers, and then calculates the expected value corresponding to the decision matrix, constructs a weight matrix according to the expected value, and then combines the polishing process parameters in the network layer, obtains the weighted supermatrix in the control layer after weighted processing, and finally constructs the limit matrix based on the weighted supermatrix. In this way, a weighted supermatrix and a limit matrix that effectively represent the expert evaluation results can be obtained. In this way, the decision information of the polishing process parameters and the response variables is given by intuitive intuitive trapezoidal fuzzy numbers, which effectively handles the uncertainty and fuzziness of the contribution rate of the polishing process parameters to the surface roughness and material removal rate, and improves the reliability of the optimization results. Furthermore, the weight corresponding to each polishing process parameter calculated can effectively reflect the expert evaluation results, and the subjective level in the polishing process can be integrated into the obtained matrix, ensuring the accuracy and reliability of the polishing optimization.
[0114] Figure 5 This is a flow chart of Example 3 of the polishing process optimization method provided in this application. Please refer to Figure 5 Based on the above embodiment, the step after constructing the grey correlation model including the polishing process parameters and the response variables further includes:
[0115] S501, calculating the signal-to-noise ratio of the response variable based on the principle of the largest value being the best;
[0116] Specifically, the principle of the largest value is the best and clearly states that the goal is to make the response variable as large as possible, and a larger response variable represents a higher grinding process efficiency.
[0117] In specific implementation, the signal-to-noise ratio of the response variable can be obtained by using a traditional signal-to-noise ratio calculation method, which will not be described in detail in this step. For example, the signal-to-noise ratio of the response variable can be calculated using the following formula:
[0118] ,
[0119] Where η is the signal-to-noise ratio, Rai is the surface roughness of the i-th experiment, and Zwi is the material removal rate of the i-th experiment.
[0120] S502 , performing data preprocessing on the signal-to-noise ratio based on the principle that the largest value is the best, to obtain normalized data.
[0121] Specifically, normalizing the signal-to-noise ratio can eliminate the impact between different data magnitudes, so that each signal-to-noise ratio data point can be evaluated under a unified standard.
[0122] In specific implementation, the signal-to-noise ratio can be processed by a traditional data preprocessing method, which will not be described in detail in this step. For example, the signal-to-noise ratio can be normalized by the following formula:
[0123] ,
[0124] in, is the initial sequence of signal-to-noise ratio, is a comparison sequence, where k=1,2,…q, i=1,2,…p, p is the number of experiments, and q is the number of targets corresponding to p.
[0125] S503: Calculate the grey relational degree based on the normalized data.
[0126] Specifically, the grey correlation degree may be calculated by using a traditional correlation degree calculation method, which is not limited in this embodiment.
[0127] A specific embodiment is given below to introduce the method of calculating the grey relational degree in detail:
[0128] (1) Determine the first normalized data as reference data.
[0129] In specific implementation, the first normalized data is used as reference data. For example, in one embodiment, is the reference sequence, the grey relational degree can be defined as follows:
[0130] ,
[0131] in, is the grey relational degree, is the reference sequence.
[0132] (2) Calculating the absolute difference between the normalized data and the reference data.
[0133] In specific implementation, this step can be calculated based on the following formula:
[0134] ,
[0135] in, is the absolute difference of k indicators.
[0136] It should be noted that is the minimum absolute difference between the reference sequence and the comparison sequence; is the maximum absolute difference between the reference sequence and the comparison sequence; is the resolution coefficient (usually 0.5).
[0137] (3) Calculating the grey relational degree based on the absolute difference.
[0138] In specific implementation, this step can be calculated based on the following formula:
[0139] ,
[0140] in, is the correlation between the reference sequence and the comparison sequence, β k is the weight of the kth response variable.
[0141] It should be noted that the larger the gray correlation degree is, the closer the matching grinding process parameter combination is to the optimal value of the response variable. In addition, the gray correlation degree also reflects the degree of influence of the grinding process parameter level on the quality characteristics.
