Accurate crown control is paramount for ensuring the quality of hot-rolled strip products.Given the multitude of influencing parameters and the intricate coupling and genetic relationships among them,the conventional ...Accurate crown control is paramount for ensuring the quality of hot-rolled strip products.Given the multitude of influencing parameters and the intricate coupling and genetic relationships among them,the conventional crown control method is no longer sufficient to meet the precision requirements of schedule-free rolling.To address this limitation,an optimization framework for hot-rolled strip crown control was developed based on model-driven digital twin(MDDT).This framework enhances the strip crown control precision by facilitating collaborative operations among physical entities,virtual models,and functional application layers.In virtual modeling,a data-driven approach that integrates the extreme gradient boosting and the improved Harris hawk optimization algorithm was firstly proposed to fit the relationship between key process parameters and strip crown,and a global-local collaborative training strategy was proposed to enhance the model adaptability to diverse working conditions.Subsequently,the influence of crucial process factors on the virtual model was examined through model responses.Furthermore,a novel optimization mode for crown control based on MDDT was established by aligning and reconstructing both the physical and virtual models,thereby enhancing the crown control precision.Finally,data trials were conducted to validate the effectiveness of the proposed framework.The results indicated that the proposed method exhibited satisfactory performance and could be effectively utilized to improve the crown control precision.展开更多
High-order asymmetric flatness defects resulting from the abnormal state of roll system are the main issue of precision rolling mill in the manufacturing process of high-strength thin strip.Due to the difficulty of mo...High-order asymmetric flatness defects resulting from the abnormal state of roll system are the main issue of precision rolling mill in the manufacturing process of high-strength thin strip.Due to the difficulty of monitoring and adjusting the abnormal state,the spatial state of roll system cannot be controlled by traditional methods.It is difficult to fundamentally improve these high-order asymmetric flatness defects.Therefore,a digital twin model of flatness control process for S6-high rolling mill was established,which could be used to analyze the influence of the abnormal state on the flatness control characteristic and propose improvement strategies.The internal relationship between the force state of side support roll system and the abnormal state of roll system was proposed.The XGBoost algorithm model was established to analyze the contribution degree of the side support roll system force to the flatness characteristic quantity.The abnormal state of roll system in the S6-high rolling mill can be diagnosed by analyzing the flatness characteristic difference between flatness value of the rolled strip and calculated characteristic value of finite element simulation.The flatness optimization model of the gray wolf optimization–long short-term memory non-dominated sorting whale optimization algorithm(GWO-LSTM-NSWOA)was established,and the decision-making selection was made from the Pareto frontier based on the flatness requirements of cold rolling to regulate the abnormal state of the roll system.The results indicate that the contribution degree of the force of the side support roll system to the flatness characteristics is more than 25%,which is the main influence of high-order asymmetric flatness defect.The performance of the GWO-LSTM flatness feature prediction model has clear advantages over back propagation and LSTM.The practical applications show that optimizing the force of side support roll system can reduce the high point of high-strength strip flatness from 13.2 to 6 IU and decrease the percentage of low-strength strip flatness defects from 1.6%to 1.2%.This optimization greatly reduced the proportion of flatness defects,improved the accuracy level of flatness control of precision rolling mill,and provided a guarantee for the stable production of thin strip.展开更多
