A method for analyzing the coupling relationship of process parameters in component docking process

Abstract

This invention discloses a method for analyzing the coupling relationship of process parameters in the component docking process, comprising: defining measurement points characterizing the docking quality characteristics of components as quality feature points, defining the positioning positions of process equipment in the docking process as process monitoring points, defining the spatial coordinates of the quality feature points as dependent variable Y, and defining the spatial coordinates of the process monitoring points as independent variable X; collecting the spatial coordinates of independent variable X at different times and the corresponding spatial coordinates of dependent variable Y at each time during the component docking process; performing data standardization on independent variable X and dependent variable Y to obtain the standardized independent variable matrix X0 and standardized dependent variable matrix Y0 after standardization of dependent variable Y and independent variable X; constructing a regression model of independent variable X and dependent variable Y using an improved partial least squares algorithm based on the standardized independent variable matrix X0 and standardized dependent variable matrix Y0; and predicting the spatial coordinates of docking quality characteristics according to the regression model. The technical solution provided by this invention solves the problem that in the existing docking process of aircraft components, the low utilization rate of attitude adjustment process data and the current lack of research on the process parameter relationship mining in the attitude adjustment process lead to a slow overall component docking and assembly process during the attitude adjustment process.

Description

Technical Field

[0001] This invention relates to, but is not limited to, the field of monitoring technology for aircraft component docking and assembly processes, specifically to a method for analyzing the correlation between quality characteristics and process equipment in docking and assembly. Background Technology

[0002] The docking of aircraft components is a critical step in aircraft manufacturing, and its quality directly affects flight safety and overall aircraft performance. Currently, the attitude adjustment principle for component docking is as follows: 1) Measure the spatial coordinates of the horizontal measurement point at the current position of the component using a laser tracker, and display the current position of the attitude adjustment positioner on the attitude adjustment industrial control computer interface; 2) Calculate the attitude adjustment target position of the positioner by fitting the measured horizontal measurement point position with the theoretical position; 3) Drive the attitude adjustment positioner to the target position, measure the spatial coordinates of the horizontal measurement point again using a laser tracker, and evaluate whether the quality parameters are qualified; 4) Repeat the above three steps until the docking attitude quality parameters are all within reasonable tolerance ranges.

[0003] As can be seen from the existing attitude adjustment principles and processes described above, the positions of the product's horizontal measurement points and the attitude adjustment positioner are key process parameters during component docking. The attitude adjustment positioner's movement is driven by the product's horizontal measurement points, and its position also characterizes the position of the horizontal measurement points to a certain extent. However, due to the low utilization rate of current attitude adjustment process data and the scarcity of research on the relationships between process parameters during the attitude adjustment process, the overall component docking and assembly process is currently slow. Summary of the Invention

[0004] The purpose of this invention is to solve the above-mentioned technical problems. This invention provides a method for analyzing the coupling relationship of process parameters in the component docking process. This method addresses the problem that the current docking process for aircraft components is slow due to the low utilization rate of attitude adjustment process data and the lack of research on the relationship between process parameters in the attitude adjustment process.

[0005] The technical solution of the present invention: In a first aspect, embodiments of the present invention provide a method for analyzing the coupling relationship of process parameters in the component docking process, comprising the following steps:

[0006] Step 1: Define the measurement point that characterizes the docking quality of the components as the quality feature point, define the positioning position of the process equipment in the docking process as the process monitoring point, and define the spatial position coordinates of the quality feature point as the dependent variable Y, and the spatial position coordinates of the process monitoring point as the independent variable X.

[0007] Step 2: Collect the spatial coordinates of the independent variable X at different times during the component docking process, as well as the spatial coordinates of the dependent variable Y at each time.

[0008] Step 3: Standardize the data of independent variable X and dependent variable Y to obtain the standardized independent variable matrix X0 and standardized dependent variable matrix Y0 after standardization of dependent variable Y and independent variable X.

[0009] Step 4: Based on the standardized independent variable matrix X0 and the standardized dependent variable matrix Y0, an improved partial least squares algorithm is used to construct a regression model for the independent variable X and the dependent variable Y.

[0010] Step 5: Based on the regression model, predict the spatial coordinates of the docking quality characteristics.

[0011] Optionally, in the process parameter coupling relationship analysis method for component docking process described above, in step 2, the spatial position coordinates of the independent variable X at different times are:

[0012]

[0013] The spatial coordinates of the dependent variable Y at each time point obtained are as follows:

[0014]

[0015] Where n represents the number of samples for independent variable X, p represents the number of samples for independent variable X in each sample; m represents the number of samples for dependent variable Y, and l represents the number of samples for dependent variable Y in each sample.