[0142] The polishing process optimization method provided in this embodiment first calculates the signal-to-noise ratio of the response variable based on the principle of the largest value being the best, then performs data preprocessing on the signal-to-noise ratio based on the principle of the largest value being the best, obtains normalized data, and finally calculates the gray correlation based on the normalized data. In this way, the data uniformity is ensured by the signal-to-noise ratio and normalization processing, making the final calculation result more reliable, and then the gray correlation is calculated by the normalized data, which can effectively characterize the need of the response variable for the polishing process. In this way, since the gray correlation fully combines the objective parameters of the response variable, the optimization process of polishing is more objective, and the reliability of the optimization is guaranteed.
[0143] Optionally, the process of determining a parameter combination of the polishing process based on the association mapping relationship and optimizing the polishing process based on the parameter combination of the polishing process may include:
[0144] Processing the parameter combination of the polishing process based on a preset analysis method to obtain a polishing response table; wherein the polishing response table includes the parameter combination of the polishing process and the corresponding polishing effect;
[0145] The polishing process is optimized based on the polishing response table.
[0146] Specifically, the preset analysis method is set in advance according to actual needs, and is not limited to this in this embodiment.
[0147] Furthermore, the polishing response table may include different polishing process parameters and corresponding polishing target achievement effects, and the achievement effect of the polishing target may be characterized by the response variable.
[0148] Furthermore, after obtaining the polishing response table, the parameter combinations with the highest signal-to-noise ratio and grey correlation were calculated. A combination with a high signal-to-noise ratio means that the combination performed stably in the experiment, with small fluctuations and good effects, while a combination with a high grey correlation indicates that the effect of the combination has the strongest correlation with the ideal reference data and the effect is close to the best.
[0149] Furthermore, the selected optimal grinding process parameters are combined, and the grinding process is optimized based on this.
[0150] The polishing process optimization method provided in this embodiment processes the polishing process parameter combination based on a preset analysis method, generates a polishing response table including parameter combinations and polishing effects, and optimizes based on this table. This can significantly improve the efficiency and quality of the polishing process, ensure that the polishing process is stable and reliable, reduce production costs, and ultimately achieve the best polishing process optimization effect.
[0151] Optionally, a specific embodiment is given below to introduce the verification process of the present invention in detail:
[0152] In specific implementation, Ti-6Al-2Sn-4Zr-2Mo material sample can be selected for experiment. The chemical composition of Ti-6Al-2Sn-4Zr-2Mo is shown in Table 1.
[0153] Table 1
[0154]
[0155] Furthermore, the aircraft engine blade grinding experiment was carried out on the ABB IRB 4600 additional robot test bench. Figure 6 For a schematic diagram of the first embodiment of the grinding process provided in this application, please refer to Figure 6 During the experiment, the robot precisely grasped the blade and pulled it close for grinding. The width of the grinding wheel is 15mm, and a high-quality cloth base is used as the substrate to ensure the overall stability and durability. Diamond is selected as the abrasive to ensure the grinding effect and durability of the grinding wheel. The bonding agent of the grinding wheel is synthetic resin, which not only increases the durability of the grinding wheel, but also ensures stable performance under high load. In terms of data acquisition and processing, Lab Windows / CVI software developed by American Instruments was used, which can efficiently display and process the data collected in the experiment. In addition, the experiment used an IPC controller to control the grinding wheel in real time, and the control cycle was accurate to 0.01 seconds, ensuring the accuracy and consistency of the entire grinding process.
[0156] Furthermore, in order to study the relationship between the grinding process parameters and the response variables, the robot machining grinding experiment was completed by changing the feed speed, grinding positive pressure and grinding wheel speed during the robot grinding process. It should be noted that in this embodiment, the response variables are characterized by surface roughness and material removal rate. The main influencing factors and parameter levels in the grinding process are shown in Table 2:
[0157] Table 2
[0158]
[0159] Furthermore, in order to ensure sufficient data and reduce experimental costs, Taguchi orthogonal experiments can be used. It should be noted that the specific content of Taguchi orthogonal experiments can be found in the description of related technologies and will not be repeated here. In the specific implementation, considering that there are three input quantities, each with three level values, and two output quantities, the L9 (3^4) orthogonal table is used to conduct the experiment, and the raw data after the grinding experiment is preprocessed. The experimental data of the two responses are converted into signal-to-noise ratios, and the multi-objective optimization of surface roughness and material removal rate is converted into the maximum signal-to-noise ratio. The results are shown in Table 3.