To achieve stable rolling,the influence of a tension mechanism of a large diameter ratio roll system on the rolling process of a strip flatness electromagnetic control rolling mill is studied.Through the analysis of t...To achieve stable rolling,the influence of a tension mechanism of a large diameter ratio roll system on the rolling process of a strip flatness electromagnetic control rolling mill is studied.Through the analysis of the rolling deformation zone,the deformation zone composition form of a large diameter ratio roll system and a calculation formula of neutral angle under tension are proposed.To analyze the effect of front and post tensions on the rolling characteristic and the strip flatness control characteristic,a three-dimensional rolling finite element(FE)model of a large diameter ratio roll system with the function of roll profile electromagnetic control is established by FE software and verified by a strip flatness electromagnetic control rolling mill.Based on the model,the strip thickness characteristic,metal transverse flow,strip flatness state,and adjustment range of the loaded roll gap are analyzed for different front and post tensions setting values.The results show that changing the front or post tension setting values can improve the single-pass reduction rate of a large diameter ratio roll system and have little effect on the flatness control ability of the strip flatness electromagnetic control rolling mill.展开更多
The slow phase transformation of microalloyed dual phase steel makes the nonuniform stress and temperature fields during the post rolling cooling process have a significant impact on the phase transformation process.G...The slow phase transformation of microalloyed dual phase steel makes the nonuniform stress and temperature fields during the post rolling cooling process have a significant impact on the phase transformation process.Given the relatively slow phase transformation of DP780 steel within the microalloyed dual phase steel series,the influence of stress on the phase transformation behavior of DP780 steel was investigated.To quantify the nonuniform thermal and stress conditions in the steel coil,a thermo-mechanical coupled finite element model of the hot-rolled strip cooling process was established.Based on the simulation data,DP780 steel was chosen as the research material,and Gleeble 3500 thermal simulation equipment was used for experimental validation.The thermal expansion curves were analyzed through regression to establish the dynamic model of DP780 steel phase transformation under stress.Subsequently,metallographic analysis was conducted to determine phase transformation type and grain size of DP780 steel.The results confirmed that the stress promotes the occurrence of semi-diffusion-type bainite transformation.Furthermore,an appropriate level of stress facilitates the growth of bainitic grains,while the increased stress inhibits the growth of ferritic grains.展开更多
Thermal-force driving of roll profile electromagnetic control technology(RPECT),which can be used to adjust the roll profile,can be affected by the sequential temperature rise between the electromagnetic stick(ES)and ...Thermal-force driving of roll profile electromagnetic control technology(RPECT),which can be used to adjust the roll profile,can be affected by the sequential temperature rise between the electromagnetic stick(ES)and electromagnetic control roll.Due to the limited space of ES and induction coil,the cross-sectional area of induction coil can be inevitably affected by changing the size of the ES induction zone,which can further change the energy input under the same electromagnetic parameters,the temperature rising effect and the bulging ability.To investigate this phenomenon,the effects of the radius of the induction zone on the thermal-force contribution ratio,the heating ability of ES and the temperature distribution were analyzed through an electromagnetic-thermal-structural finite element model.To ensure that the results are applicable to RPECT,the thermal energy conversion ability and thermal-force roll crown control ability under different lengths of the induction zone were analyzed.It was found that whether the current density regulation mode or the current frequency regulation mode is adopted,the cases with 20 or 25 mm radius of the induction zone have the great thermal energy conversion ability and the good thermal-force roll crown control ability.The reasonable adjustment of the length of the induction zone can reduce the radius required for the maximum energy efficiency regulation.Combined with the results of the simulation analysis,the optimization of ES based on the control ability maximization requirement is achieved,which provides the base for the design and configuration of ES in RPECT.展开更多
Donghua steel continuous casting-rolling(DSCCR)production line is an endless rolling production line independently integrated and developed by a Chinese enterprise.To solve the problem of out-of-control shape caused b...Donghua steel continuous casting-rolling(DSCCR)production line is an endless rolling production line independently integrated and developed by a Chinese enterprise.To solve the problem of out-of-control shape caused by insufficient adjustment ability of work roll bending force of DSCCR finishing mills,the backup roll contours of the finishing mill were optimized considering multi-objectives,and varying contact length backup roll(VCR)contours matching the concave work rolls were designed based on the rapid rolls-strip deformation calculation model,which integrated the elastic deformation of the rolls and the fast plastic deformation of the rolled strip.The simulation results showed that VCR configuration can significantly enhance the control efficiency of the work roll bending force and increase the transverse stiffness of rolling mill compared with the conventional roll configuration.In addition,VCR backup roll can improve the contact state between backup roll and work roll.The industrial production showed that the bending force setup value was more reasonable and the strip crown control accuracy was improved.展开更多
