Method and system for predicting quality of tire cord steel decarburized layer

Abstract

The invention discloses a tire cord steel decarburized layer quality prediction method and system, and the method comprises the steps: S1, obtaining sample data in a tire cord steel production process, the sample data comprising scalar data and time sequence data, and lengthening the scalar data to obtain multivariable time sequence data; s2, fitting the multivariable time series data into multivariable function type data by utilizing B spline primary function fitting; s3, projecting the multivariable function type data to a high-dimensional function type feature space by using a function type data kernel method; s4, calculating a function type data projection hyperplane enabling the intra-class distance to be minimum and the inter-class distance to be maximum based on the function type features under the high dimension; and S5, projecting each sample to obtain a final sample classification result. By adopting the method, the quality of the tire cord steel decarburized layer can be accurately predicted in real time.

Description

Technical Field

[0001] The invention relates to the technical field of quality prediction of a decarburized layer of a cord steel, and in particular to a method and system for quality prediction of a decarburized layer of a cord steel. Background Art

[0002] In the production process of cord steel, the detection of the decarburization layer depth (decarburization layer quality) is usually carried out after the cord steel wire rod comes off the line, and usually requires the use of metallographic detection methods, which are high-cost and lagging, and cannot meet the needs of modern steel companies for real-time acquisition of product quality. Summary of the invention

[0003] The present invention provides a method and system for predicting the quality of the decarburized layer of cord steel, which is used to solve the problems existing in the above-mentioned prior art. The technical solution is as follows:

[0004] On the one hand, a method for predicting the quality of a decarburized layer of a cord steel is provided, comprising:

[0005] S1. Obtain sample data in the production process of cord steel, wherein the sample data includes scalar data and time series data, and stretch the scalar data to obtain multivariate time series data;

[0006] S2, fitting the multivariate time series data into multivariate function data using B-spline basis function fitting;

[0007] S3, projecting the multivariate functional data into a high-dimensional functional feature space using a functional data kernel method;

[0008] S4. Based on the functional features in high dimensions, calculate the functional data projection hyperplane that minimizes the intra-class distance and maximizes the inter-class distance;

[0009] S5. Project each sample to obtain the final sample classification result.

[0010] Optionally, the step of stretching the scalar data in S1 to obtain multivariate time series data specifically includes:

[0011] The scalar data is converted into a constant function whose domain size is the heating furnace's in-furnace time, and is concatenated with the time series data in the variable dimension to obtain the multivariate time series data.

[0012] Optionally, the S2 specifically includes:

[0013] For functional data fitting, the function x obtained after fitting can be expressed as the inner product of the coefficient vector and the basis function: Where c is the coefficient vector, T represents the matrix transpose, is the basis function;

[0014] The coefficient vector c is determined by the least square method. At this time, the weighted fitting error term SMSSE(y i,j |c i,j )=(y i,j -Ψ j c i,j ) T W i,j (y i,j -Ψ j c i,j ),in, is the basis function numerical matrix of the jth variable, D j is the number of basis functions used when fitting functional data for the jth variable, is the lth basis function used when fitting the jth variable function data, L j is the length of observation time of the jth variable, W i,j is the jth curve weight matrix of the i-th sample, which is defined as the inverse of the covariance matrix calculated from the fitted residuals of all observation points of the curve of the j-th variable of the i-th sample;

[0015] Functional data fitting based on the least squares method will lose a certain degree of smoothness, so it is necessary to introduce a roughness penalty term to obtain the least squares fitting error PENSSE with a penalty term. m (y i,j |c i,j )=(y i,j -Ψ j c i,j ) T W i,j (y i,j -Ψ j c i,j )+λc i,j T R j c i,j , where R j is a weight matrix that reflects the relationship between variables. The coefficient vector c is solved using the least squares method. i,j =(Ψ j T W i,j Ψ j +λR j ) -1 Ψ j T W i,j y i,j , where λ is the smoothness coefficient.

[0016] Optionally, the S3 specifically includes:

[0017] Select the functional Gaussian radial basis kernel function k er , the multivariable functional data is projected into a high-dimensional space through a high-dimensional mapping, thereby converting the linearly inseparable low-dimensional multivariable functional data into linearly separable multivariable functional data in high dimensions. The inner product of the multivariable functional data in high dimensions can be expressed as k er (x1(t),x2(t))=exp(||x1(t)-x2(t)|| 2 / S 2 ), where S is the parameter of the kernel function, ||x1(t)-x2(t)|| 2 It is expressed as:

[0018]

[0019] Therefore, the final functional kernel function k er It can be expressed as a basis function Operation.