[0016] Optionally, in the process parameter coupling relationship analysis method for component docking process described above, the formulas for the standardized independent variable matrix X0 and the standardized dependent variable matrix Y0 obtained through standardization processing in step three are as follows:

[0017]

[0018] Wherein, any element in the standardized independent variable matrix X0 This represents the average value of the data in the j-th column of the independent variable X. This represents the standard deviation of the data in the j-th column of the independent variable X;

[0019] Standardize any element in the dependent variable matrix Y0 This represents the average value of the data in the j-th column of matrix Y. Let represent the standard deviation of the data in the j-th column of matrix Y.

[0020] Optionally, in the above-described method for analyzing the coupling relationship of process parameters in the component docking process, the method for constructing the regression model of independent variable X and dependent variable Y in step four is as follows:

[0021] Adopt the partial least squares regression modeling method based on Renyi entropy to construct the regression model of independent variable X and dependent variable Y.

[0022] Optionally, in the process parameter coupling relationship analysis method for component docking process as described above, the calculation process of Renyi entropy in step 4 is as follows:

[0023] Step 41, calculate the corresponding independent variable covariance matrix K of the standardized independent variable matrix X0 and the standardized dependent variable matrix Y0 x and the dependent variable covariance matrix K y , and the calculation formula is as follows:

[0024]

[0025] Step 42, solve the corresponding eigenvalues x of the independent variable covariance matrix K y and the dependent variable covariance matrix K and the eigenvectors

[0026] Step 43, calculate the Renyi entropy corresponding to each eigenvalue in the independent variable covariance matrix K x , and the calculation formula is:

[0027]

[0028] where E x represents the unit vector with the same dimension as the eigenvector , k represents the number of eigenvalues of the independent variable covariance matrix K x .

[0029] Optionally, in the process of constructing the regression model by using the partial least squares regression modeling method based on Renyi entropy in step 4 of the process parameter coupling relationship analysis method for component docking process as described above, the method for extracting the principal components and determining the number of principal components includes:

[0030] Step 44, arrange the Renyi entropy in descending order, and calculate the cumulative information rate ω corresponding to the eigenvalues and eigenvectors of the independent variable covariance matrix K x , and the calculation formula is:

[0031]

[0032] Step 45, solve the maximum cumulative information rate and the corresponding position coordinate j, which requires 1 < j < k, and calculate the information attenuation rate corresponding to the eigenvalues and eigenvectors of the first j , and the calculation formula is as follows:

[0033]

[0034] Step 46: Calculate the information decay rate when the cumulative information rate ω is at its maximum. If the maximum number of principal components k′ corresponding to the attenuation rate is greater than the preset attenuation rate, then k′ satisfies 0. <k′<p;

[0035] Step 47: Based on the determined number of principal components k′, use partial least squares fitting to fit the regression model of independent variable X and dependent variable Y, and determine the fitting coefficient matrix A and constant term C of the regression model.

[0036] Optionally, in the process parameter coupling relationship analysis method for component docking process described above, the regression model for fitting the independent variable X and dependent variable Y in step 47 is as follows:

[0037] Y T =AX T +C;

[0038] Where A is the coefficient of the independent variable in the regression model, and C is the constant term.

[0039] Optionally, in the process parameter coupling relationship analysis method for component docking process described above,

[0040] The preset attenuation rate is between [0.1, 0.2].

[0041] This invention also provides a computer-readable storage medium, including: a memory and a processor;

[0042] The memory is configured to store executable instructions;

[0043] The processor is specifically configured to implement, when executing the executable instructions stored in the memory, a process parameter coupling relationship analysis method for component docking processes as described above.

[0044] The beneficial effects of this invention are as follows: Addressing the problem of unclear influence mechanisms of process parameters during component docking and assembly, leading to low efficiency due to repeated measurements, this invention provides a method for analyzing the coupling relationship of process parameters during component docking. This method collects data on the position of the attitude adjustment locator and the position of the product's horizontal measurement points during the component docking and assembly process. It constructs a process parameter regression model based on improved partial least squares (PLS). By optimizing principal component extraction and the number of principal components in the PLS model using Reiny entropy optimization, it solves the problem of multiple iterations in high-dimensional data regression caused by cross-validation in traditional PLS regression model construction. This improves the computational efficiency for constructing regression fitting models for high-dimensional data. The improved PLS regression model can predict the spatial position of the product's horizontal measurement points in real time based on the attitude adjustment locator data, effectively reducing the number of repeated measurements during the attitude adjustment process, improving attitude adjustment efficiency, and providing theoretical support for attitude adjustment trajectory planning and product attitude evaluation. Attached Figure Description

[0045] The accompanying drawings are used to provide a further understanding of the technical solution of the present invention and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the technical solution of the present invention and do not constitute a limitation on the technical solution of the present invention.