The method to predict roll deformation precisely and efficiently is vital for the strip shape control of a six-high rolling mill. Traditional calculation methods of roll deformation, such as the finite element method ...The method to predict roll deformation precisely and efficiently is vital for the strip shape control of a six-high rolling mill. Traditional calculation methods of roll deformation, such as the finite element method and the influence function method, have been widely used due to their accuracies. However, the required calculation time is too long to be applied to the realtime control. Therefore, a rapid calculation method for predicting roll deformation of a six-high rolling mill was proposed, which employed the finite difference method to calculate the roll deflection and used a polynomial to describe the nonlinear relationship between roll flattening and roll contact pressure. Furthermore, a new correction strategy was proposed in the iteration, where the roll center flattening and the roll flattening deviation were put forward and corrected simultaneously in the iteration process according to the static equilibrium of roll. Finally, by the comparison with traditional methods, the proposed method was proved to be more efficient and it was suitable for the online calculation of the strip shape control.展开更多
Mechanical performance prediction is the key to the transformation and upgrading of steel enterprises to intelligent manufacturing.Due to time-varying manufacturing data,the traditional prediction model of mechanical ...Mechanical performance prediction is the key to the transformation and upgrading of steel enterprises to intelligent manufacturing.Due to time-varying manufacturing data,the traditional prediction model of mechanical properties of hotrolled strip may cause performance degradation or even failure in its use.An MDA-JITL model was thus proposed to handle the modeling problem of complex time-varying data.Relevant parameters were first chosen and normalized.Then,a distance measurement method combining the importance of data attributes and time characteristics was designed to select the most suitable samples for on-line local modeling.After that,using the chosen dataset,a linear local model was created to predict target sample.Finally,an uncertainty evaluation method was designed to evaluate the uncertainty of prediction results.Furthermore,the appropriate dataset partition and off-line simulation experiment scheme were created based on the peculiarities of hot-rolling production.The suggested model performs much better than the classic global model when applied to actual production data from a steel plant.The stability of its prediction accuracy is demonstrated in a simulation prediction for up to five months.Moreover,there is a high link between the uncertainty evaluation metrics and the prediction error of the model,reducing the field sampling rate by 30%in industrial applications in the latest year.展开更多
To reveal the generation mechanism of the quarter buckle in the process of hot-rolled temper rolling,the elastic–plastic finite element method is used to calculate the deformation of the roll and the strip during the...To reveal the generation mechanism of the quarter buckle in the process of hot-rolled temper rolling,the elastic–plastic finite element method is used to calculate the deformation of the roll and the strip during the temper rolling.The change of the cross section of the strip and the distribution of the longitudinal stress are analysed under different bending forces to obtain the boundary conditions of the quarter buckle.The generation of the quarter buckle is further analysed from the bending and contact flattening of the roll system and the elastic recovery of the strip after rolling.We found that the quarter buckle is closely related to the high-order distribution of the contact flattening between the work roll and the strip and is less affected by the bending deformation of the roll and the contact flattening between the rolls.Finally,a new work roll contour of the locally variable crown is proposed to change the distribution of contact flattening between the work roll and the strip.It is verified through theoretical calculations and industrial applications that the new contour can effectively improve the quarter buckle.展开更多