[0020] Optionally, the S4 specifically includes:

[0021] The problem of minimum intra-class distance and maximum inter-class distance is transformed into the problem of maximizing Rayleigh quotient of multivariable function data. in, is the between-class scatter matrix, is the intra-class scatter matrix, p is the number of categories of the data processed, μ is the category center of one of the categories of data, N i is the number of samples in the i-th category, x i,j (t) is the jth multivariate function in the i-th class sample. The ω corresponding to the maximum Rayleigh quotient is the high-dimensional function-type projection hyperplane that minimizes the intra-class distance and maximizes the inter-class distance.

[0022] Since the multivariate functional data is linearly separable after being projected into the high-dimensional feature space, the high-dimensional functional projection hyperplane ω is converted into a linear representation of the high-dimensional multivariate functional data. Among them, φ(x)=[φ1(x),φ2(x),...,φ I (x)] T Represents the projection relationship from low dimension to high dimension, α=[α1,α2,...,α I ] T is the I×1-dimensional projection coefficient vector, I is the total number of samples, and the projection of φ(x) to the functional projection hyperplane ω in the high-dimensional space is expressed as φ(x)ω=φ(x)φ(x) T α=k er α, so solving the ω corresponding to the maximum Rayleigh quotient is transformed into solving the α problem;

[0023] Using the transformation form of the high-dimensional function-type projection hyperplane, and using the function-type data kernel function to transform the inner product of the multivariable function-type data in the high-dimensional form, the Rayleigh quotient of the function-type data is transformed into in,

[0024] is the jth sample of the kth class, k er (x i (t),x j (t)) is the kernel function, K i I×N i The matrix of For each element For K i The element K in row a and column b in i [a][b]=k er (x a (t),x b i (t));

[0025] The solution of Rayleigh quotient is further transformed into solving the optimization problem At this time, the Lagrange multiplier method is used, and the derivative of α can be obtained to obtain Mα=δNα. After solving the eigenvalue δ and the eigenvector α, a functional projection hyperplane is obtained that can distinguish whether the decarburized layer is qualified or not to the greatest extent.

[0026] On the other hand, a cord steel decarburization layer quality prediction system is provided, the system comprising:

[0027] An acquisition module is used to acquire sample data in the production process of cord steel, wherein the sample data includes scalar data and time series data, and the scalar data is stretched to obtain multivariate time series data;

[0028] A fitting module, used for fitting the multivariate time series data into multivariate function data by using B-spline basis function fitting;

[0029] A mapping module, used for mapping the multivariate functional data to a high-dimensional functional feature space using a functional data kernel method;

[0030] A calculation module is used to calculate the functional data projection hyperplane that minimizes the intra-class distance and maximizes the inter-class distance based on the functional features in high dimensions;

[0031] The projection module is used to project each sample to obtain the final sample classification result.

[0032] Optionally, the acquisition module is specifically used to:

[0033] The scalar data is converted into a constant function whose domain size is the heating furnace's in-furnace time, and is concatenated with the time series data in the variable dimension to obtain the multivariate time series data.

[0034] Optionally, the fitting module is specifically used to:

[0035] For functional data fitting, the function x obtained after fitting can be expressed as the inner product of the coefficient vector and the basis function: Where c is the coefficient vector, T represents the matrix transpose, is the basis function;

[0036] The coefficient vector c is determined by the least square method. At this time, the weighted fitting error term SMSSE(y i,j |c i,j )=(y i,j -Ψ j c i,j ) T W i,j (y i,j -Ψ j c i,j ),in, is the basis function numerical matrix of the jth variable, D j is the number of basis functions used when fitting functional data for the jth variable, is the lth basis function used when fitting the jth variable function data, L j is the length of observation time of the jth variable, W i,j is the jth curve weight matrix of the i-th sample, which is defined as the inverse of the covariance matrix calculated from the fitted residuals of all observation points of the curve of the j-th variable of the i-th sample;

[0037] Functional data fitting based on the least squares method will lose a certain degree of smoothness, so it is necessary to introduce a roughness penalty term to obtain the least squares fitting error PENSSE with a penalty term. m (y i,j |c i,j )=(y i,j -Ψ j c i,j ) T W i,j (y i,j -Ψ j c i,j )+λc i,j T R j c i,j , where R jis a weight matrix that reflects the relationship between variables. The coefficient vector c is solved using the least squares method. i,j =(Ψ j T W i,j Ψ j +λR j ) -1 Ψ j T W i,j y i,j , where λ is the smoothness coefficient.