[0046] Figure 1 A flowchart of a method for analyzing the coupling relationship of process parameters in a component docking process provided in an embodiment of the present invention;

[0047] Figure 2 This is a schematic diagram of the changes in Renyi entropy accumulation information rate and information decay rate with the number of principal components in an embodiment of the present invention;

[0048] Figure 3 This is a schematic diagram illustrating the deviation between the predicted and actual values ​​based on partial least squares fitting in an embodiment of the present invention.

[0049] Figure 4 This is a schematic diagram illustrating the deviation between the predicted and actual positions of points based on partial least squares fitting in an embodiment of the present invention. Detailed Implementation

[0050] To make the purpose, technical solutions and advantages of the present invention more clearly understood, the embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be noted that, unless there is a conflict, the embodiments and features in the embodiments of the present application can be combined with each other in any manner.

[0051] As explained in the background section above, the importance of aircraft component docking in aircraft manufacturing is based on the existing aircraft component docking process and attitude adjustment principle. However, the following specific problems still exist in the attitude adjustment process: 1) After the attitude adjustment is completed, although the position of the attitude adjustment locator is known, the position of the horizontal measurement point needs to be remeasured; 2) The position of the horizontal measurement point needs to be measured once for each attitude adjustment, resulting in multiple repeated measurements. At the same time, due to the low measurement efficiency of the laser tracker, the overall component docking and assembly process is slow.

[0052] To address the aforementioned issues, a key problem urgently needs to be solved in the component docking and assembly process: how to uncover the correlation between the position of the attitude adjustment locator and the horizontal measurement point during the docking and assembly process, explore the pattern of the horizontal measurement point changing with the attitude adjustment position, and use the position data of the attitude adjustment locator to characterize the position of the horizontal measurement point, thereby reducing the need for multiple measurements and improving attitude adjustment efficiency. Based on the above analysis, this invention provides a method for analyzing the coupling relationship of process parameters in the component docking process.

[0053] The present invention provides the following specific embodiments, which can be combined with each other. For the same or similar concepts or processes, they may not be described again in some embodiments.

[0054] Figure 1 This is a flowchart of a method for analyzing the coupling relationship of process parameters in a component docking process, provided in an embodiment of the present invention. Figure 1 The diagram illustrates the key process parameter coupling relationship analysis process during component docking. This invention provides a method for analyzing the process parameter coupling relationship during component docking, comprising the following steps:

[0055] Step 1: Define the measurement point that characterizes the docking quality of the components as the quality feature point, define the positioning position of the process equipment in the docking process as the process monitoring point, and define the spatial position coordinates of the quality feature point as the dependent variable Y, and the spatial position coordinates of the process monitoring point as the independent variable X.

[0056] Step 2: Collect the spatial coordinates of the independent variable X at different times during the component docking process, as well as the spatial coordinates of the dependent variable Y at each time.

[0057] Step 3: Standardize the data of independent variable X and dependent variable Y to obtain the standardized independent variable matrix X0 and standardized dependent variable matrix Y0 after standardization of dependent variable Y and independent variable X.

[0058] Step 4: Based on the standardized independent variable matrix X0 and the standardized dependent variable matrix Y0, an improved partial least squares algorithm is used to construct a regression model for the independent variable X and the dependent variable Y.

[0059] Step 5: Based on the regression model, predict the spatial coordinates of the docking quality characteristics.

[0060] In one implementation of this invention, the spatial coordinates of the independent variable X obtained in step 2 above at different times are:

[0061]

[0062] The spatial coordinates of the dependent variable Y at each time point obtained are as follows:

[0063]

[0064] Where n represents the number of samples of independent variable X, p represents the number of samples of independent variable X in each sample; m represents the number of samples of dependent variable Y, and l represents the number of samples of dependent variable Y in each sample.