With the increasing demand on higher strip quality, the profile and flatness of hot rolling strips have become subjects of concern, particularly for compact strip product(CSP) hot strip mills. Based on the roll contou...With the increasing demand on higher strip quality, the profile and flatness of hot rolling strips have become subjects of concern, particularly for compact strip product(CSP) hot strip mills. Based on the roll contour, control model, and rolling process, a comprehensive shape control technology is proposed and applied to CSP hot strip mill of Lianyuan steel, which includes optimization and design of the work roll contour and varying contact back-up roll(VCR) plus backup roll contour, analysis of the flatness feedback control model, as well as improvement of the rolling process control system. The application of the technology has significantly improved the shape control performance. The roll wear is improved and the general roll consumption of the finishing mill is reduced by 29.86%.The percentages that satisfy the control target ranges of the average strip flatness and crown are increased by approximately 15.40%and 14.82%, respectively. The rejection rate of grade Q235 due to shape quality problem is reduced monthly by 39.69%, which creates significant economic benefits for the plant.展开更多
The cross-section profile is a key signal for evaluating hot-rolled strip quality,and ignoring its defects can easily lead to a final failure.The characteristics of complex curve,significant irregular fluctuation and ...The cross-section profile is a key signal for evaluating hot-rolled strip quality,and ignoring its defects can easily lead to a final failure.The characteristics of complex curve,significant irregular fluctuation and imperfect sample data make it a challenge of recognizing cross-section defects,and current industrial judgment methods rely excessively on human decision making.A novel stacked denoising autoencoders(SDAE)model optimized with support vector machine(SVM)theory was proposed for the recognition of cross-section defects.Firstly,interpolation filtering and principal component analysis were employed to linearly reduce the data dimensionality of the profile curve.Secondly,the deep learning algorithm SDAE was used layer by layer for greedy unsupervised feature learning,and its final layer of back-propagation neural network was replaced by SVM for supervised learning of the final features,and the final model SDAE_SVM was obtained by further optimizing the entire network parameters via error back-propagation.Finally,the curve mirroring and combination stitching methods were used as data augmentation for the training set,which dealt with the problem of sample imbalance in the original data set,and the accuracy of cross-section defect prediction was further improved.The approach was applied in a 1780-mm hot rolling line of a steel mill to achieve the automatic diagnosis and classification of defects in cross-section profile of hot-rolled strip,which helps to reduce flatness quality concerns in downstream processes.展开更多
基金financially supported by the National Key Research and Development Program of China(Grant No.2023YFB3710204)Guangxi Science and Technology Major Program(Grant No.AA23023028-1)+1 种基金Natural Science Foundation of Heilongjiang Province of China for Distinguished Young Scientists(Grant No.JQ2022E007)Xinjiang Production and Construction Corps Science and Technology Plan(Grant No.2023AA003).
文摘Accurate crown control is paramount for ensuring the quality of hot-rolled strip products.Given the multitude of influencing parameters and the intricate coupling and genetic relationships among them,the conventional crown control method is no longer sufficient to meet the precision requirements of schedule-free rolling.To address this limitation,an optimization framework for hot-rolled strip crown control was developed based on model-driven digital twin(MDDT).This framework enhances the strip crown control precision by facilitating collaborative operations among physical entities,virtual models,and functional application layers.In virtual modeling,a data-driven approach that integrates the extreme gradient boosting and the improved Harris hawk optimization algorithm was firstly proposed to fit the relationship between key process parameters and strip crown,and a global-local collaborative training strategy was proposed to enhance the model adaptability to diverse working conditions.Subsequently,the influence of crucial process factors on the virtual model was examined through model responses.Furthermore,a novel optimization mode for crown control based on MDDT was established by aligning and reconstructing both the physical and virtual models,thereby enhancing the crown control precision.Finally,data trials were conducted to validate the effectiveness of the proposed framework.The results indicated that the proposed method exhibited satisfactory performance and could be effectively utilized to improve the crown control precision.
基金financially supported by the National Key Research and Development Program of China(Grant No.2023YFB3812602).
