Seamless steel tube production monitoring and abnormity tracing method based on Stacking fusion multiple models

Abstract

The invention discloses a seamless steel tube production monitoring and abnormity tracing method based on Stacking fusion multiple models, and belongs to the technical field of mechanical automation control, and the method comprises the steps: obtaining production process data in a historical production process; screening out excellent sample data; preprocessing the excellent sample data, and performing dimension reduction processing on the excellent sample data by adopting a dimension reduction algorithm based on Stacking fusion multiple models to obtain a corresponding feature matrix; collecting production process data in the current production process in real time; preprocessing the real-time data, and performing dimension reduction processing on the real-time data by adopting a dimension reduction algorithm based on Stacking fusion multiple models to obtain a corresponding feature matrix; based on this, whether the current production process is abnormal or not is judged, and abnormity tracing is carried out when abnormity occurs. On-line monitoring and abnormity tracing of seamless steel pipe production can be realized, and the production efficiency and the product qualification rate of the seamless steel pipe are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of mechanical automation control, and in particular to a seamless steel pipe production monitoring and abnormality tracing method based on Stacking fusion multi-model. Background Art

[0002] As an extremely commonly used special steel product, the quality monitoring algorithms for each process of seamless steel pipe are gradually being paid attention to by major seamless steel pipe manufacturers.

[0003] The continuous rolling production process of seamless steel pipes has a great impact on the size and quality of the final product. At the same time, the continuous rolling production process is very complicated and has various management and control problems that have not yet been solved. In the actual on-site production process, manual sampling and experience judgment methods are often used to monitor the quality of seamless steel pipes and analyze abnormalities. A large amount of available production data has not been effectively utilized, the quality control efficiency is low, and most quality anomalies have not been effectively handled. Therefore, there is an urgent need for an automated intelligent monitoring system that uses production data to achieve real-time monitoring, alarms and analysis of specific processes to improve production efficiency and product quality. Summary of the invention

[0004] The present invention provides a seamless steel pipe production monitoring and abnormality tracing method based on Stacking fusion multi-model, so as to solve the technical problem of low quality control efficiency in the traditional scheme of monitoring and abnormality analysis of the quality of seamless steel pipes by manual sampling and experience judgment.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] On the one hand, the present invention provides a method for seamless steel pipe production monitoring and abnormal tracing based on Stacking fusion multi-model, the method for seamless steel pipe production monitoring and abnormal tracing comprising:

[0007] Obtain production process data during the historical continuous rolling production of seamless steel pipes and record them as historical data;

[0008] The production process data corresponding to each seamless steel pipe is used as a sample data, and the sample data that meets the preset requirements is screened out from the historical data using the preset quality evaluation index, and recorded as excellent sample data;

[0009] Preprocessing the excellent sample data using a preset preprocessing algorithm to obtain an original matrix corresponding to the excellent sample data, and performing dimensionality reduction processing on the original matrix corresponding to the excellent sample data using a dimensionality reduction algorithm based on Stacking fusion multi-model to obtain a feature matrix corresponding to the excellent sample data;

[0010] Real-time collection of production process data of the seamless steel pipe during the current continuous rolling production process, recorded as real-time data;

[0011] The real-time data is preprocessed by using a preset preprocessing algorithm to obtain an original matrix corresponding to the real-time data, and the original matrix corresponding to the real-time data is reduced in dimension by using a dimensionality reduction algorithm based on Stacking fusion multi-model to obtain a feature matrix corresponding to the real-time data;

[0012] Based on the feature matrix corresponding to the real-time data and the feature matrix corresponding to the excellent sample data, it is determined whether the current continuous rolling production process of the seamless steel pipe is abnormal, and abnormality tracing is performed when an abnormality occurs.

[0013] Furthermore, the production process data include: rolling force, speed, torque and current of each roller of the six stands of the continuous rolling mill, mill inlet temperature and mandrel outlet linear speed.

[0014] Furthermore, the quality evaluation index includes: the head and tail wall thickness out-of-tolerance length ratio P ht , Maximum standard tolerance percentage of wall thickness after sawing P ot , Maximum eccentricity of wall thickness after sawing P ecc 、The proportion of head and tail outer diameter out-of-tolerance length P hd And the maximum standard tolerance percentage P of the outer diameter after sawing od ;in,

[0015]

[0016] Among them, l ht l is the length of the head wall thickness tolerance; et l is the length of the tail wall thickness tolerance; t is the total length of the seamless steel pipe;

[0017]

[0018] Among them, T i represents the wall thickness of all sampling points after sawing; T s Indicates the target wall thickness value; R t Indicates the maximum wall thickness error percentage coefficient allowed by production requirements;

[0019]

[0020] Among them, T max Indicates the maximum wall thickness on the sampling unit circle; T min Indicates the minimum wall thickness on the sampling unit circle;

[0021]

[0022] Among them, lhd Indicates the length of the head outer diameter out of tolerance; l ed Indicates the out-of-tolerance length of the tail outer diameter;

[0023]

[0024] Among them, D i Indicates the outer diameter of all sampling points after sawing; D s Indicates the target outer diameter value; R d Indicates the maximum percentage error coefficient of the outer diameter allowed by production requirements.