[0038] Optionally, the mapping module is specifically used to:

[0039] Select the functional Gaussian radial basis kernel function k er , the multivariable functional data is projected into a high-dimensional space through a high-dimensional mapping, thereby converting the linearly inseparable low-dimensional multivariable functional data into linearly separable multivariable functional data in high dimensions. The inner product of the multivariable functional data in high dimensions can be expressed as k er (x1(t),x2(t))=exp(||x1(t)-x2(t)|| 2 / S 2 ), where S is the parameter of the kernel function, ||x1(t)-x2(t)|| 2 It is expressed as:

[0040]

[0041] Therefore, the final functional kernel function k er It can be expressed as a basis function Operation.

[0042] Optionally, the calculation module is specifically used to:

[0043] The problem of minimum intra-class distance and maximum inter-class distance is transformed into the problem of maximizing Rayleigh quotient of multivariable function data. in, is the between-class scatter matrix, is the intra-class scatter matrix, p is the number of categories of the data processed, μ is the category center of one of the categories of data, N i is the number of samples in the i-th category, x i,j (t) is the jth multivariate function in the i-th class sample. The ω corresponding to the maximum Rayleigh quotient is the high-dimensional function-type projection hyperplane that minimizes the intra-class distance and maximizes the inter-class distance.

[0044] Since the multivariate functional data is linearly separable after being projected into the high-dimensional feature space, the high-dimensional functional projection hyperplane ω is converted into a linear representation of the high-dimensional multivariate functional data. Among them, φ(x)=[φ1(x),φ2(x),...,φ I (x)] T Represents the projection relationship from low dimension to high dimension, α=[α1,α2,...,α I ] T is the I×1-dimensional projection coefficient vector, I is the total number of samples, and the projection of φ(x) to the functional projection hyperplane ω in the high-dimensional space is expressed as φ(x)ω=φ(x)φ(x) T α=k er α, so solving the ω corresponding to the maximum Rayleigh quotient is transformed into solving the α problem;

[0045] Using the transformation form of the high-dimensional function-type projection hyperplane, and using the function-type data kernel function to transform the inner product of the multivariable function-type data in the high-dimensional form, the Rayleigh quotient of the function-type data is transformed into in,

[0046] is the jth sample of the kth class, k er (x i (t),x j (t)) is the kernel function, K i I×N i The matrix of For each element For K i The element K in row a and column b in i [a][b]=k er (x a (t),x b i (t));

[0047] The solution of Rayleigh quotient is further transformed into solving the optimization problem At this time, the Lagrange multiplier method is used, and the derivative of α can be obtained to obtain Mα=δNα. After solving the eigenvalue δ and the eigenvector α, a functional projection hyperplane is obtained that can distinguish whether the decarburized layer is qualified or not to the greatest extent.

[0048] On the other hand, an electronic device is provided, comprising a processor and a memory, wherein the memory stores at least one instruction, and the at least one instruction is loaded and executed by the processor to implement the above-mentioned method for predicting the quality of the decarburized layer of cord steel.

[0049] On the other hand, a computer-readable storage medium is provided, wherein at least one instruction is stored in the storage medium, and the at least one instruction is loaded and executed by a processor to implement the above-mentioned method for predicting the quality of the decarburized layer of cord steel.

[0050] The beneficial effects brought about by the technical solution provided by the present invention include at least:

[0051] By adopting the present invention, the quality of the decarburized layer of the cord steel can be accurately predicted in real time. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0053] Figure 1 It is a flow chart of a method for predicting the quality of a decarburized layer of a cord steel provided by an embodiment of the present invention;

[0054] Figure 2 It is a schematic diagram of a method for stretching scalar data provided by an embodiment of the present invention;

[0055] Figure 3 This is a block diagram of a cord steel decarburization layer quality prediction system provided by an embodiment of the present invention;

[0056] Figure 4 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0058] The embodiment of the present invention provides a method for predicting the quality of the decarburized layer of a cord steel, which can be implemented by an electronic device, and the electronic device can be a terminal or a server. Figure 1 A flow chart of a method for predicting the quality of a cord steel decarburization layer is shown, and the processing flow of the method may include the following steps:

[0059] S1. Obtain sample data in the production process of cord steel, wherein the sample data includes scalar data and time series data, and stretch the scalar data to obtain multivariate time series data;

[0060] The production process of cord steel generally includes: converter-refining-continuous casting-heating furnace-rolling process, the scalar data includes the composition analysis data of the converter, refining and continuous casting processes, and the time series data includes the process control data of continuous casting, heating furnace and rolling.