文摘High-order asymmetric flatness defects resulting from the abnormal state of roll system are the main issue of precision rolling mill in the manufacturing process of high-strength thin strip.Due to the difficulty of monitoring and adjusting the abnormal state,the spatial state of roll system cannot be controlled by traditional methods.It is difficult to fundamentally improve these high-order asymmetric flatness defects.Therefore,a digital twin model of flatness control process for S6-high rolling mill was established,which could be used to analyze the influence of the abnormal state on the flatness control characteristic and propose improvement strategies.The internal relationship between the force state of side support roll system and the abnormal state of roll system was proposed.The XGBoost algorithm model was established to analyze the contribution degree of the side support roll system force to the flatness characteristic quantity.The abnormal state of roll system in the S6-high rolling mill can be diagnosed by analyzing the flatness characteristic difference between flatness value of the rolled strip and calculated characteristic value of finite element simulation.The flatness optimization model of the gray wolf optimization–long short-term memory non-dominated sorting whale optimization algorithm(GWO-LSTM-NSWOA)was established,and the decision-making selection was made from the Pareto frontier based on the flatness requirements of cold rolling to regulate the abnormal state of the roll system.The results indicate that the contribution degree of the force of the side support roll system to the flatness characteristics is more than 25%,which is the main influence of high-order asymmetric flatness defect.The performance of the GWO-LSTM flatness feature prediction model has clear advantages over back propagation and LSTM.The practical applications show that optimizing the force of side support roll system can reduce the high point of high-strength strip flatness from 13.2 to 6 IU and decrease the percentage of low-strength strip flatness defects from 1.6%to 1.2%.This optimization greatly reduced the proportion of flatness defects,improved the accuracy level of flatness control of precision rolling mill,and provided a guarantee for the stable production of thin strip.
基金supported by the Natural Science Foundation of Hebei Province of China(Grant No.E2021203129).
文摘To achieve stable rolling,the influence of a tension mechanism of a large diameter ratio roll system on the rolling process of a strip flatness electromagnetic control rolling mill is studied.Through the analysis of the rolling deformation zone,the deformation zone composition form of a large diameter ratio roll system and a calculation formula of neutral angle under tension are proposed.To analyze the effect of front and post tensions on the rolling characteristic and the strip flatness control characteristic,a three-dimensional rolling finite element(FE)model of a large diameter ratio roll system with the function of roll profile electromagnetic control is established by FE software and verified by a strip flatness electromagnetic control rolling mill.Based on the model,the strip thickness characteristic,metal transverse flow,strip flatness state,and adjustment range of the loaded roll gap are analyzed for different front and post tensions setting values.The results show that changing the front or post tension setting values can improve the single-pass reduction rate of a large diameter ratio roll system and have little effect on the flatness control ability of the strip flatness electromagnetic control rolling mill.
基金supported by the National Natural Science Foundation of China(Grant No.52004029).
文摘The slow phase transformation of microalloyed dual phase steel makes the nonuniform stress and temperature fields during the post rolling cooling process have a significant impact on the phase transformation process.Given the relatively slow phase transformation of DP780 steel within the microalloyed dual phase steel series,the influence of stress on the phase transformation behavior of DP780 steel was investigated.To quantify the nonuniform thermal and stress conditions in the steel coil,a thermo-mechanical coupled finite element model of the hot-rolled strip cooling process was established.Based on the simulation data,DP780 steel was chosen as the research material,and Gleeble 3500 thermal simulation equipment was used for experimental validation.The thermal expansion curves were analyzed through regression to establish the dynamic model of DP780 steel phase transformation under stress.Subsequently,metallographic analysis was conducted to determine phase transformation type and grain size of DP780 steel.The results confirmed that the stress promotes the occurrence of semi-diffusion-type bainite transformation.Furthermore,an appropriate level of stress facilitates the growth of bainitic grains,while the increased stress inhibits the growth of ferritic grains.
基金This work was supported by the National Natural Science Foundation of China(Grant Nos.U1560206 and 51975510)the Natural Science Foundation of China in Hebei Province(Grant No.E2021203129).