[0065] In one implementation of this invention, the formulas for the standardized independent variable matrix X0 and the standardized dependent variable matrix Y0 obtained through the standardization process in step three above are as follows:

[0066]

[0067] Wherein, any element in the above-mentioned standardized independent variable matrix X0 This represents the average value of the data in the j-th column of the independent variable X. This represents the standard deviation of the data in the j-th column of the independent variable X;

[0068] Standardize any element in the dependent variable matrix Y0 This represents the average value of the data in the j-th column of matrix Y. Let represent the standard deviation of the data in the j-th column of matrix Y.

[0069] In one implementation of this invention, the method for constructing the regression model of independent variable X and dependent variable Y in step four above is as follows:

[0070] A partial least squares regression model based on Renyi entropy was used to construct a regression model for independent variable X and dependent variable Y.

[0071] In this implementation, the calculation process of Renyi entropy in step four is as follows:

[0072] Step 41: Calculate the covariance matrix K of the independent variables corresponding to the standardized independent variable matrix X0 and the standardized dependent variable matrix Y0, respectively. x and the covariance matrix of the dependent variable K y The calculation formula is as follows:

[0073]

[0074] Step 42, solve the covariance matrix K of independent variables respectively x and the covariance matrix K of dependent variables y corresponding eigenvalues and eigenvectors

[0075] Step 43, calculate the Renyi entropy corresponding to each eigenvalue in the covariance matrix K of independent variables x The calculation formula is:

[0076]

[0077] where, E x represents the unit vector with the same dimension as the eigenvector k represents the number of eigenvalues of the covariance matrix K of independent variables x of.

[0078] Furthermore, in the process of constructing the regression model by using the partial least squares regression modeling method based on Renyi entropy in Step 4, the methods for principal component extraction and determination of the number of principal components include:

[0079] Step 44, sort the Renyi entropy in descending order, and calculate the cumulative information rate ω corresponding to the eigenvalues and eigenvectors of the covariance matrix K of independent variables x The calculation formula is:

[0080]

[0081] Step 45, solve the maximum cumulative information rate and the corresponding position coordinate j, which requires 1 < j < k, and calculate the information attenuation rate corresponding to the first j eigenvalues and eigenvectors The calculation formula is as follows:

[0082]

[0083] Step 46, calculate the information attenuation rate when the cumulative information rate ω is the largest, and the maximum number of principal components k′ corresponding to when the information attenuation rate is greater than the preset attenuation rate, then k′ satisfies 0 < k′ < p;

[0084] Step 47, according to the determined number of principal components k′, use the partial least squares fitting method to fit the regression model of independent variable X and dependent variable Y, and determine the fitting coefficient matrix A and constant term C of the regression model.

[0085] Specifically, in the implementation, the regression model for fitting independent variable X and dependent variable Y in Step 47 is:

[0086] YT =AX T +C;

[0087] Where A is the coefficient of the independent variable in the regression model, and C is the constant term.

[0088] The preset attenuation rate in step 46 above is between [0.1, 0.2].

[0089] To address the problem of unclear influence mechanisms of process parameters during component docking and assembly, leading to low efficiency due to repeated measurements, this invention provides a method for analyzing the coupling relationship of process parameters during component docking. This method collects data on the position of the attitude adjustment locator and the position of the product's horizontal measurement points during the component docking and assembly process. It then constructs a process parameter regression model based on improved partial least squares (PLS). By optimizing principal component extraction and the number of principal components in the PLS model using Reiny entropy optimization, it solves the problem of multiple iterations in high-dimensional data regression caused by cross-validation in traditional PLS regression model construction. This improves the computational efficiency for building regression fitting models for high-dimensional data. The improved PLS regression model can predict the spatial position of the product's horizontal measurement points in real time based on the attitude adjustment locator data, effectively reducing the number of repeated measurements during the attitude adjustment process, improving attitude adjustment efficiency, and providing theoretical support for attitude trajectory planning and product attitude evaluation.

[0090] Based on the method for analyzing the coupling relationship of process parameters in the component docking process provided in the above embodiments of the present invention, the present invention also provides a computer-readable storage medium, characterized in that it includes: a memory and a processor;

[0091] The memory is configured to store executable instructions;

[0092] The processor is specifically configured to implement, when executing the executable instructions stored in the memory, a process parameter coupling relationship analysis method for component docking processes as provided in any of the above embodiments.