[0061] Optionally, the step of stretching the scalar data in S1 to obtain multivariate time series data specifically includes:

[0062] The scalar data is converted into a constant function whose domain size is the heating furnace time (such as Figure 2 As shown, the scalar data C content is taken as a function with a constant value, that is, a straight line whose length is the furnace time), and is spliced ​​with the time series data in the variable dimension to obtain the multivariate time series data.

[0063] In the embodiment of the present invention, the multivariate time series data is represented in a matrix form, where each row of the matrix represents a sample, and each column of the matrix represents each univariate time series data.

[0064] S2, fitting the multivariate time series data into multivariate function data using B-spline basis function fitting;

[0065] Optionally, the S2 specifically includes:

[0066] For functional data fitting, the function x obtained after fitting can be expressed as the inner product of the coefficient vector and the basis function: Where c is the coefficient vector, T represents the matrix transpose, is a basis function (for example, the basis function selects a 5th-order B-spline basis function with strong fitting ability);

[0067] The coefficient vector c is determined by the least square method. At this time, the weighted fitting error term SMSSE(y i,j |c i,j )=(y i,j -Ψ j c i,j ) T W i,j (y i,j -Ψ j c i,j ),in, is the basis function numerical matrix of the jth variable, D j is the number of basis functions used when fitting functional data for the jth variable, is the lth basis function used when fitting the jth variable function data, L j is the length of observation time of the jth variable, W i,j is the jth curve weight matrix of the i-th sample, which is defined as the inverse of the covariance matrix calculated from the fitted residuals of all observation points of the curve of the j-th variable of the i-th sample;

[0068] Functional data fitting based on the least squares method will lose a certain degree of smoothness, so it is necessary to introduce a roughness penalty term to obtain the least squares fitting error PENSSE with a penalty term. m (y i,j |ci,j )=(y i,j -Ψ j c i,j ) T W i,j (y i,j -Ψ j c i,j )+λc i,j T R j c i,j , where R j is a weight matrix that reflects the relationship between variables. The coefficient vector c is solved using the least squares method. i,j =(Ψ j T W i,j Ψ j +λR j ) -1 Ψ j T W i,j y i,j , where λ is the smoothness coefficient.

[0069] The embodiment of the present invention converts each numerical element of the matrix into each functional element through the functional data fitting.

[0070] S3, projecting the multivariate functional data into a high-dimensional functional feature space using a functional data kernel method;

[0071] Optionally, the S3 specifically includes:

[0072] Select the functional Gaussian radial basis kernel function k er , the multivariable functional data is projected into a high-dimensional space through a high-dimensional mapping, thereby converting the linearly inseparable low-dimensional multivariable functional data into linearly separable multivariable functional data in high dimensions. The inner product of the multivariable functional data in high dimensions can be expressed as k er (x1(t),x2(t))=exp(||x1(t)-x2(t)|| 2 / S 2 ), where S is the parameter of the kernel function, ||x1(t)-x2(t)|| 2 It is expressed as:

[0073]

[0074] Therefore, the final functional kernel function k er It can be expressed as a basis function Operation.

[0075] S4. Based on the functional features in high dimensions, calculate the functional data projection hyperplane that minimizes the intra-class distance and maximizes the inter-class distance;

[0076] Optionally, the S4 specifically includes:

[0077] The problem of minimum intra-class distance and maximum inter-class distance is transformed into the problem of maximizing Rayleigh quotient of multivariable function data. in, is the between-class scatter matrix, is the intra-class scatter matrix, p is the number of categories of the processed data (the number of categories in the embodiment of the present invention is 2, qualified and unqualified), μ is the category center of one category of data, N i is the number of samples in the i-th category, x i,j (t) is the jth multivariate function in the i-th class sample. The ω corresponding to the maximum Rayleigh quotient is the high-dimensional function-type projection hyperplane that minimizes the intra-class distance and maximizes the inter-class distance.

