A method for predicting the mechanical properties after steel pipe production based on machine learning

Abstract

The present invention discloses a method for predicting the mechanical properties after steel pipe production based on machine learning, which relates to the field of artificial intelligence prediction. Aiming at the limitations of existing prediction methods in dealing with complex production processes and multi-factor influences, the present invention collects multi-parameters and microstructure information of steel pipe production, introduces a hierarchical weighted entropy weight method to screen strongly correlated features, uses an improved multi-modal deep learning model DeepFM to fuse multi-type data for unified modeling, enhances the model interpretability and generates a confidence interval through the integrated decision tree XGBoost, and finally deploys the model on edge devices to achieve real-time prediction. The present invention can comprehensively consider influencing factors, improve prediction accuracy, effectively fuse multi-modal data, help adjust process parameters according to the prediction during the production process, improve production efficiency and product quality, and is applicable to the field of steel pipe production.

Description

Technical Field

[0001] The present invention belongs to the field of artificial intelligence prediction, and particularly relates to a method for predicting the mechanical properties of steel pipes after production based on machine learning. Background Art

[0002] In the process of steel pipe production, mechanical properties are the key indicators to measure the quality of steel pipes, which directly affect the safety and reliability of steel pipes in practical applications. However, the existing methods for predicting the mechanical properties of steel pipes often rely on empirical formulas and physical models, and these methods have certain limitations when dealing with complex production processes and multi-factor influences. For example, in actual production, the mechanical properties of steel pipes are affected by many factors such as the chemical composition of raw materials, production process parameters, and microstructure information. Moreover, there are complex non-linear relationships between these factors, and it is difficult for traditional methods to accurately capture and model them. In addition, the fusion processing of multi-modal information such as the time-series changes of dynamic parameters and microstructure image data during the production process is also a problem that traditional methods are difficult to effectively solve. With the continuous improvement of the quality requirements for steel pipes in industrial production, as well as the wide application of digital and intelligent technologies in the manufacturing industry, there is an urgent need for a more accurate, efficient and capable of comprehensively processing multi-source data method for predicting the mechanical properties of steel pipes after production, in order to improve the quality and production efficiency of steel pipe production, reduce production costs, and meet the growing market demand. Summary of the Invention

[0003] The present invention aims at the technical problems existing in the prediction of steel pipe properties, and proposes a method for predicting the mechanical properties of steel pipes after production based on machine learning.

[0004] In order to achieve the above object, the technical solution adopted by the present invention is as follows: including the following steps:

[0005] S1. First, collect various parameters in steel pipe production, including the chemical composition of raw materials, quenching temperature, tempering temperature, rolling temperature, rolling speed, cooling method, cooling time, and the microstructure information of steel pipes, including grain size and grain boundary distribution;

[0006] S2. Secondly, introduce a method based on hierarchical weighted entropy weight to screen out the features strongly related to mechanical properties and remove redundant and invalid features;

[0007] S3. For the dynamic parameters in the production process, introduce a long short-term memory network LSTM to capture the influence of the time-series changes of these parameters on mechanical properties. For the image data of the microstructure of steel pipes, construct a deep learning model based on CNN to extract image features, fuse the image features, time-series data and other static process parameters, and use the improved multi-modal deep learning model DeepFM to realize the unified modeling of multi-dimensional data and improve the prediction accuracy;

[0008] S4. Then, the interpretability of the deep learning model is enhanced through the integrated decision tree XGBoost, the influence degree of different production parameters on the mechanical properties of steel pipes is analyzed, a confidence interval is generated for the prediction result, so that the process parameters can be adjusted according to the uncertainty of the prediction during the production process to ensure that the actual product performance meets the requirements;

[0009] S5. Finally, the model is deployed to the edge device of the production line to achieve real-time prediction of mechanical properties, avoid delays, and improve production efficiency;

[0010] The implementation steps of using the improved multi-modal deep learning model DeepFM to achieve unified modeling of multi-dimensional data in step S3 are as follows:

[0011] S31. First, the input data is divided into time-series dynamic data, image data, and static process data. The feature sets of these multi-modal data are X = {x1, x2... x n}, where n is the total number of features;

[0012] S32. Secondly, linear processing is performed on the input features, and the formula is: where w0 is the bias term, x i is the i-th feature in the multi-modal data, and g(x i ) is the function used for feature evaluation. p k represents the distribution probability of feature x i ;