[0025] Furthermore, the preset quality evaluation index is used to select sample data that meets the preset requirements from the historical data and record them as excellent sample data, including:

[0026] The quality evaluation index of each seamless steel pipe is calculated according to the production process data corresponding to each seamless steel pipe, and the production process data corresponding to the seamless steel pipe whose quality evaluation index is within a preset range is used as excellent sample data.

[0027] Furthermore, the execution process of the preprocessing algorithm includes:

[0028] For the production process data to be processed, the dynamic time warping method is used to transform the data into a standard matrix that meets the statistical process control requirements, so that their lengths are standardized and equal.

[0029] Furthermore, the execution process of the dimensionality reduction algorithm based on Stacking fusion multi-model includes:

[0030] For the production process data to be reduced in dimension, a variety of different data processing algorithms are first used to perform dimensionality reduction processing on them respectively to obtain a variety of dimensionality reduction processing results; then the Stacking ensemble learning technology is used to integrate and fuse the dimensionality reduction processing results of a variety of different data processing algorithms to achieve the dimensionality reduction of the production process data to be reduced in dimension.

[0031] Furthermore, the multiple different data processing algorithms include: a kernel principal component analysis algorithm, a kernel partial least squares algorithm and a kernel entropy component analysis algorithm.

[0032] Furthermore, each data processing algorithm is optimized using the Adam algorithm.

[0033] Further, based on the feature matrix corresponding to the real-time data and the feature matrix corresponding to the excellent sample data, it is judged whether the current continuous rolling production process of the seamless steel pipe is abnormal, and abnormality tracing is performed when an abnormality occurs, including:

[0034] By using the Procrustes analysis algorithm, the dissimilarity between the characteristic matrix corresponding to the real-time data and the characteristic matrix corresponding to the excellent sample data is calculated, and the calculated dissimilarity is compared with a preset dissimilarity threshold value. When the calculated dissimilarity exceeds the preset dissimilarity threshold value, it is determined that the current continuous rolling production process is abnormal;

[0035] When it is determined that the current continuous rolling production process is abnormal, the production process data of the seamless steel pipe continuous rolling production process in which the abnormality occurs is collected and recorded as abnormal data;

[0036] The abnormal contribution rate of each variable in the abnormal data is calculated, and the single variable or multiple variables with the largest abnormal contribution rate are taken as the cause of the abnormality in the current seamless steel pipe continuous rolling production process to achieve abnormality tracing.

[0037] Furthermore, the formula for calculating the abnormal contribution rate of each variable in the abnormal data is:

[0038]

[0039] Among them, Contri j represents the abnormal contribution rate of the jth variable in the abnormal data; t i represents the principal component of the i-th feature matrix corresponding to the abnormal data; T represents the transpose of the matrix; N represents the number of principal components of the feature matrix corresponding to the abnormal data; x j represents the standardized data of the jth variable in the abnormal data; H(p)i represents the information entropy of the principal component of the i-th feature matrix corresponding to the abnormal data.

[0040] On the other hand, the present invention further provides an electronic device, comprising a processor and a memory; wherein the memory stores at least one instruction, and the instruction is loaded and executed by the processor to implement the above method.

[0041] In yet another aspect, the present invention further provides a computer-readable storage medium, wherein at least one instruction is stored in the storage medium, and the instruction is loaded and executed by a processor to implement the above method.

[0042] The beneficial effects brought about by the technical solution provided by the present invention include at least:

[0043] Compared with the traditional method of monitoring the quality of seamless steel pipes and analyzing abnormalities based on manual experience, the production data statistical process control processing method of the present invention has formulated a size judgment index, and uses Stacking fusion multi-model, Procrustes analysis algorithm and kernel space abnormality contribution algorithm to realize the functions of real-time monitoring, real-time alarm, real-time analysis and timely tracing of abnormalities in the continuous rolling production process of seamless steel pipes, which can solve the problem of low efficiency in traditional continuous rolling process monitoring and abnormality analysis. Experiments were conducted on multiple real cases to prove the actual effectiveness and field practicability of the algorithm, and to improve the production efficiency and product qualification rate of seamless steel pipes. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] 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.