[0078] Since the multivariate functional data is linearly separable after being projected into the high-dimensional feature space, the high-dimensional functional projection hyperplane ω is converted into a linear representation of the high-dimensional multivariate functional data. Among them, φ(x)=[φ1(x),φ2(x),...,φ I (x)] T Represents the projection relationship from low dimension to high dimension, α=[α1,α2,...,α I ] T is the I×1-dimensional projection coefficient vector, I is the total number of samples, and the projection of φ(x) to the functional projection hyperplane ω in the high-dimensional space is expressed as φ(x)ω=φ(x)φ(x) T α=k er α, so solving the ω corresponding to the maximum Rayleigh quotient is transformed into solving the α problem;

[0079] Using the transformation form of the high-dimensional function-type projection hyperplane, and using the function-type data kernel function to transform the inner product of the multivariable function-type data in the high-dimensional form, the Rayleigh quotient of the function-type data is transformed into in,

[0080] is the jth sample of the kth class, k er (x i (t),x j (t)) is the kernel function, K i I×N i The matrix of For each element For K iThe element K in row a and column b in i [a][b]=k er (x a (t),x b i (t));

[0081] The solution of Rayleigh quotient is further transformed into solving the optimization problem At this time, the Lagrange multiplier method is used, and the derivative of α can be obtained to obtain Mα=δNα. After solving the eigenvalue δ and the eigenvector α, a functional projection hyperplane is obtained that can distinguish whether the decarburized layer is qualified or not to the greatest extent.

[0082] S5. Project each sample to obtain the final sample classification result.

[0083] According to the high-dimensional function-type projection hyperplane form, each sample is projected, and its Euclidean distance with the center of each category is calculated, and the category with the smallest Euclidean distance is selected as the final category to which the sample belongs.

[0084] like Figure 3 As shown, an embodiment of the present invention further provides a cord steel decarburization layer quality prediction system, the system comprising:

[0085] An acquisition module 310 is used to acquire sample data in the production process of the cord steel, wherein the sample data includes scalar data and time series data, and the scalar data is stretched to obtain multivariate time series data;

[0086] A fitting module 320 is used to fit the multivariate time series data into multivariate function data by using B-spline basis function fitting;

[0087] A mapping module 330, used to map the multivariate functional data to a high-dimensional functional feature space using a functional data kernel method;

[0088] A calculation module 340 is used to calculate a functional data projection hyperplane that minimizes the intra-class distance and maximizes the inter-class distance based on the functional features in high dimensions;

[0089] The projection module 350 is used to project each sample to obtain a final sample classification result.

[0090] Optionally, the acquisition module is specifically used to:

[0091] The scalar data is converted into a constant function whose domain size is the heating furnace's in-furnace time, and is concatenated with the time series data in the variable dimension to obtain the multivariate time series data.

[0092] Optionally, the fitting module is specifically used to:

[0093] For functional data fitting, the function x obtained after fitting can be expressed as the inner product of the coefficient vector and the basis function: Where c is the coefficient vector, T represents the matrix transpose, is the basis function;

[0094] The coefficient vector c is determined by the least square method. At this time, the weighted fitting error term SMSSE(y i,j |c i,j )=(y i,j -Ψ j c i,j ) T W i,j (y i,j -Ψ j c i,j ),in, is the basis function numerical matrix of the jth variable, D j is the number of basis functions used when fitting functional data for the jth variable, is the lth basis function used when fitting the jth variable function data, L j is the length of observation time of the jth variable, W i,j is the jth curve weight matrix of the i-th sample, which is defined as the inverse of the covariance matrix calculated from the fitted residuals of all observation points of the curve of the j-th variable of the i-th sample;

[0095] Functional data fitting based on the least squares method will lose a certain degree of smoothness, so it is necessary to introduce a roughness penalty term to obtain the least squares fitting error PENSSE with a penalty term. m (y i,j |c i,j )=(y i,j -Ψ j c i,j ) T W i,j (y i,j -Ψ j c i,j )+λc i,j T R j c i,j , where R j is a weight matrix that reflects the relationship between variables. The coefficient vector c is solved using the least squares method. i,j =(Ψ j T W i,j Ψ j +λR j ) -1 Ψ jT W i,j y i,j , where λ is the smoothness coefficient.