[0160] Table 3
[0161]
[0162] Furthermore, the surface roughness was measured in the direction perpendicular to the polishing path using Keyence's VR-6000 3D profiler (sampling length of 5.6 mm). Five measurement points were selected on the machined surface of each blade before and after grinding, and the average measured value Ra of the surface roughness was taken as the final result. The measurement results are shown in Table 4:
[0163] Table 4
[0164]
[0165] The material removal rate (MRR) is indirectly determined by measuring the mass change of the blade before and after grinding. The calculation formula is as follows:
[0166] ,
[0167] Where MRR is the material removal rate.
[0168] Furthermore, three experts were used to make decision evaluations on the grinding process parameters and response variables. To ensure the objectivity and accuracy of the evaluation, seven language variables were set and corresponding conversion standards were formulated, as shown in Table 5, so as to convert the experts' language evaluations into quantifiable values:
[0169] Table 5
[0170]
[0171] Furthermore, the decision matrix constructed based on all expert evaluation structures is as follows:
[0172]
[0173] Combining the decision matrices, the resulting comprehensive decision matrix R is shown in Table 6.
[0174] Table 6
[0175] Grinding process parameters Surface roughness Material removal rate Feed speed vf ⟨ (0.73, 0.80, 0.87, 0.93),(0.70, 0.80, 0.87, 0.70) > ⟨ (0, 0.07, 0.13, 0.20),(0, 0.07,0.13, 0.20) > Grinding positive pressure Fn ⟨ (0, 0.07, 0.13, 0.20),(0,0.07, 0.13, 0.20) > ⟨ (0.73, 0.80, 0.87, 0.93),(0.70,0.80, 0.87, 0.87) > Grinding wheel speed vg ⟨ (0.23, 0.33, 0.43, 0.53),(0.13, 0.30, 0.43, 0.63) > ⟨ (0.43, 0.53, 0.63, 0.73),(0.33,0.53, 0.63, 0.83) >
[0176] Furthermore, the expected value of each element in the comprehensive decision matrix is calculated, and the comprehensive decision expectation matrix U is constructed by the expected value. The obtained expectation matrix U is as follows:
[0177]
[0178] The weight of each index is calculated. The main process parameter affecting the surface roughness is the feed speed vf, followed by the grinding wheel speed vg, while the grinding positive pressure Fn has little effect on the surface roughness; the grinding positive pressure Fn is the main process parameter affecting the material removal rate, followed by the grinding wheel speed vg, as shown in Table 7:
[0179] Table 7
[0180]
[0181] Please refer to Table 7 and repeat the above operations to obtain a surface roughness weight of 0.5743 and a material removal rate weight of 0.4257.
[0182] Furthermore, according to the criterion that the larger the feature, the better, the signal-to-noise ratio is normalized in the interval [0,1], and then the absolute value between the reference sequence and the comparison sequence of the k indicators is calculated. , and then determine the grayscale correlation, where the resolution coefficient is set to 0.5. The results are shown in Table 8.
[0183] Table 8
[0184]
[0185] Please refer to Table 8. The grey correlation degree of Group 6 is the largest, indicating that this group of process parameters best meets the multi-objective requirements. The calculation results show that the grinding process optimization method used in this method can process the evaluation data more comprehensively, and the grey correlation degree is higher.
[0186] Furthermore, the mean method is used to calculate the grey relational degree and Taguchi orthogonal test results, and the resulting response table is shown in Table 9:
[0187] Table 9
[0188]
[0189] Please refer to Table 9. The range of the grinding wheel speed vf is the largest, indicating that the grinding wheel speed vf has the greatest impact on the two response variables of aircraft engine blade grinding.
[0190] Figure 7This is a schematic diagram of the calculation results of the polishing process optimization method provided in this application. Figure 7 It can be seen that the grey correlation degree is the largest at level 2-2-3, and the optimized process parameter combination is grinding wheel speed 4000 mm / min, grinding positive pressure 80 N, and feed speed 600 mm / min.
[0191] Furthermore, in order to verify the validity of the results, the grinding wheel speed was 4000 mm / min, the grinding positive pressure was 80 N, and the feed speed was 600 mm / min. The grinding results are shown in Table 10:
[0192] Table 10
[0193]
[0194] Please refer to Table 10. The surface roughness after grinding optimization is 0.309µm and the material removal rate is 0.0605mm 2 / min, and compared with the polishing results before optimization, such as Figure 7 As shown in the figure, SR is reduced by 7.76%, MRR is increased by 22.72%, and GRG is increased by 7.18%, achieving the goal of reducing surface roughness and increasing material removal rate at the same time.