[0013] S33. Then, in order to enhance the understanding of complex multi-modal features, third-order feature interactions are added, and the formula is as follows: where v i , v j , v k are the embedding vectors of features x i , x j , x k , <v i , v j , v k > represents the triple inner product of the embedding vectors of the third-order interaction features, and x i x j x k represents the third-order interaction of features;

[0014] S34. Combining the deep feature representations of multi-modal data, high-dimensional feature interactions are learned through a deep neural network, and the higher-order relationships of features are further extracted through a multi-layer non-linear network;

[0015] S35. Combine the linear part, third-order feature interactions, and deep part, and finally output the prediction result.

[0016] Preferably, the implementation steps of introducing the method based on hierarchical weighted entropy weight to achieve feature selection in step 2 are as follows:

[0017] S21. First, according to different data sources and attributes in the steel pipe production process, the features are divided into multiple categories, and the data of each category is preprocessed to ensure numerical normalization and avoid calculation deviation caused by different dimensions between features;

[0018] S22. For each feature category dataset with m samples and n features in each sample, for each feature x i The proportion of the feature value obtained through normalization Calculate the entropy value of the feature according to the normalized feature value where To ensure the normalization of the entropy value, calculate the weight of each feature according to the entropy value Obtain the feature weight distribution within this category;

[0019] S23. Calculate the importance evaluation between categories, use the correlation analysis method to evaluate the contribution degree of each feature category to the mechanical properties, assign an initial category weight coefficient to each category, and normalize the category weights to ensure that the importance of different categories adds up to 1;

[0020] S24. Multiply the feature weight in each category by the category weight to calculate the final weight W of each feature final (x i ) = W k ×W i , where W final (x i ) is the final weight of feature x i , W k is the weighted coefficient of the category where this feature is located, and W i is the feature weight within the category;

[0021] S25. According to the finally calculated feature weight distribution, set a threshold of 0.32, and retain the features with weights higher than the threshold.

[0022] Preferably, the implementation process of the integrated decision tree XGBoost in step S4 is as follows:

[0023] S41. First, prepare the multi-modal data processed by the deep learning model DeepFM and obtain the prediction output result of the model;

[0024] S42. Use the XGBoost model to analyze the feature importance, perform regression on the features by training the XGBoost model, and obtain the importance score of each feature: where T iAll decision trees that contain feature x i |T i | is the number of decision trees, and Δy j is the prediction change caused by the introduction of feature x i ;

[0025] S43. Train the output of the deep learning model through the XGBoost model to generate more accurate prediction results. The objective function of model training is: where m is the number of samples, y i is the actual label, is the prediction result, λ is the regularization parameter, and Ω(f k ) is the penalty term for model complexity;

[0026] S44. Then, in order to generate the confidence interval of the prediction result, use the output distribution of XGBoost to calculate the confidence interval. The formula for setting the confidence interval is: where is the predicted value, z is the critical value of the standard normal distribution, and σ is the standard deviation of the prediction result;

[0027] S45. Finally, make adjustments according to the prediction result generated by the model and its confidence interval. When the confidence interval is large, consider increasing the control of relevant parameters to reduce uncertainty. When the importance of certain features is high, give priority to monitoring and optimizing these features to ensure that the performance of the final product meets the requirements.

[0028] Compared with the prior art, the advantages and positive effects of the present invention are as follows: comprehensively collect various parameters and microstructure information of steel pipe production, can comprehensively consider the influence of multiple factors, and compared with the existing methods that rely on empirical formulas and physical models, the prediction is more accurate. Screen features based on hierarchical weighted entropy weights, overcome the deficiencies of traditional entropy weight methods, and improve the accuracy. Improve the multi-modal deep learning model DeepFM to effectively fuse multi-type data, improve the comprehensiveness and accuracy of prediction, and make up for the defects of traditional models in fusing multi-modal data. The integrated XGBoost enhances the model interpretability, generates confidence intervals to assist production adjustment, and can also perform real-time prediction, improving production efficiency and quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0030] Figure 1 is a stress-strain curve diagram simulated according to the prediction result of mechanical properties;

[0031] Figure 2 It is a schematic diagram of the microstructure of steel pipe materials; Specific implementation manners

[0032] In order to more clearly understand the above objects, features and advantages of the present invention, the present invention will be further described below with reference to the accompanying drawings and embodiments. It should be noted that, without conflict, the embodiments of the present application and the features in the embodiments may be combined with each other.