[0096] Optionally, the mapping module is specifically used to:

[0097] Select the functional Gaussian radial basis kernel function k er , the multivariable functional data is projected into a high-dimensional space through a high-dimensional mapping, thereby converting the linearly inseparable low-dimensional multivariable functional data into linearly separable multivariable functional data in high dimensions. The inner product of the multivariable functional data in high dimensions can be expressed as k er (x1(t),x2(t))=exp(||x1(t)-x2(t)|| 2 / S 2 ), where S is the parameter of the kernel function, ||x1(t)-x2(t)|| 2 It is expressed as:

[0098]

[0099] Therefore, the final functional kernel function k er It can be expressed as a basis function Operation.

[0100] Optionally, the calculation module is specifically used to:

[0101] The problem of minimum intra-class distance and maximum inter-class distance is transformed into the problem of maximizing Rayleigh quotient of multivariable function data. in, is the between-class scatter matrix, is the intra-class scatter matrix, p is the number of categories of the data processed, μ is the category center of one of the categories of data, N i is the number of samples in the i-th category, x i,j (t) is the jth multivariate function in the i-th class sample. The ω corresponding to the maximum Rayleigh quotient is the high-dimensional function-type projection hyperplane that minimizes the intra-class distance and maximizes the inter-class distance.

[0102] Since the multivariate functional data is linearly separable after being projected into the high-dimensional feature space, the high-dimensional functional projection hyperplane ω is converted into a linear representation of the high-dimensional multivariate functional data. Among them, φ(x)=[φ1(x),φ2(x),...,φ I (x)] T Represents the projection relationship from low dimension to high dimension, α=[α1,α2,...,α I ] Tis the I×1-dimensional projection coefficient vector, I is the total number of samples, and the projection of φ(x) to the functional projection hyperplane ω in the high-dimensional space is expressed as φ(x)ω=φ(x)φ(x) T α=k er α, so solving the ω corresponding to the maximum Rayleigh quotient is transformed into solving the α problem;

[0103] Using the transformation form of the high-dimensional function-type projection hyperplane, and using the function-type data kernel function to transform the inner product of the multivariable function-type data in the high-dimensional form, the Rayleigh quotient of the function-type data is transformed into in,

[0104] is the jth sample of the kth class, k er (x i (t),x j (t)) is the kernel function, K i I×N i The matrix of For each element For K i The element K in row a and column b in i [a][b]=k er (x a (t),x b i (t));

[0105] The solution of Rayleigh quotient is further transformed into solving the optimization problem At this time, the Lagrange multiplier method is used, and the derivative of α can be obtained to obtain Mα=δNα. After solving the eigenvalue δ and the eigenvector α, a functional projection hyperplane is obtained that can distinguish whether the decarburized layer is qualified or not to the greatest extent.

[0106] The functional structure of a cord steel decarburization layer quality prediction system provided in an embodiment of the present invention corresponds to the cord steel decarburization layer quality prediction method provided in an embodiment of the present invention, which will not be described in detail here.

[0107] Figure 4 It is a structural schematic diagram of an electronic device 400 provided in an embodiment of the present invention. The electronic device 400 may have relatively large differences due to different configurations or performances, and may include one or more processors (centra lprocessing in its, CPU) 401 and one or more memories 402, wherein at least one instruction is stored in the memory 402, and the at least one instruction is loaded and executed by the processor 401 to implement the steps of the above-mentioned cord steel decarburization layer quality prediction method.

[0108] In an exemplary embodiment, a computer-readable storage medium is also provided, such as a memory including instructions, which can be executed by a processor in a terminal to complete the above-mentioned cord steel decarburization layer quality prediction method. For example, the computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0109] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0110] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for predicting the quality of a decarburized layer of a cord steel, characterized in that: The method comprises: S1. Obtain sample data in the production process of cord steel, wherein the sample data includes scalar data and time series data, and stretch the scalar data to obtain multivariate time series data; S2, fitting the multivariate time series data into multivariate function data using B-spline basis function fitting; S3, projecting the multivariate functional data into a high-dimensional functional feature space using a functional data kernel method; S4. Based on the functional features in high dimensions, calculate the functional data projection hyperplane that minimizes the intra-class distance and maximizes the inter-class distance; S5. Project each sample to obtain the final sample classification result.

2. The method according to claim 1, characterized in that In S1, the scalar data is stretched to obtain multivariate time series data, which specifically includes: The scalar data is converted into a constant function whose domain size is the heating furnace's in-furnace time, and is concatenated with the time series data in the variable dimension to obtain the multivariate time series data.