[0045] Figure 1 It is a schematic diagram of the execution flow of the seamless steel pipe production monitoring and abnormality tracing method based on Stacking fusion multi-model provided in an embodiment of the present invention;

[0046] Figure 2 It is a system block diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0047] In order to make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0048] First of all, it should be noted that in the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "exemplary" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "exemplarily" is intended to present the concept in a concrete way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0049] First embodiment

[0050] This embodiment combines various difficulties in quality control of seamless steel pipe production process, establishes a state monitoring and abnormality tracing method for continuous rolling process, takes statistical process control as the core, and provides a seamless steel pipe production monitoring and abnormality tracing method based on Stacking fusion multi-model. The method can be implemented by an electronic device, which can be a terminal or a server. The execution process of the method is as follows: Figure 1 As shown, the following steps are included:

[0051] S1, obtaining the production process data of the historical continuous rolling production process of the seamless steel pipe, and recording it as historical data;

[0052] The production process data selected for monitoring in this embodiment include effective variables such as process variables of each roll of the six stands of the continuous rolling mill, mill inlet temperature, and mandrel state variables, and redundant or inefficient variables are removed. Specifically, the production process data selected for monitoring in this embodiment include: rolling force, speed, torque and current of each roll of the six stands of the continuous rolling mill, mill inlet temperature, and mandrel outlet linear speed.

[0053] S2, taking the production process data corresponding to each steel pipe as a sample data, using the preset quality evaluation index, screening out the sample data that meets the preset requirements from the historical data, and recording them as excellent sample data;

[0054] Among them, this embodiment designs quality evaluation indicators according to the actual production situation on site, including: the head and tail wall thickness out-of-tolerance length ratio P ht , Maximum standard tolerance percentage of wall thickness after sawing P ot , Maximum eccentricity of wall thickness after sawing P ecc 、The proportion of head and tail outer diameter out-of-tolerance length P hd And the maximum standard tolerance percentage P of the outer diameter after sawing od ;in,

[0055]

[0056] Among them, l ht l is the length of the head wall thickness tolerance; et l is the length of the tail wall thickness tolerance; t It is the total length of the seamless steel pipe; the wall thickness tolerance length is determined by the wall thickness requirement range of the production process.

[0057]

[0058] Among them, the sawing length is collected from the production record sheet; T i represents the wall thickness of all sampling points after sawing; T s Indicates the target wall thickness value; R t It indicates the maximum wall thickness error percentage coefficient allowed by production requirements, usually with a value of 0.1.

[0059]

[0060] Among them, T max Indicates the maximum wall thickness on the sampling unit circle; T minIt indicates the minimum wall thickness on the sampling unit circumference; the eccentricity after sawing represents the unevenness of wall thickness and eccentricity. A larger value indicates that the wall thickness of the steel pipe is unevenly distributed in some parts and the center of the circular hole is eccentric. The eccentricity after sawing can measure the uniformity of the wall thickness of the finished seamless steel pipe.

[0061]

[0062] Among them, l hd Indicates the length of the head outer diameter out of tolerance; l ed Indicates the out-of-tolerance length of the tail outer diameter; l t It is the total length of the seamless steel pipe; the outer diameter tolerance length is determined by the outer diameter requirement range of the production process.

[0063]

[0064] Among them, D i Indicates the outer diameter of all sampling points after sawing; D s Indicates the target outer diameter value; R d It indicates the maximum percentage coefficient of outer diameter error allowed by production requirements, usually with a value of 0.005-0.01.

[0065] Furthermore, the preset quality evaluation index is used to screen out sample data that meets the preset requirements from the historical data and record them as excellent sample data. Specifically, the quality evaluation index of each seamless steel pipe is calculated based on the production process data corresponding to each seamless steel pipe, and the production process data corresponding to the seamless steel pipe whose quality evaluation index is within the preset range is taken as excellent sample data, so as to construct an excellent sample library.

[0066] S3, preprocessing the excellent sample data using a preset preprocessing algorithm to obtain an original matrix corresponding to the excellent sample data, and performing dimensionality reduction processing on the original matrix corresponding to the excellent sample data using a dimensionality reduction algorithm based on Stacking fusion multi-models to obtain a feature matrix corresponding to the excellent sample data;

[0067] Furthermore, the execution process of the preprocessing algorithm includes: preprocessing the raw data of different lengths and time dislocations, and using the dynamic time warping method to transform the data into a standard matrix that meets the statistical process control, so that its length standard is equal. The specific calculation formula is as follows:

[0068] (1) For the collected data array X (n×m), the data standardization process is as follows:

[0069]

[0070] (2) Let the two time series to be calculated be X and Y, with lengths |X| and |Y| respectively. The format of the normalization path is W = w1,w2,...,w K , where Max(|X|,|Y|)<=K<=|X|+|Y|. k The form is (i,j), where i represents the i coordinate in X and j represents the j coordinate in Y. The normalization path W must start from w1=(1,1) and end at w K =(|X|,|Y|), where i and j in w(i,j) in W must be monotonically increasing:

[0071] w k =(i,j),w k+1 =(i',j')i≤i'≤i+1,j≤j'≤j+1

[0072] Find the shortest normalized path of the sequence trajectory, with a distance of D(|X|,|Y|):

[0073] D(i,j)=Dist(i,j)+min[D(i-1,j),D(i,j-1),D(i-1,j-1)]

[0074] Furthermore, the execution process of the dimensionality reduction algorithm based on Stacking fusion of multiple models is specifically as follows: for new regularized data, kernel principal component analysis, kernel partial least squares method, and kernel entropy component analysis are used to process them respectively, and then data analysis is performed based on Stacking integration and fusion of the above multi-layer models to achieve data dimensionality reduction.

[0075] (1) The kernel principal component analysis (KPCA) process is as follows:

[0076] First, the nonlinear mapping of the data is defined as θ:R N →F, R N Represents the sample space, F represents the high-dimensional mapping space, and the two are transformed by a nonlinear mapping function θ. Then the sample covariance matrix C in the F space is expressed as:

[0077]

[0078] Introducing the kernel function, according to its kernel function theory, we can get the kernel matrix K, and the expressions of each element in the kernel matrix are:

[0079] K ij =K(x i ,x j )=(Φ(x i )·Φ(x j ))

[0080] This scheme uses the Gaussian kernel function as the kernel matrix mapping function. The Gaussian kernel function mapping formula is:

[0081]

[0082] In the formula, K(x i ,x j ) is the element in the i-th row and j-th column of the kernel matrix, x i is the variable array of column i of the original data matrix, x j is the variable array of the jth column of the original data matrix, σ is the Gaussian kernel function coefficient, and the output is the kernel space data matrix.

[0083] (2) The process of KPLS algorithm is as follows:

[0084] Set nonlinear mapping function It can transform the sample x i (1,2,…,N) is mapped from the original variable space to the feature space F, that is:

[0085]

[0086] In the formula, the dimension m of the feature space F can be arbitrarily large or even infinite-dimensional. The matrix X is mapped to the matrix F in the feature space:

[0087]

[0088] KPLS requires zero mean treatment for φ. First, we need to find The average

[0089]

[0090] In the formula, I N =[1 1…1] T ∈R N , and then find the zero mean

[0091]

[0092] Therefore, nonlinear data (X,Y) is mapped to linear data Then, the PLS algorithm is used to construct KPLS model between and Y:

[0093]

[0094] Where Y∈R N×γ It is called the scoring matrix, P∈R M×γ ,yes The load matrix, Q∈R1×γ is the loading matrix of Y and Y r They are and the residual matrix of Y, γ is the number of latent variables.

[0095] (3) The process of KECA is as follows:

[0096] Using Renyi entropy as the information entropy function, the quadratic Renyi entropy formula is as follows:

[0097] H(P)=-log∫p 2 (x)dx

[0098] Substituting the kernel function into the quadratic Renyi entropy, the derivation is as follows:

[0099]

[0100] In the formula, I is an N×1 column vector (all elements are 1), and the kernel matrix K is subjected to eigenvalue decomposition, and the eigenvalues ​​λ in the formula are obtained. i and the eigenvector σ i , and finally the quadratic Renyi entropy of each dimension of the kernel matrix is ​​obtained. The dimension with extremely small information entropy has less information, mostly composed of useless information and noise, which is discarded to obtain the final kernel entropy principal component matrix.

[0101] (4) Establish a prediction model based on the Stacking algorithm to fuse multi-layer models and perform dimensionality reduction processing on the continuous rolling production process data. Among them, the training method of each model in the Stacking framework is uniformly optimized using the Adam algorithm, which uses the first-order moment estimation and second-order moment estimation of the gradient to dynamically adjust the learning rate of each parameter in the basic model. The process of Adam algorithm optimizing the basic model is as follows:

[0102]

[0103] θ=θ+Δθ

[0104] in, is the gradient, m is the number of samples, L is the objective function, θ is the initial weight of the basic model, s and r are the gradients respectively. The first-order moment estimate and the second-order moment estimate of and are the bias corrections for the first-order moment estimation and the second-order moment estimation, ρ1 is the first-order momentum decay coefficient, ρ2 is the second-order momentum decay coefficient, δ is the smoothing term, and Δθ is the change in θ.

[0105] The loss function L is calculated as follows:

[0106]

[0107] Where N is the number of training samples, and Y i are the predicted value and the true value.

[0108] The model adjusts the weights and bias vectors of the single model through back-propagation training, and finally obtains a comprehensive dimension reduction matrix. This model can be used to achieve dimensionality reduction of data.