[0033] Many specific details are set forth in the following description in order to fully understand the present invention. However, the present invention may be implemented in other ways different from those described herein. Therefore, the present invention is not limited by the limitations of the specific embodiments disclosed in the following specification.

[0034] Embodiment. In a steel pipe production enterprise, in order to achieve accurate prediction of the mechanical properties of steel pipes after production, thereby improving product quality, optimizing the production process and enhancing production efficiency, and solving the limitation problems existing in the existing prediction methods due to relying on empirical formulas and physical models when dealing with complex production processes and multi-factor influences, a prediction method for the mechanical properties of steel pipes after production based on machine learning is adopted to implement.

[0035] First, a variety of parameters in steel pipe production need to be collected, including the chemical composition of raw materials, quenching temperature, tempering temperature, rolling temperature, rolling speed, cooling method, cooling time, and the microstructure information of steel pipes, including grain size and grain boundary distribution; the accurate collection of these data provides rich original information for subsequent accurate prediction of mechanical properties. By collecting parameters from multiple aspects, the factors affecting the mechanical properties of steel pipes can be comprehensively considered, laying a solid foundation for accurate prediction.

[0036] Considering that the defect of the entropy weight method lies in ignoring the feature category differences and not considering the importance between categories, resulting in unreasonable weight distribution and inaccurate feature screening. The hierarchical weighted entropy weight method overcomes these problems, considers the category characteristics, evaluates the category importance, accurately screens features, and improves the prediction accuracy. First, according to different data sources and attributes in the steel pipe production process, the features are divided into multiple categories, and the data of each category are preprocessed to ensure numerical normalization and avoid calculation deviations caused by different dimensions between features; for each feature category dataset, there are m samples, and each sample has n features. For each feature x i The proportion of the feature value obtained through normalization processing Calculate the entropy value of the feature according to the normalized feature value Wherein Used to ensure the normalization of the entropy value, and calculate the weight of each feature according to the entropy value Obtain the feature weight distribution within the category; calculate the importance evaluation between categories, use the correlation analysis method to evaluate the contribution of each feature category to the mechanical properties, assign an initial category weight coefficient to each category, and normalize the category weights to ensure that the importance of different categories sums up to 1; multiply the feature weight in each category by the category weight to calculate the final weight W of each feature final (x i ) = W k ×W i , where W final (x i ) is the final weight of feature x i , W k is the weighted coefficient of the category where the feature is located, and W i is the feature weight within the category; finally, according to the finally calculated feature weight distribution, set a threshold of 0.32 and retain the features with weights higher than the threshold.

[0037] To address different parameters in the production process, the present invention adopts an improved multi-modal deep learning model DeepFM. Traditional methods are difficult to effectively fuse the time-series dynamic data, image data, and static process parameters in steel pipe production. These data types have different structures and feature representation methods, and traditional models cannot fully explore the complex correlation relationships between them, resulting in information loss or insufficient utilization. The DeepFM model can effectively fuse multi-modal data such as time-series dynamic data, image data, and static process parameters. Through reasonable structural design, different types of data are processed separately, and their features are organically combined inside the model, making full use of the information in various types of data and improving the comprehensiveness and accuracy of prediction. First, the input data is divided into time-series dynamic data, image data, and static process data. The feature sets of these multi-modal data are X = {x1, x2... x n}, where n is the total number of features; secondly, linear processing is performed on the input features, and the formula is: where w0 is the bias term, x i is the i-th feature in the multi-modal data, and g(x i ) is the function for feature evaluation. p k represents the distribution probability of feature x i ; then, to enhance the understanding of complex multi-modal features, third-order feature interactions are added, and the formula is as follows: where v i , v j , v k are the embedding vectors of features x i , x j , x k , <v i , v j,v k > represents the triple inner product of the embedding vectors of the third-order interaction features, x i x j x k represents the third-order interaction of features; combined with the deep feature representation of multi-modal data, high-dimensional feature interactions are learned through a deep neural network, and the high-order relationships of features are further extracted through a multi-layer non-linear network; the linear part, the third-order feature interaction, and the deep part are combined to finally output the prediction result