[0093] Application Cases

[0094] This embodiment provides a method for analyzing the coupling relationship of process parameters in the component docking process. This application case uses the process parameter coupling relationship analysis of a large aircraft wing attitude adjustment process as an example. This application case focuses on the wing docking and assembly attitude adjustment process, using the position data of the attitude adjustment locator group and the horizontal measurement point position data as a basis. Utilizing data mining knowledge, a partial least squares process coupling relationship model based on Renyi entropy is constructed. The horizontal measurement point position is predicted using the attitude adjustment locator position data for each attitude adjustment, providing theoretical support for the next attitude adjustment direction and magnitude, thereby improving attitude adjustment efficiency. The specific algorithm includes the following steps:

[0095] Step 1: Define the horizontal measurement point of the product during the wing docking and attitude adjustment process as the dependent variable, and the corresponding position data of the attitude adjustment positioner group as the independent variable;

[0096] Step 2: Collect position data of the attitude adjustment locator group and the corresponding horizontal measurement point position data at the same time. The position data of the attitude adjustment locator group contains 12 monitoring data points and 15 sample sets, forming a 15×12 independent variable X. The horizontal measurement point data contains 36 monitoring data points and 15 sample sets. In this calculation, 3 monitoring data points are used to form a 15×3 dependent variable matrix Y. The calculation method of other dependent variable points is the same and will not be described again.

[0097] Step 3: Standardize the independent variable X and dependent variable Y to obtain the standardized independent variable matrix X0 and standardized dependent variable matrix Y0, which can be represented as:

[0098]

[0099] Step 4: Based on the standardized independent variable matrix X0 and the standardized dependent variable matrix Y0, the partial least squares method of Renyi entropy is used to construct a regression model for the independent variable X and the dependent variable Y.

[0100] First, the Renyi entropy of the independent and dependent variable matrices is calculated, and the calculation process is shown below:

[0101] 4-1 Calculate the covariance matrix K of the independent variable X and the dependent variable Y respectively. x and the covariance matrix of the dependent variable K y , means as follows:

[0102]

[0103] 4-2 Solve for the covariance matrix K of the independent variable. x and the covariance matrix of the dependent variable K y Corresponding eigenvalues and eigenvectors The calculation result is expressed as follows: D x =diag(-2.9360×10) -16 -2.3271×10 -16 1.3748×10 -16 5.54355×10 -16 2.6464×10 -6 ,0.0023,0.1088,0.3900,0.6906,1.3666,2.4731,6.1686);

[0104]

[0105] D y = diag(0.2902, 0.9043, 1.6056)

[0106]

[0107] 4-3 Calculate the covariance matrix K of independent variables according to the formula x The Renyi entropy corresponding to each eigenvalue, and the calculation results are as follows: H x = [-1.48×10 -18 , -2.9142×10 -17 , 5.8467×10 -19 , 1.8441×10 -17 , 9.4898×10 -07 , 0.0126, 0.0113, 0.4384, 0.0342, 4.0609, 4.5243, 0.2127];

[0108] 4-4 Sort in descending order, calculate the cumulative information rate ω corresponding to the eigenvalues and their eigenvectors, and the calculation results are as follows:

[0109] ω x = [0.4868,​​​​​​​​​​​​​​​​​​​​​​​​

[0115] Y T =AX T +C.

[0116] Step 5: Based on the regression model determined in Step 4, predict the spatial coordinates of the horizontal measurement points during the 10 attitude adjustments. The error between these predictions and the actual measurements is as follows: Figure 2 As shown.

[0117] This application case focuses on the attitude adjustment process of a large aircraft wing docking. Based on the position data of the attitude adjustment locator and its corresponding product horizontal measurement point position data, this study explores the coupling influence between the attitude adjustment locator position data and the product horizontal measurement point position data through data collection, analysis, and mining. A partial least squares regression modeling method based on Renyi entropy is proposed. This algorithm is used to perform fitting regression modeling on the attitude adjustment process data of 15 large aircraft wing docking processes, and the position data of the No. 1 horizontal measurement point in 10 wing attitude adjustment processes are predicted. The deviation between the predicted spatial position and the actual measured spatial position is shown below. Figure 3 As shown, it can be seen that the deviation between the spatial position of the product measurement point calculated by this algorithm and the actual position is at most about 0.5mm and at least about 0.16mm. It can completely predict the current wing attitude in real time through the attitude adjustment positioner during the attitude adjustment process, which can effectively reduce the number of measurements in the attitude adjustment process, and only the inspection measurement is required at the final attitude.