[0109] S4, real-time collection of production process data of the seamless steel pipe in the current continuous rolling production process, recorded as real-time data;

[0110] S5, preprocessing the real-time data using a preset preprocessing algorithm to obtain an original matrix corresponding to the real-time data, and performing dimensionality reduction processing on the original matrix corresponding to the real-time data using a dimensionality reduction algorithm based on Stacking fusion multi-models to obtain a feature matrix corresponding to the real-time data;

[0111] S6, judging whether the current continuous rolling production process of the seamless steel pipe is abnormal based on the feature matrix corresponding to the real-time data and the feature matrix corresponding to the excellent sample data, and tracing the abnormality when an abnormality occurs;

[0112] Specifically, based on the feature matrix corresponding to the real-time data and the feature matrix corresponding to the excellent sample data, it is determined whether the current continuous rolling production process of the seamless steel pipe is abnormal, and abnormality tracing is performed when an abnormality occurs, including:

[0113] S61, calculating the dissimilarity between the feature matrix corresponding to the real-time data and the feature matrix corresponding to the excellent sample data through the Procrustes analysis algorithm, comparing the calculated dissimilarity with a preset dissimilarity threshold, and determining that the current continuous rolling production process is abnormal when the calculated dissimilarity exceeds the preset dissimilarity threshold;

[0114] The specific method for calculating the dissimilarity is as follows:

[0115] Taking two different batches of data matrices as an example, X and Y are two different data matrices, and the dissimilarity formula is derived as follows:

[0116] First move the centers of both matrices to the origin:

[0117]

[0118] In the formula, is the central matrix of X, is the centered matrix of Y.

[0119] Then normalize the two matrices to have the same size, that is, normalize their dissimilarity to 1:

[0120]

[0121] In the formula, is the normalized matrix of X, is the normalized matrix of Y.

[0122] Then find a rotation matrix R such that As much as possible Alignment. This rotation matrix is ​​calculated by singular value decomposition (SVD):

[0123]

[0124] R=UVT

[0125] Where U and V are orthogonal matrices, Σ is a diagonal matrix, and R is a rotation matrix.

[0126] Finally, calculate the dissimilarity between the aligned matrices:

[0127]

[0128] Where d(·) is the dissimilarity between matrices, and its range is [0,1]. The larger the dissimilarity index, the greater the difference between the two original data matrices X and Y.

[0129] Among them, when there are multiple excellent samples, the average of the dissimilarity between the feature matrix corresponding to the real-time data and the feature matrix corresponding to each excellent sample data can be used as the final dissimilarity value to realize production process monitoring.

[0130] S62, when it is determined that the current continuous rolling production process is abnormal, the production process data of the seamless steel pipe continuous rolling production process in which the abnormality occurs is collected and recorded as abnormal data;

[0131] S63, based on the effective production data in each production process, the abnormal contribution rate of each variable in the abnormal data is calculated by using the abnormal contribution rate method, and the single variable or multiple variables with the largest abnormal contribution rate are used as the cause of the abnormality of the current seamless steel pipe continuous rolling production process, so as to realize abnormality tracing;

[0132] Among them, the formula for calculating the abnormal contribution rate of each variable in the abnormal data is:

[0133]

[0134] Among them, Contri j represents the abnormal contribution rate of the jth monitoring parameter; t i represents the principal component of the i-th Stacking fusion matrix; T represents the transpose of the matrix; N represents the number of principal components of the Stacking fusion matrix; xj represents the standardized data of the jth monitoring parameter; H(p) i Represents the principal component information entropy of the i-th Stacking fusion matrix.

[0135] In summary, this embodiment provides a method for seamless steel pipe production monitoring and abnormality tracing based on Stacking fusion multi-models. It uses statistical process control, combined with the corresponding Stacking integrated fusion model and abnormal contribution rate algorithm, to achieve process monitoring and abnormality tracing of the steel pipe continuous rolling process, and ultimately ensure that the seamless steel pipe maintains normal production and achieves the goal of improving the yield rate. This data-driven method for seamless steel pipe production process monitoring and abnormality tracing is efficient and convenient, based on a large amount of actual production data, and can quickly adjust parameters, training sets, and models according to actual working conditions, so that it can adapt to the production scenarios of various working conditions and various steel types.

[0136] Second embodiment

[0137] This embodiment describes the implementation process of the seamless steel pipe production monitoring and abnormal tracing method based on Stacking fusion multi-model provided by the present invention in combination with a specific application scenario, which includes the following steps:

[0138] S1, select effective variables in the continuous rolling production process for monitoring.