文摘Thermal-force driving of roll profile electromagnetic control technology(RPECT),which can be used to adjust the roll profile,can be affected by the sequential temperature rise between the electromagnetic stick(ES)and electromagnetic control roll.Due to the limited space of ES and induction coil,the cross-sectional area of induction coil can be inevitably affected by changing the size of the ES induction zone,which can further change the energy input under the same electromagnetic parameters,the temperature rising effect and the bulging ability.To investigate this phenomenon,the effects of the radius of the induction zone on the thermal-force contribution ratio,the heating ability of ES and the temperature distribution were analyzed through an electromagnetic-thermal-structural finite element model.To ensure that the results are applicable to RPECT,the thermal energy conversion ability and thermal-force roll crown control ability under different lengths of the induction zone were analyzed.It was found that whether the current density regulation mode or the current frequency regulation mode is adopted,the cases with 20 or 25 mm radius of the induction zone have the great thermal energy conversion ability and the good thermal-force roll crown control ability.The reasonable adjustment of the length of the induction zone can reduce the radius required for the maximum energy efficiency regulation.Combined with the results of the simulation analysis,the optimization of ES based on the control ability maximization requirement is achieved,which provides the base for the design and configuration of ES in RPECT.
基金This work was financially supported by the National Natural Science Foundation of China(51774139)the Natural Science Foundation of Hebei Province(E2020209014).
文摘Donghua steel continuous casting-rolling(DSCCR)production line is an endless rolling production line independently integrated and developed by a Chinese enterprise.To solve the problem of out-of-control shape caused by insufficient adjustment ability of work roll bending force of DSCCR finishing mills,the backup roll contours of the finishing mill were optimized considering multi-objectives,and varying contact length backup roll(VCR)contours matching the concave work rolls were designed based on the rapid rolls-strip deformation calculation model,which integrated the elastic deformation of the rolls and the fast plastic deformation of the rolled strip.The simulation results showed that VCR configuration can significantly enhance the control efficiency of the work roll bending force and increase the transverse stiffness of rolling mill compared with the conventional roll configuration.In addition,VCR backup roll can improve the contact state between backup roll and work roll.The industrial production showed that the bending force setup value was more reasonable and the strip crown control accuracy was improved.
基金This work was financially supported by the National Natural Science Foundation of China （51674028）, and Fundamental Research Funds for the Central Universities （FRF-IC- 16-001）.
文摘The method to predict roll deformation precisely and efficiently is vital for the strip shape control of a six-high rolling mill. Traditional calculation methods of roll deformation, such as the finite element method and the influence function method, have been widely used due to their accuracies. However, the required calculation time is too long to be applied to the realtime control. Therefore, a rapid calculation method for predicting roll deformation of a six-high rolling mill was proposed, which employed the finite difference method to calculate the roll deflection and used a polynomial to describe the nonlinear relationship between roll flattening and roll contact pressure. Furthermore, a new correction strategy was proposed in the iteration, where the roll center flattening and the roll flattening deviation were put forward and corrected simultaneously in the iteration process according to the static equilibrium of roll. Finally, by the comparison with traditional methods, the proposed method was proved to be more efficient and it was suitable for the online calculation of the strip shape control.
基金This work was supported by the National Natural Science Foundation of China(No.52004029)the Fundamental Research Funds for the Central Universities(FRF-TT-20-06).
文摘Mechanical performance prediction is the key to the transformation and upgrading of steel enterprises to intelligent manufacturing.Due to time-varying manufacturing data,the traditional prediction model of mechanical properties of hotrolled strip may cause performance degradation or even failure in its use.An MDA-JITL model was thus proposed to handle the modeling problem of complex time-varying data.Relevant parameters were first chosen and normalized.Then,a distance measurement method combining the importance of data attributes and time characteristics was designed to select the most suitable samples for on-line local modeling.After that,using the chosen dataset,a linear local model was created to predict target sample.Finally,an uncertainty evaluation method was designed to evaluate the uncertainty of prediction results.Furthermore,the appropriate dataset partition and off-line simulation experiment scheme were created based on the peculiarities of hot-rolling production.The suggested model performs much better than the classic global model when applied to actual production data from a steel plant.The stability of its prediction accuracy is demonstrated in a simulation prediction for up to five months.Moreover,there is a high link between the uncertainty evaluation metrics and the prediction error of the model,reducing the field sampling rate by 30%in industrial applications in the latest year.