[0195] The polishing process optimization method provided in this embodiment has been verified through experiments and effectively improves the polishing quality obtained by the polishing process. It can process the evaluation data more comprehensively and with higher correlation, thereby achieving coordinated optimization of processing efficiency and surface quality.
[0196] Corresponding to the aforementioned embodiment of a polishing process optimization method, the present application also provides an embodiment of a polishing process optimization device.
[0197] The embodiment of a polishing process optimization device of the present application can be applied to a polishing process optimization device. The device embodiment can be implemented by software, or by hardware or a combination of software and hardware. Taking software implementation as an example, as a device in a logical sense, the processor of the polishing process optimization device in which it is located reads the corresponding computer program instructions in the non-volatile memory into the memory and runs them. From the hardware level, if Figure 8 As shown, it is a hardware structure diagram of the grinding process optimization device where the grinding process optimization device of this application is located, except Figure 8 In addition to the processor, memory, network interface, and non-volatile memory shown, the polishing process optimization device in which the device in the embodiment is located may also include other hardware according to the actual function of the polishing process optimization device, which will not be described in detail.
[0198] Fig. 9This is a schematic diagram of the structure of the first embodiment of the grinding process optimization device provided in this application. Fig. 9 The device provided in this embodiment includes a construction module 910, a determination module 920 and an optimization module 930; wherein,
[0199] The construction module 910 is used to construct a polishing process evaluation model including the mutual influence between subjective factors and polishing process parameters; the polishing process evaluation model includes a network layer and a control layer, the network layer includes a number of the polishing process parameters, and the control layer includes a number of process targets; the polishing process parameters in the network layer influence each other; the process targets in the control layer are independent of each other, and any process target is associated with at least one process parameter;
[0200] The construction module 910 is further used to construct a grey correlation model including the polishing process parameters and the response variables; the grey correlation model is used to determine the grey correlation degree between the polishing process parameters and the response variables;
[0201] The determination module 920 is used to determine the weight matrix and limit matrix of the process target based on the polishing process evaluation model;
[0202] The determination module 920 is further used to introduce the weight matrix and the limit matrix into the grey correlation degree determined by the grey correlation model to determine the correlation mapping relationship;
[0203] The optimization module 930 is used to determine a parameter combination of the polishing process based on the association mapping relationship, and optimize the polishing process based on the parameter combination of the polishing process.
[0204] The device of this embodiment can be used to perform Figure 1 The steps, specific implementation principles and implementation processes of the method embodiment shown are similar and will not be repeated here.
[0205] Optionally, the process goal is to optimize the process combination; the control layer also includes a number of process decision criteria, and the process decision criteria include at least minimum surface roughness and maximum material removal rate; and the grinding process parameters include at least feed speed, grinding pressure and grinding wheel speed.
[0206] Optionally, the construction module 910 is specifically used to construct a decision matrix through the polishing process parameters and the process objectives; the decision matrix includes expert evaluation results represented by intuitive trapezoidal fuzzy numbers;
[0207] The construction module 910 is further specifically used to calculate the expected value corresponding to the decision matrix, and construct the weight matrix according to the expected value;
[0208] The construction module 910 is further specifically used to combine the polishing process parameters in the network layer and obtain the weighted super matrix in the control layer after weighted processing;
[0209] The construction module 910 is further specifically configured to construct the limit matrix based on the weighted super matrix.
[0210] Optionally, the construction module 910 is further specifically used to calculate the signal-to-noise ratio of the response variable based on the principle of the largest value being the best;
[0211] The construction module 910 is further specifically used to perform data preprocessing on the signal-to-noise ratio based on the principle of the largest value being the best, to obtain normalized data;
[0212] The construction module 910 is further specifically configured to calculate the grey relational degree based on the normalized data.