[0038] To enhance the interpretability of the deep learning model, the present invention generates a confidence interval for the prediction result by integrating the decision tree XGBoost, so that the process parameters can be adjusted according to the prediction uncertainty during the production process to ensure that the actual product performance meets the requirements. First, prepare the multi-modal data processed by the deep learning model DeepFM and obtain the prediction output result of the model; use the XGBoost model to analyze the feature importance, and perform regression on the features through training the XGBoost model to obtain the importance score of each feature: where T i is all the decision trees containing the feature x i |T i | is the number of decision trees, and Δy j is the prediction change caused by the introduction of the feature x i ; train the output of the deep learning model through the XGBoost model to generate a more accurate prediction result, and the objective function of the model training is: where m is the number of samples, y i is the actual label, is the prediction result, λ is the regularization parameter, and Ω(f k ) is the penalty term for the model complexity; then, to generate the confidence interval of the prediction result, use the output distribution of XGBoost to calculate the confidence interval, and the formula for setting the confidence interval is: where is the predicted value, z is the critical value of the standard normal distribution, and σ is the standard deviation of the prediction result; finally, adjust according to the prediction result and its confidence interval generated by the model. When the confidence interval is large, consider increasing the control of the relevant parameters to reduce the uncertainty. When the importance of some features is high, give priority to monitoring and optimizing these features to ensure that the final product performance meets the requirements.

[0039] After the model training optimization is finally completed, it is deployed to the edge devices on the production line. These edge devices are located at the production site and can collect various types of data on steel pipe production in real time, such as the chemical composition of raw materials, various process parameters, and microstructure information. The collected data is immediately input into the model, and the model quickly calculates and outputs the real-time mechanical property prediction results. This process is closely coherent, avoiding the delays caused by data transmission and remote computing. The production process can adjust in a timely manner according to this result to ensure efficient production, stable product quality, effectively improve production efficiency, and ensure the smooth operation of the production line.

[0040] Figure 1 The stress-strain curve simulated according to the prediction results of the mechanical properties. In the figure, the curve trends of the predicted values and the true values can be clearly seen. The predicted value curve is relatively close to the true value curve, which indicates that the method for predicting the mechanical properties after steel pipe production based on machine learning proposed by the present invention has high accuracy. In the low strain region, both the predicted values and the true values show small stress changes. As the strain increases, the stress gradually rises, and within a certain strain range, the growth trends of the two are basically the same. Figure 2 Shows the detailed information of the microstructure of the steel pipe material. From the figure, characteristics such as the morphology, size of the grains, and the clarity of the grain boundaries can be observed. These microstructure information is of great significance for understanding the mechanical properties of the steel pipe. Smaller and uniform grain sizes usually help to improve the strength and toughness of the steel pipe because smaller grains can hinder the movement of dislocations, thereby enhancing the material's ability to resist deformation. And the uniformity of the grain boundary distribution also affects the mechanical properties of the material. Uniformly distributed grain boundaries can make the stress transfer more evenly in the material, reduce the stress concentration phenomenon, and thus improve the overall performance of the steel pipe.

[0041] The above are only the preferred embodiments of the present invention, and are not limitations on the present invention in other forms. Any person skilled in the art may use the technical content disclosed above to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change, and modification made to the above embodiments based on the technical essence of the present invention still belong to the protection scope of the technical solution of the present invention.