[0139] This case uses the production data of a seamless steel pipe plant with the same specification and furnace number on a certain day (a total of 600 pieces) as an example, including the continuous rolling inlet temperature T0, the rolling force F of the six-stand three-roller rolling mill, and the rolling speed of the three-roller rolling mill. 1-1 ~F 6-1 、F 1-2 ~F 6-2 、F 1-3 ~F 6-3 , speed v 1-1 ~v 6-1 、v 1-2 ~v 6-2 、v 1-3 ~v 6-3 , torque M 1-1 ~M 6-1 、M 1-2 ~M 6-2 、M 1-3 ~M 6-3 , current I 1-1 ~I 6-1 ,I 1-2 ~I 6-2 ,I 1-3 ~I 6-3 , outlet linear velocity v s1 -v s6 , as shown in Table 1:

[0140] Table 1 Parameters of each sample on site

[0141]

[0142]

[0143] S2, based on the data obtained in S1, calculates the quality evaluation index according to the actual production data samples on site, selects samples with excellent quality, and builds an excellent sample database, as shown in Table 2;

[0144] Table 2 Quality evaluation indicators

[0145]

[0146] S3: preprocessing the raw data of some usage parameters;

[0147] The dynamic time warping method is used to transform the data into a standard matrix that satisfies statistical process control. The linear interpolation method is used to interpolate and extract the time series of parameters of different lengths, and the sampling length of each parameter is unified without destroying its original characteristics. The length sampling point allocation of the parameter is realized in a physical sense.

[0148] S4: Data dimension reduction is achieved through data analysis based on Stacking integrated fusion multi-layer model, and the dissimilarity of the reduced dimension matrix is ​​calculated through Procrustes analysis algorithm to realize process monitoring of continuous rolling production.

[0149] Among them, in this embodiment, according to the fault production record table, it is known that the 448th steel pipe has serious rolling problems, and the entire steel pipe has cracked during the continuous rolling process and is finally treated as scrap. However, after experiments, it was found that production anomalies occurred at the 348th steel pipe, as shown in Table 3. Because the first 100 steel pipes were used as the training set, and the number of training sets was added, it was exactly the 448th steel pipe, which proved that the production anomaly positioning was effective. The production data was retrieved and compared with the normal steel pipe, and the abnormal characteristics of the production data were obvious.

[0150] Table 3 Abnormal dissimilarity values

[0151]

[0152] S5: Utilize the effective production data in each production process, and analyze the monitoring parameters using the abnormal contribution rate method, calculate and analyze the single or several parameters that have the greatest impact on the abnormality, and realize the abnormal tracing of production abnormalities. Specifically, in this embodiment, based on the step in S4, the process monitoring finds that the abnormality occurs, and then extracts the abnormal sample data, and performs the abnormal contribution rate analysis. The results are shown in Table 4:

[0153] Table 4 Abnormal contribution rate analysis

[0154]

[0155]

[0156] It can be seen from Table 4 that the corresponding process parameter number shows that the abnormal contribution rate of the rolling mill torque is the largest, and the torque reflects the combined effect of rolling force and speed. According to historical data records, the rolling mill set speed was adjusted in subsequent work, and production returned to normal.

[0157] In summary, this case proves that the seamless steel pipe continuous rolling production process detection and anomaly tracing method based on the Stacking integrated fusion multi-layer model provided by the present invention can solve the problem of low efficiency existing in traditional continuous rolling process monitoring and anomaly analysis methods, and can realize online monitoring and anomaly tracing of the seamless steel pipe production process, which is of great significance to improving the production efficiency and product qualification rate of seamless steel pipes.

[0158] Third embodiment

[0159] This embodiment provides an electronic device, such as Figure 2 As shown, the electronic device includes: a processor and a memory; wherein the processor and the memory can be connected via a communication bus; the memory stores at least one instruction, and the instruction is loaded and executed by the processor to implement the method of the first embodiment. In addition, the electronic device may also include a transceiver, the processor and the transceiver can be connected via a communication bus, and the transceiver is used to communicate with other devices.

[0160] Next, combine Figure 2 The following is a detailed introduction to the various components of the electronic device:

[0161] Among them, the processor is the control center of the electronic device, and the electronic device may include multiple processors, each of which may be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). The processor here may be a processor or a general term for multiple processing elements. For example, the processor is one or more central processing units (CPUs), or other general-purpose processors, application specific integrated circuits (ASICs), or one or more integrated circuits configured to implement an embodiment of the present invention, such as one or more microprocessors (digital signal processors, DSPs), or one or more field programmable gate arrays (field programmable gate arrays, FPGAs), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor may execute various functions of the electronic device by running or executing software programs stored in the memory and calling data stored in the memory.

[0162] In a specific implementation, as an embodiment, the processor may include one or more CPUs, such as Figure 2 The CPU0 and CPU1 shown in the figure are, of course, only exemplary.

[0163] The memory is used to store the software program for executing the solution of the present invention, and the execution is controlled by the processor. The specific implementation method can refer to the above method embodiment and will not be repeated here.