3. The method according to claim 1, characterized in that The S2 specifically includes: For functional data fitting, the function x obtained after fitting can be expressed as the inner product of the coefficient vector and the basis function: Where c is the coefficient vector, T represents the matrix transpose, is the basis function; The coefficient vector c is determined by the least square method. At this time, the weighted fitting error term SMSSE(y i,j |c i,j )=(y i,j -Ψ j c i,j ) T W i,j (y i,j -Ψ j c i,j ),in, is the basis function numerical matrix of the jth variable, D j is the number of basis functions used when fitting functional data for the jth variable, is the lth basis function used when fitting the jth variable function data, L j is the length of observation time of the jth variable, W i,j is the jth curve weight matrix of the i-th sample, which is defined as the inverse of the covariance matrix calculated from the fitted residuals of all observation points of the curve of the j-th variable of the i-th sample; Functional data fitting based on the least squares method will lose a certain degree of smoothness, so it is necessary to introduce a roughness penalty term to obtain the least squares fitting error PENSSE with a penalty term. m (y i,j |c i,j )=(y i,j -Ψ j c i,j ) T W i,j (y i,j -Ψ j c i,j )+λc i,j T R j c i,j , where R j is a weight matrix that reflects the relationship between variables, and the coefficient vector is solved using the least squares method. Where λ is the smoothness coefficient.

4. The method according to claim 3, characterized in that The S3 specifically includes: Select the functional Gaussian radial basis kernel function k er , the multivariable functional data is projected into a high-dimensional space through a high-dimensional mapping, thereby converting the linearly inseparable low-dimensional multivariable functional data into linearly separable multivariable functional data in high dimensions. The inner product of the multivariable functional data in high dimensions can be expressed as k er (x1(t),x2(t))=exp(||x1(t)-x2(t)|| 2 / S 2 ), where S is the parameter of the kernel function, ||x1(t)-x2(t)|| 2 It is expressed as: Therefore, the final functional kernel function k er It can be expressed as a basis function Operation.

5. The method according to claim 4, characterized in that The S4 specifically includes: The problem of minimum intra-class distance and maximum inter-class distance is transformed into the problem of maximizing Rayleigh quotient of multivariable function data. in, is the between-class scatter matrix, is the intra-class scatter matrix, p is the number of categories of the data processed, μ is the category center of one of the categories of data, N i is the number of samples in the i-th category, x i,j (t) is the jth multivariate function in the i-th class sample. The ω corresponding to the maximum Rayleigh quotient is the high-dimensional function-type projection hyperplane that minimizes the intra-class distance and maximizes the inter-class distance. Since the multivariate functional data is linearly separable after being projected into the high-dimensional feature space, the high-dimensional functional projection hyperplane ω is converted into a linear representation of the high-dimensional multivariate functional data. Among them, φ(x)=[φ1(x),φ2(x),...,φ I (x)] T Represents the projection relationship from low dimension to high dimension, α=[α1,α2,...,α I ] T is the I×1-dimensional projection coefficient vector, I is the total number of samples, and the projection of φ(x) to the functional projection hyperplane ω in the high-dimensional space is expressed as φ(x)ω=φ(x)φ(x) T α=k er α, so solving the ω corresponding to the maximum Rayleigh quotient is transformed into solving the α problem; Using the transformation form of the high-dimensional function-type projection hyperplane, and using the function-type data kernel function to transform the inner product of the multivariable function-type data in the high-dimensional form, the Rayleigh quotient of the function-type data is transformed into in, is the jth sample of the kth class, k er (x i (t),x j (t)) is the kernel function, K i I×N i The matrix of For each element For K i The element K in row a and column b in i [a][b]=k er (x a (t),x b i (t)); The solution of Rayleigh quotient is further transformed into solving the optimization problem At this time, the Lagrange multiplier method is used, and the derivative of α can be obtained to obtain Mα=δNα. After solving the eigenvalue δ and the eigenvector α, a functional projection hyperplane is obtained that can distinguish whether the decarburized layer is qualified or not to the greatest extent.

6. A cord steel decarburization layer quality prediction system, characterized in that: The system comprises: An acquisition module is used to acquire sample data in the production process of cord steel, wherein the sample data includes scalar data and time series data, and the scalar data is stretched to obtain multivariate time series data; A fitting module, used for fitting the multivariate time series data into multivariate function data by using B-spline basis function fitting; A mapping module, used for mapping the multivariate functional data to a high-dimensional functional feature space using a functional data kernel method; A calculation module is used to calculate the functional data projection hyperplane that minimizes the intra-class distance and maximizes the inter-class distance based on the functional features in high dimensions; The projection module is used to project each sample to obtain the final sample classification result.