[0213] Optionally, the determination module 920 is specifically used to obtain the expert evaluation result and convert the expert evaluation result into the intuitive trapezoidal fuzzy number;
[0214] The determination module 920 is further specifically used to construct a decision information matrix based on all the intuitive trapezoidal fuzzy numbers;
[0215] The determination module 920 is further specifically used to process the decision information matrix using intuitive trapezoidal fuzzy weighted average to obtain a comprehensive fuzzy number evaluation;
[0216] The construction module 910 is further specifically used to construct the decision matrix based on the comprehensive fuzzy number evaluation.
[0217] Optionally, the determination module 920 is further specifically configured to determine the first normalized data as reference data;
[0218] The determination module 920 is further specifically configured to calculate an absolute difference between the normalized data and the reference data;
[0219] The determination module 920 is further specifically configured to calculate the grey relational degree based on the absolute difference.
[0220] Optionally, the determination module 920 is further specifically configured to calculate a plurality of parameter weights corresponding to a plurality of the polishing process parameters based on the weight matrix and the limit matrix;
[0221] The determination module 920 is further specifically used to introduce the parameter weight into the grey correlation model to screen out the target parameter combination with the smallest grinding process distance from the response variable;
[0222] The determination module 920 is further specifically configured to determine the target parameter combination as the association mapping relationship.
[0223] Optionally, the determination module 920 is further specifically configured to process the corresponding grey correlation degree based on the parameter weight to obtain the optimal correlation degree of each polishing process parameter;
[0224] The determination module 920 is further specifically configured to determine the parameter combination of the polishing process with the optimal correlation as the target parameter combination.
[0225] Optionally, the determination module 920 is further specifically used to process the parameter combination of the polishing process based on a preset analysis method to obtain a polishing response table; wherein the polishing response table includes the parameter combination of the polishing process and the corresponding polishing effect;
[0226] The optimization module 930 is specifically configured to optimize the polishing process based on the polishing response table.
[0227] Please continue to refer to Figure 8 The present application also provides a polishing process optimization device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of any one of the methods provided in the first aspect of the present application are implemented.
[0228] The present application also provides a computer-readable storage medium having a computer program stored thereon, and when the program is executed by a processor, the steps of any one of the methods provided in the present application are implemented.
[0229] The implementation process of the functions and effects of each unit in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and will not be repeated here.
[0230] For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The device embodiment described above is only schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present application scheme. A person of ordinary skill in the art can understand and implement it without paying any creative work.
[0231] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.
Claims
1. A grinding process optimization method, It is characterized in that The method comprises: Constructing a polishing process evaluation model including the mutual influence between subjective factors and polishing process parameters; the polishing process evaluation model includes a network layer and a control layer, the network layer includes a number of the polishing process parameters, and the control layer includes a number of process targets; the polishing process parameters in the network layer influence each other; the process targets in the control layer are independent of each other, and any process target is associated with at least one process parameter; Constructing a grey correlation model including the grinding process parameters and the response variables; the grey correlation model is used to determine the grey correlation degree between the grinding process parameters and the response variables; Determine a weight matrix and a limit matrix of the process target based on the polishing process evaluation model; The step of determining the weight matrix and the limit matrix of the process target based on the polishing process evaluation model includes: A decision matrix is constructed by using the polishing process parameters and the process objectives; the decision matrix includes expert evaluation results represented by intuitive trapezoidal fuzzy numbers; Calculating the expected value corresponding to the decision matrix, and constructing the weight matrix according to the expected value; Combining the polishing process parameters in the network layer and obtaining a weighted supermatrix in the control layer after weighted processing; constructing the limit matrix based on the weighted supermatrix; The weight matrix is constructed based on the following formula: , , , , Among them, the ij is the expected value, is a weight vector, S is a decision matrix, and U is a weight matrix; Introducing the weight matrix and the limit matrix into the grey correlation degree determined by the grey correlation model to determine the correlation mapping relationship; The step of introducing the weight matrix and the limit matrix into the grey correlation degree determined by the grey correlation model to determine the correlation mapping relationship comprises: Calculate a plurality of parameter weights corresponding to a plurality of the polishing process parameters based on the weight matrix and the limit matrix; Introducing the parameter weight into the grey correlation model, and screening out the target parameter combination with the smallest grinding process distance from the response variable; Determining the target parameter combination as the association mapping relationship; Based on the association mapping relationship, a parameter combination of the polishing process is determined, and the polishing process is optimized based on the parameter combination of the polishing process.