[0164] Optionally, the memory may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory may be integrated with the processor or exist independently and accessed through the interface circuit ( Figure 2 (not shown) is coupled to the processor, which is not specifically limited in this embodiment of the present invention.

[0165] The transceiver may include a receiver and a transmitter ( Figure 2 The receiver is used to implement the receiving function, and the transmitter is used to implement the sending function. The transceiver can be integrated with the processor or exist independently and communicate with the electronic device through the interface circuit ( Figure 2 (not shown) is coupled to the processor, which is not specifically limited in this embodiment of the present invention.

[0166] In addition, it should be noted that Figure 2 The structure of the electronic device shown in the figure does not constitute a limitation on the device, and the actual device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. In addition, the technical effects achieved by the electronic device when executing the method of the first embodiment above can refer to the technical effects described in the first embodiment above, so they are not repeated here.

[0167] Fourth embodiment

[0168] This embodiment provides a computer-readable storage medium, which stores at least one instruction, and the instruction is loaded and executed by a processor to implement the method of the first embodiment. The computer-readable storage medium may be a ROM, a random access memory, a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc. The instructions stored therein may be loaded by a processor in a terminal to execute the method.

[0169] In addition, it should be noted that the present invention can be provided as a method, an apparatus or a computer program product. Therefore, the embodiment of the present invention can be in the form of a full or partial hardware embodiment, a full or partial software embodiment or an embodiment combining software and hardware. Moreover, when implemented using software, the embodiment of the present invention can be in the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program codes. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center containing one or more available media sets. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a DVD), or a semiconductor medium. The semiconductor medium may be a solid state hard disk.

[0170] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0171] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device, so that a series of operation steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable terminal device provide for implementing the process in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0172] It should also be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of more restrictions, the elements defined by the sentence "including one..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements. In addition, the term "and / or" is only an association relationship describing the associated objects, indicating that there can be three relationships, for example, A and / or B, which can represent: A exists alone, A and B exist at the same time, and B exists alone, wherein A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding. "At least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can be represented by: a, b, c, ab, ac, bc or abc, where a, b, c can be single or plural.

[0173] In addition, it can be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0174] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0175] In several embodiments provided by the present invention, it should be understood that the disclosed equipment, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of functional modules / units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point, the coupling or direct coupling or communication connection between each other shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms. The unit described as a separate component may or may not be physically separated, and the component displayed as a unit may or may not be a physical unit, that is, it may be located in one place, or it may be distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0176] If the method is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0177] Finally, it should be noted that the above is only a preferred embodiment of the present invention. It should be pointed out that although the preferred embodiment of the present invention has been described, for ordinary technicians in this technical field, once the basic creative concept of the present invention is known, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the protection scope of the present invention. Therefore, the attached claims are intended to be interpreted as including the preferred embodiment and all changes and modifications that fall within the scope of the embodiments of the present invention.

Claims

1. A seamless steel pipe production monitoring and abnormal tracing method based on Stacking fusion multi-model, characterized in that: The seamless steel pipe production monitoring and abnormal tracing method based on Stacking fusion multi-model includes: Obtain production process data during the historical continuous rolling production of seamless steel pipes and record them as historical data; The production process data corresponding to each seamless steel pipe is used as a sample data, and the sample data that meets the preset requirements is screened out from the historical data using the preset quality evaluation index, and recorded as excellent sample data; Preprocessing the excellent sample data using a preset preprocessing algorithm to obtain an original matrix corresponding to the excellent sample data, and performing dimensionality reduction processing on the original matrix corresponding to the excellent sample data using a dimensionality reduction algorithm based on Stacking fusion multi-model to obtain a feature matrix corresponding to the excellent sample data; Real-time collection of production process data of the seamless steel pipe during the current continuous rolling production process, recorded as real-time data; The real-time data is preprocessed by using a preset preprocessing algorithm to obtain an original matrix corresponding to the real-time data, and the original matrix corresponding to the real-time data is reduced in dimension by using a dimensionality reduction algorithm based on Stacking fusion multi-model to obtain a feature matrix corresponding to the real-time data; Based on the feature matrix corresponding to the real-time data and the feature matrix corresponding to the excellent sample data, it is determined whether the current continuous rolling production process of the seamless steel pipe is abnormal, and abnormality tracing is performed when an abnormality occurs.

2. The method for seamless steel pipe production monitoring and abnormal tracing based on Stacking fusion multi-model according to claim 1 is characterized in that: The production process data include: rolling force, speed, torque and current of each roller of the six stands of the continuous rolling mill, the inlet temperature of the rolling mill and the outlet linear speed of the mandrel.