[0118] While the embodiments disclosed in this invention are as described above, they are merely illustrative of the embodiments to facilitate understanding of the invention and are not intended to limit the invention. Any person skilled in the art to which this invention pertains may make any modifications and variations in the form and details of the implementation without departing from the spirit and scope disclosed herein; however, the scope of patent protection for this invention shall still be determined by the scope defined in the appended claims.

Claims

1. A method for analyzing the coupling relationship of process parameters in component docking processes, characterized in that, include; Step 1: Define the measurement point that characterizes the docking quality of the components as the quality feature point, define the positioning position of the process equipment in the docking process as the process monitoring point, and define the spatial position coordinates of the quality feature point as the dependent variable Y, and the spatial position coordinates of the process monitoring point as the independent variable X. Step 2: Collect the spatial coordinates of the independent variable X at different times during the component docking process, as well as the spatial coordinates of the dependent variable Y at each time. Step 3: Standardize the independent variable X and dependent variable Y to obtain the standardized independent variable matrix after standardization of dependent variable Y and independent variable X. Standardized dependent variable matrix ; Step 4, based on the standardized independent variable matrix Standardized dependent variable matrix An improved partial least squares algorithm is used to construct the independent variable X and the dependent variable. The regression model; Step 5: Based on the regression model, predict the spatial coordinates of the docking quality characteristics; In step two, the spatial coordinates of the independent variable X at different times are: ； The spatial coordinates of the dependent variable Y at each time point obtained are as follows: ； Where n represents the number of samples for independent variable X, p represents the number of samples of independent variable X in each sample; m represents the number of samples for dependent variable Y, and l represents the number of samples of dependent variable Y in each sample; The pyramidalized independent variable matrix obtained through pyramidalization in step three and standardized dependent variable matrix The formulas are expressed as follows: ； ； Wherein, the standardized independent variable matrix any element in , This represents the average value of the data in the j-th column of the independent variable X. This represents the standard deviation of the data in the j-th column of the independent variable X; Standardized dependent variable matrix any element in , This represents the average value of the data in the j-th column of matrix Y. This represents the standard deviation of the data in the j-th column of matrix Y; The method for constructing the regression model of independent variable X and dependent variable Y in step four is as follows: The partial least squares regression modeling method based on Renyi entropy is used to construct the independent variable X and the dependent variable X. The regression model; The calculation process of Renyi entropy in step four is as follows: Step 41, calculate the standardized independent variable matrix respectively. Standardized dependent variable matrix The corresponding independent variable covariance matrix and the covariance matrix of the dependent variables The calculation formula is as follows: ； Step 42, solve for the covariance matrix of the independent variables respectively. and the covariance matrix of the dependent variables Corresponding eigenvalues , and eigenvectors , ; Step 43, calculate the covariance matrix of the independent variables. The Renyi entropy corresponding to each feature value is calculated using the following formula: ； in Representation and eigenvectors Unit vectors of equal dimension, where k represents the covariance matrix of the independent variables. The number of eigenvalues; In step four, the partial least squares regression modeling method based on Renyi entropy is used to construct the regression model. The methods for principal component extraction and determining the number of principal components include: Step 44, Renyi entropy Sort in descending order and calculate the covariance matrix of the independent variables. Cumulative information rate of eigenvalues ​​and their eigenvectors The calculation formula is: ； Step 45: Solve for the maximum cumulative information rate and the corresponding position coordinates j, requiring that the following conditions are met. And calculate the information decay rate corresponding to the first j eigenvalues ​​and eigenvectors. The calculation formula is as follows: ； Step 46, calculate the cumulative information rate. At its maximum, the information decay rate The maximum number of principal components corresponding to a decay rate greater than the preset decay rate ,but Satisfying 0< <p; Step 47, based on the determined number of principal components The independent variables were fitted using the partial least squares fitting method. With dependent variable The regression model is determined, and the fitting coefficient matrix A and constant term C of the regression model are determined.

2. The method for analyzing the coupling relationship of process parameters in the component docking process according to claim 1, characterized in that, In step 47, the fitted independent variables With dependent variable The regression model is as follows: ； Where A is the fitting coefficient matrix of the regression model, and C is the constant term.

3. The method for analyzing the coupling relationship of process parameters in the component docking process according to claim 1, characterized in that, The preset attenuation rate is between [0.1, 0.2].

4. A computer-readable storage medium, characterized in that, include: Memory and processor; The memory is configured to store executable instructions; The processor is specifically configured to implement the process parameter coupling relationship analysis method for component docking process as described in any one of claims 1 to 3 when executing the executable instructions stored in the memory.

Images