基金This work was financially supported by the National Natural Science Foundation of China(52004029)the Open Foundation of Engineering Research Centre for Metallurgical Automation and Measurement Technology of Ministry of Education(MADTOF-2019A01)the Postdoctoral Science Foundation of China(2021M690352).
文摘To reveal the generation mechanism of the quarter buckle in the process of hot-rolled temper rolling,the elastic–plastic finite element method is used to calculate the deformation of the roll and the strip during the temper rolling.The change of the cross section of the strip and the distribution of the longitudinal stress are analysed under different bending forces to obtain the boundary conditions of the quarter buckle.The generation of the quarter buckle is further analysed from the bending and contact flattening of the roll system and the elastic recovery of the strip after rolling.We found that the quarter buckle is closely related to the high-order distribution of the contact flattening between the work roll and the strip and is less affected by the bending deformation of the roll and the contact flattening between the rolls.Finally,a new work roll contour of the locally variable crown is proposed to change the distribution of contact flattening between the work roll and the strip.It is verified through theoretical calculations and industrial applications that the new contour can effectively improve the quarter buckle.
基金supported by Program for New Century Excellent Talents in University of Ministry of Education of China(No.NCET-10-0223)the Fundamental Research Funds for the Central Universities of China(No.FRF-TP-11-003A)the 2012 National Science and Technology Support Program of China(No.2012BAF09B04)
文摘With the increasing demand on higher strip quality, the profile and flatness of hot rolling strips have become subjects of concern, particularly for compact strip product(CSP) hot strip mills. Based on the roll contour, control model, and rolling process, a comprehensive shape control technology is proposed and applied to CSP hot strip mill of Lianyuan steel, which includes optimization and design of the work roll contour and varying contact back-up roll(VCR) plus backup roll contour, analysis of the flatness feedback control model, as well as improvement of the rolling process control system. The application of the technology has significantly improved the shape control performance. The roll wear is improved and the general roll consumption of the finishing mill is reduced by 29.86%.The percentages that satisfy the control target ranges of the average strip flatness and crown are increased by approximately 15.40%and 14.82%, respectively. The rejection rate of grade Q235 due to shape quality problem is reduced monthly by 39.69%, which creates significant economic benefits for the plant.
基金supported by the National Natural Science Foundation of China(No.52004029)the Joint Doctoral Program of China Scholarship Council(CSC)(202006460073)Liuzhou Science and Technology Plan Project,China(2021AAD0102).
文摘The cross-section profile is a key signal for evaluating hot-rolled strip quality,and ignoring its defects can easily lead to a final failure.The characteristics of complex curve,significant irregular fluctuation and imperfect sample data make it a challenge of recognizing cross-section defects,and current industrial judgment methods rely excessively on human decision making.A novel stacked denoising autoencoders(SDAE)model optimized with support vector machine(SVM)theory was proposed for the recognition of cross-section defects.Firstly,interpolation filtering and principal component analysis were employed to linearly reduce the data dimensionality of the profile curve.Secondly,the deep learning algorithm SDAE was used layer by layer for greedy unsupervised feature learning,and its final layer of back-propagation neural network was replaced by SVM for supervised learning of the final features,and the final model SDAE_SVM was obtained by further optimizing the entire network parameters via error back-propagation.Finally,the curve mirroring and combination stitching methods were used as data augmentation for the training set,which dealt with the problem of sample imbalance in the original data set,and the accuracy of cross-section defect prediction was further improved.The approach was applied in a 1780-mm hot rolling line of a steel mill to achieve the automatic diagnosis and classification of defects in cross-section profile of hot-rolled strip,which helps to reduce flatness quality concerns in downstream processes.