3. The method for seamless steel pipe production monitoring and abnormal tracing based on Stacking fusion multi-model according to claim 1 is characterized in that: The quality evaluation indicators include: the head and tail wall thickness tolerance length ratio P ht , Maximum standard tolerance percentage of wall thickness after sawing P ot , Maximum eccentricity of wall thickness after sawing P ecc 、The proportion of head and tail outer diameter out-of-tolerance length P hd And the maximum standard tolerance percentage P of the outer diameter after sawing od ;in, Among them, l ht l is the length of the head wall thickness tolerance; et l is the length of the tail wall thickness tolerance; t is the total length of the seamless steel pipe; Among them, T i represents the wall thickness of all sampling points after sawing; T s Indicates the target wall thickness value; R t Indicates the maximum wall thickness error percentage coefficient allowed by production requirements; Among them, T max Indicates the maximum wall thickness on the sampling unit circle; T min Indicates the minimum wall thickness on the sampling unit circle; Among them, l hd Indicates the length of the head outer diameter out of tolerance; l ed Indicates the out-of-tolerance length of the tail outer diameter; Among them, D i Indicates the outer diameter of all sampling points after sawing; D s Indicates the target outer diameter value; R d Indicates the maximum percentage error coefficient of the outer diameter allowed by production requirements.

4. The method for seamless steel pipe production monitoring and abnormal tracing based on Stacking fusion multi-model as claimed in claim 3 is characterized in that: The method of using the preset quality evaluation index to select sample data that meets the preset requirements from the historical data and record them as excellent sample data includes: The quality evaluation index of each seamless steel pipe is calculated according to the production process data corresponding to each seamless steel pipe, and the production process data corresponding to the seamless steel pipe whose quality evaluation index is within a preset range is used as excellent sample data.

5. The method for seamless steel pipe production monitoring and abnormal tracing based on Stacking fusion multi-model according to claim 1 is characterized in that: The execution process of the preprocessing algorithm includes: For the production process data to be processed, the dynamic time warping method is used to transform the data into a standard matrix that meets the statistical process control requirements, so that their lengths are standardized and equal.

6. The method for seamless steel pipe production monitoring and abnormal tracing based on Stacking fusion multi-model according to claim 1 is characterized in that: The execution process of the dimensionality reduction algorithm based on Stacking fusion multi-model includes: For the production process data to be reduced in dimension, a variety of different data processing algorithms are first used to perform dimensionality reduction processing on them respectively to obtain a variety of dimensionality reduction processing results; then the Stacking ensemble learning technology is used to integrate and fuse the dimensionality reduction processing results of a variety of different data processing algorithms to achieve the dimensionality reduction of the production process data to be reduced in dimension.

7. The method for seamless steel pipe production monitoring and abnormal tracing based on Stacking fusion multi-model according to claim 6 is characterized in that: The multiple different data processing algorithms include: a kernel principal component analysis algorithm, a kernel partial least squares algorithm and a kernel entropy component analysis algorithm.

8. The method for seamless steel pipe production monitoring and abnormal tracing based on Stacking fusion multi-model according to claim 6 is characterized in that: Each data processing algorithm is optimized using the Adam algorithm.

9. The method for seamless steel pipe production monitoring and abnormal tracing based on Stacking fusion multi-model according to claim 1, characterized in that: Based on the feature matrix corresponding to the real-time data and the feature matrix corresponding to the excellent sample data, it is judged whether the current continuous rolling production process of the seamless steel pipe is abnormal, and abnormality tracing is performed when an abnormality occurs, including: By using the Procrustes analysis algorithm, the dissimilarity between the characteristic matrix corresponding to the real-time data and the characteristic matrix corresponding to the excellent sample data is calculated, and the calculated dissimilarity is compared with a preset dissimilarity threshold value. When the calculated dissimilarity exceeds the preset dissimilarity threshold value, it is determined that the current continuous rolling production process is abnormal; When it is determined that the current continuous rolling production process is abnormal, the production process data of the seamless steel pipe continuous rolling production process in which the abnormality occurs is collected and recorded as abnormal data; The abnormal contribution rate of each variable in the abnormal data is calculated, and the single variable or multiple variables with the largest abnormal contribution rate are taken as the cause of the abnormality in the current seamless steel pipe continuous rolling production process to achieve abnormality tracing.

10. The method for seamless steel pipe production monitoring and abnormality tracing based on Stacking fusion multi-model according to claim 9, characterized in that: The formula for calculating the abnormal contribution rate of each variable in the abnormal data is: Among them, Contri j represents the abnormal contribution rate of the jth variable in the abnormal data; t i represents the principal component of the i-th feature matrix corresponding to the abnormal data; T represents the transpose of the matrix; N represents the number of principal components of the feature matrix corresponding to the abnormal data; x j represents the standardized data of the jth variable in the abnormal data; H(p)i represents the information entropy of the principal component of the i-th feature matrix corresponding to the abnormal data.

Images