Claims

1. A method for predicting the mechanical properties of steel pipes after production based on machine learning, characterized in that: The following steps are involved: S1. First, collect various parameters in steel pipe production including chemical composition of raw materials, quenching temperature, tempering temperature, rolling temperature, rolling speed, cooling method, cooling time, and microstructure information of steel pipe including grain size and grain boundary distribution; S2. Secondly, a method based on hierarchical weighted entropy weights is introduced to screen out features that are strongly related to mechanical properties and remove redundant and invalid features; S3. For the dynamic parameters in the production process, the long short-term memory network LSTM is introduced to capture the influence of the time series changes of these parameters on the mechanical properties. For the image data of the microstructure of the steel pipe, a deep learning model based on CNN is constructed to extract image features, integrate image features, time series data with other static process parameters, and use the improved multimodal deep learning model DeepFM to achieve unified modeling of multi-dimensional data and improve prediction accuracy; S4. Then, by integrating the decision tree XGBoost, the interpretability of the deep learning model is enhanced, the influence of different production parameters on the mechanical properties of the steel pipe is analyzed, and a confidence interval is generated for the prediction results, so that the process parameters can be adjusted according to the uncertainty of the prediction during the production process to ensure that the actual product performance meets the requirements; S5. Finally, the model is deployed on the edge devices of the production line to achieve real-time mechanical property prediction, avoid delays, and improve production efficiency; In step S3, the improved multimodal deep learning model DeepFM is used to implement unified modeling of multi-dimensional data in the following steps: S31. First, the input data is divided into time series dynamic data, image data and static process data. The feature set of these multimodal data is X = {x1, x2...x n }, where n is the total number of features; S32, then perform linear processing on the input features, the formula is: Where w0 is the bias term, x i is the i-th feature in the multimodal data, g(x i ) is the function used for feature evaluation. p k Represents feature x i The distribution probability of S33. Then, in order to enhance the understanding of complex multimodal features, the third-order feature interaction is added. The formula is as follows: where v i ,v j ,v k is feature x i , x j , x k The embedding vector of i ,v j ,v k > represents the triple inner product of the embedding vector of the third-order interaction feature, x i x j x k Represents the third-order interaction of features; S34, combined with the deep feature representation of multimodal data, high-dimensional feature interactions are learned through deep neural networks, and high-order relations of features are further extracted through multi-layer nonlinear networks; S35, combine the linear part, the third-order feature interaction and the deep part, and finally output the prediction result 2. The method for predicting mechanical properties of steel pipes after production based on machine learning according to claim 1, characterized in that: The steps for implementing feature selection by introducing the method based on hierarchical weighted entropy weights in step S2 are as follows: S21. First, according to the different data sources and attributes in the steel pipe production process, the features are divided into multiple categories, and the data of each category is preprocessed to ensure the normalization of the values ​​and avoid calculation deviations caused by different dimensions between features; S22. For each feature category, there are m samples in the data set, and each sample has n features. For each feature x i The proportion of eigenvalues ​​obtained through normalization Calculate the entropy value of the feature based on the normalized eigenvalue in Used to ensure the normalization of entropy values ​​and calculate the weight of each feature based on the entropy value Get the feature weight distribution within the category; S23, calculate the importance evaluation between categories, use the correlation analysis method to evaluate the contribution of each feature category to the mechanical properties, assign an initial category weight coefficient to each category, and normalize the weights of each category to ensure that the sum of the importance of different categories is 1; S24. Multiply the feature weight in each category by the category weight to calculate the final weight W of each feature final (x i )=W k ×W i , where W final (x i ) is the feature x i The final weight, W k is the weight coefficient of the category to which the feature belongs, W i is the feature weight within the category; S25. According to the finally calculated feature weight distribution, a threshold of 0.32 is set, and features with weights higher than the threshold are retained.

3. The method for predicting mechanical properties of steel pipes after production based on machine learning according to claim 1, characterized in that: The implementation process of integrating decision tree XGBoost in step S4 is as follows: S41, first prepare multimodal data processed by the deep learning model DeepFM, and obtain the prediction output result of the model; S42. Use the XGBoost model to analyze feature importance. By training the XGBoost model to regress the features, we can get the importance score of each feature: Where T i is a feature x i All decision trees of |T i | is the number of decision trees, Δy j is due to the feature x i changes in forecasts due to the introduction of S43. The output of the deep learning model is trained through the XGBoost model to generate more accurate prediction results. The objective function of the model training is: Where m is the number of samples, y i is the actual label, is the prediction result, λ is the regularization parameter, Ω(f k ) is the penalty term for model complexity; S44. Then, in order to generate a confidence interval for the prediction result, the output distribution of XGBoost is used to calculate the confidence interval, and the formula for setting the confidence interval is: in is the predicted value, z is the critical value of the standard normal distribution, and σ is the standard deviation of the predicted result; S45. Finally, make adjustments based on the prediction results and confidence intervals generated by the model. When the confidence interval is large, consider increasing the control of related parameters to reduce uncertainty. When the importance of features is high, give priority to monitoring and optimizing these features to ensure that the final product performance meets the requirements.

Images