7. The system according to claim 6, characterized in that The acquisition module is specifically used for: The scalar data is converted into a constant function whose domain size is the heating furnace's in-furnace time, and is concatenated with the time series data in the variable dimension to obtain the multivariate time series data.

8. The system according to claim 6, characterized in that The fitting module is specifically used for: For functional data fitting, the function x obtained after fitting can be expressed as the inner product of the coefficient vector and the basis function: Where c is the coefficient vector, T represents the matrix transpose, is the basis function; The coefficient vector c is determined by the least square method. At this time, the weighted fitting error term SMSSE(y i,j |c i,j )=(y i,j -Ψ j c i,j ) T W i,j (y i,j -Ψ j c i,j ),in, is the basis function numerical matrix of the jth variable, D j is the number of basis functions used when fitting functional data for the jth variable, is the lth basis function used when fitting the jth variable function data, L j is the length of observation time of the jth variable, W i,j is the jth curve weight matrix of the i-th sample, which is defined as the inverse of the covariance matrix calculated from the fitted residuals of all observation points of the curve of the j-th variable of the i-th sample; Functional data fitting based on the least squares method will lose a certain degree of smoothness, so it is necessary to introduce a roughness penalty term to obtain the least squares fitting error PENSSE with a penalty term. m (y i,j |c i,j )=(y i,j -Ψ j c i,j ) T W i,j (y i,j -Ψ j c i,j )+λc i,j T R j c i,j , where R j is a weight matrix that reflects the relationship between variables, and the coefficient vector is solved using the least squares method. Where λ is the smoothness coefficient.

9. The system according to claim 8, characterized in that The mapping module is specifically used for: Select the functional Gaussian radial basis kernel function k er , the multivariable functional data is projected into a high-dimensional space through a high-dimensional mapping, thereby converting the linearly inseparable low-dimensional multivariable functional data into linearly separable multivariable functional data in high dimensions. The inner product of the multivariable functional data in high dimensions can be expressed as k er (x1(t),x2(t))=exp(||x1(t)-x2(t)|| 2 / S 2 ), where S is the parameter of the kernel function, ||x1(t)-x2(t)|| 2 It is expressed as: Therefore, the final functional kernel function k er It can be expressed as a basis function Operation.

10. The system according to claim 9, characterized in that The computing module is specifically used for: The problem of minimum intra-class distance and maximum inter-class distance is transformed into the problem of maximizing Rayleigh quotient of multivariable function data. in, is the between-class scatter matrix, is the intra-class scatter matrix, p is the number of categories of the data processed, μ is the category center of one of the categories of data, N i is the number of samples in the i-th category, x i,j (t) is the jth multivariate function in the i-th class sample. The ω corresponding to the maximum Rayleigh quotient is the high-dimensional function-type projection hyperplane that minimizes the intra-class distance and maximizes the inter-class distance. Since the multivariate functional data is linearly separable after being projected into the high-dimensional feature space, the high-dimensional functional projection hyperplane ω is converted into a linear representation of the high-dimensional multivariate functional data. Among them, φ(x)=[φ1(x),φ2(x),...,φ I (x)] T Represents the projection relationship from low dimension to high dimension, α=[α1,α2,...,α I ] T is the I×1-dimensional projection coefficient vector, I is the total number of samples, and the projection of φ(x) to the functional projection hyperplane ω in the high-dimensional space is expressed as φ(x)ω=φ(x)φ(x) T α=k er α, so solving the ω corresponding to the maximum Rayleigh quotient is transformed into solving the α problem; Using the transformation form of the high-dimensional function-type projection hyperplane, and using the function-type data kernel function to transform the inner product of the multivariable function-type data in the high-dimensional form, the Rayleigh quotient of the function-type data is transformed into in, is the jth sample of the kth class, k er (x i (t),x j (t)) is the kernel function, K i I×N i The matrix of For each element For K i The element K in row a and column b in i [a][b]=k er (x a (t),x b i (t)); The solution of Rayleigh quotient is further transformed into solving the optimization problem At this time, the Lagrange multiplier method is used, and the derivative of α can be obtained to obtain Mα=δNα. After solving the eigenvalue δ and the eigenvector α, a functional projection hyperplane is obtained that can distinguish whether the decarburized layer is qualified or not to the greatest extent.

Images