2. The method according to claim 1, It is characterized in that The process goal is to optimize the process combination; the control layer also includes a number of process decision criteria, and the process decision criteria include at least the minimum surface roughness and the maximum material removal rate; the grinding process parameters include at least the feed speed, the grinding pressure and the grinding wheel speed.
3. The method according to claim 1, It is characterized in that After constructing the grey correlation model including the polishing process parameters and the response variables, the method further includes: Calculate the signal-to-noise ratio of the response variable based on the principle of maximum value being the best; Based on the principle that the maximum value is the best, data preprocessing is performed on the signal-to-noise ratio to obtain normalized data; The grey relational degree is calculated based on the normalized data.
4. The method according to claim 1, It is characterized in that The constructing a decision matrix through the polishing process parameters and the process objectives includes: Obtaining the expert evaluation result, and converting the expert evaluation result into the intuitive trapezoidal fuzzy number; Constructing a decision information matrix based on all the intuitive trapezoidal fuzzy numbers; The decision information matrix is processed by using intuitive trapezoidal fuzzy weighted average to obtain a comprehensive fuzzy number evaluation; The decision matrix is constructed based on the comprehensive fuzzy number evaluation.
5. The method according to claim 3, It is characterized in that The calculating the grey relational degree based on the normalized data comprises: Determining the first normalized data as reference data; Calculating the absolute difference between the normalized data and the reference data; The grey relational degree is calculated based on the absolute difference.
6. The method according to claim 1, It is characterized in that The step of introducing the parameter weight into the grey correlation degree and selecting the target parameter combination with the smallest grinding process distance from the response variable comprises: Based on the grey correlation degree corresponding to the parameter weight processing, an optimal correlation degree of each grinding process parameter is obtained; The parameter combination of the grinding process with the optimal correlation is determined as the target parameter combination.
7. The method according to claim 1, It is characterized in that The step of determining a parameter combination of the polishing process based on the association mapping relationship, and optimizing the polishing process based on the parameter combination of the polishing process includes: Processing the parameter combination of the polishing process based on a preset analysis method to obtain a polishing response table; wherein the polishing response table includes the parameter combination of the polishing process and the corresponding polishing effect; The polishing process is optimized based on the polishing response table.
8. A grinding process optimization device, It is characterized in that The device comprises a construction module, a determination module and an optimization module; wherein, The construction module is used to construct a polishing process evaluation model including the mutual influence between subjective factors and polishing process parameters; the polishing process evaluation model includes a network layer and a control layer, the network layer includes a number of the polishing process parameters, and the control layer includes a number of process targets; the polishing process parameters in the network layer influence each other; the process targets in the control layer are independent of each other, and any process target is associated with at least one process parameter; The construction module is further used to construct a grey correlation model including the polishing process parameters and the response variables; the grey correlation model is used to determine the grey correlation degree between the polishing process parameters and the response variables; The determination module is used to determine the weight matrix and limit matrix of the process target based on the polishing process evaluation model; The step of determining the weight matrix and the limit matrix of the process target based on the polishing process evaluation model includes: A decision matrix is constructed by using the polishing process parameters and the process objectives; the decision matrix includes expert evaluation results represented by intuitive trapezoidal fuzzy numbers; Calculating the expected value corresponding to the decision matrix, and constructing the weight matrix according to the expected value; Combining the polishing process parameters in the network layer and obtaining a weighted supermatrix in the control layer after weighted processing; constructing the limit matrix based on the weighted supermatrix; The weight matrix is constructed based on the following formula: , , , , Among them, the ij is the expected value, is a weight vector, S is a decision matrix, and U is a weight matrix; The determination module is further used to introduce the weight matrix and the limit matrix into the grey correlation degree determined by the grey correlation model to determine the correlation mapping relationship; The step of introducing the weight matrix and the limit matrix into the grey correlation degree determined by the grey correlation model to determine the correlation mapping relationship comprises: Calculate a plurality of parameter weights corresponding to a plurality of the polishing process parameters based on the weight matrix and the limit matrix; Introducing the parameter weight into the grey correlation model, and screening out the target parameter combination with the smallest grinding process distance from the response variable; Determining the target parameter combination as the association mapping relationship; The optimization module is used to determine a parameter combination of the polishing process based on the association mapping relationship, and optimize the polishing process based on the parameter combination of the polishing process.
