With the development of automation and informatization in the steelmaking industry,the human brain gradually fails to cope with an increasing amount of data generated during the steelmaking process.Machine learning te...With the development of automation and informatization in the steelmaking industry,the human brain gradually fails to cope with an increasing amount of data generated during the steelmaking process.Machine learning technology provides a new method other than production experience and metallurgical principles in dealing with large amounts of data.The application of machine learning in the steelmaking process has become a research hotspot in recent years.This paper provides an overview of the applications of machine learning in the steelmaking process modeling involving hot metal pretreatment,primary steelmaking,secondary refining,and some other aspects.The three most frequently used machine learning algorithms in steelmaking process modeling are the artificial neural network,support vector machine,and case-based reasoning,demonstrating proportions of 56%,14%,and 10%,respectively.Collected data in the steelmaking plants are frequently faulty.Thus,data processing,especially data cleaning,is crucially important to the performance of machine learning models.The detection of variable importance can be used to optimize the process parameters and guide production.Machine learning is used in hot metal pretreatment modeling mainly for endpoint S content prediction.The predictions of the endpoints of element compositions and the process parameters are widely investigated in primary steelmaking.Machine learning is used in secondary refining modeling mainly for ladle furnaces,Ruhrstahl–Heraeus,vacuum degassing,argon oxygen decarburization,and vacuum oxygen decarburization processes.Further development of machine learning in the steelmaking process modeling can be realized through additional efforts in the construction of the data platform,the industrial transformation of the research achievements to the practical steelmaking process,and the improvement of the universality of the machine learning models.展开更多
The steelmaking process scheduling problem by considering variable electricity price （SMSPVEP） was in- vestigated. A decomposition approach was proposed for the SMSPVEP. At the first stage, mathematical program-ming...The steelmaking process scheduling problem by considering variable electricity price （SMSPVEP） was in- vestigated. A decomposition approach was proposed for the SMSPVEP. At the first stage, mathematical program-ming was utilized to minimize the maximum completion time for each cast without considering variable electricity price. At the second stage, based on obtained relative schedules of all casts, a mathematical model was formulated with an objective of minimizing the energy cost for all casts scheduling problem. The two-stage models were tested on randomly generated instances based on the practical process in a Chinese steelmaking plant. Computational results demonstrate the effectiveness of the proposed approach.展开更多
The models, algorithms and implementation results of a computerized scheduling system were introduced for the steelmaking-continuous casting process （SCCP） of a steel plant in China. The scheduling of SCCP in this p...The models, algorithms and implementation results of a computerized scheduling system were introduced for the steelmaking-continuous casting process （SCCP） of a steel plant in China. The scheduling of SCCP in this plant required that each cast plan should be processed on time, the charges in the same cast should be processed con- tinuously on the same caster, and the waiting time of the charges which are in front of each caster cannot exceed the given threshold. At the same time, the processing time of charges cannot be conflicted mutually in the same convert- ers or refining furnaces. Based on the research background, a hybrid optimal scheduling approach and its application were discussed. Aiming at the main equipment scheduling, an optimal scheduling method was proposed which con- sisted of equipment assignment algorithm based on dynamic program （DP） technique and conflict elimination algo rithm based on linear program （LP） technique. The approach guarantees that the charges are continuously processed on the same caster. Meanwhile, the requirement for high temperature ladle can also be satisfied due to the ladle matching function. Numerical results demonstrate solution quality, computational efficiency, and values of the mod els and algorithm.展开更多
To promote sustainability, it has become increasingly vital to properly account material and energy flows in industrial production processes. Therefore, a generic process-level input-output （IO） model was developed ...To promote sustainability, it has become increasingly vital to properly account material and energy flows in industrial production processes. Therefore, a generic process-level input-output （IO） model was developed to provide an integrated energy （material） accounting and analysis approach for industrial production processes. By extending the existing processlevel IO models, the production, usage, export and loss of by-products were explicitly considered in the proposed IO model. Moreover, the by-products allocation procedures were incorporated into the proposed IO model to reflect individual contributions of products to energy consumption. Finally, the proposed model enabled calculating embodied energy of main products and total energy consumption under hierarchical accounting scope. Plant managers, energy management consultants, governmental officials and academic researchers could use this input-output model to account material and energy flows, thus calculating energy consumption indicators of a production process with their specific system boundary requirements. The accounting results could be further used for energy labeling, identifying bottlenecks of production activities, evaluating industrial symbiosis effects, improving materials and energy utilization efficiency, etc. The model could also be used as a planning tool to determine the effect that a particular change of technology and supply chains may have on the industrial production processes. The proposed model was tested and applied in a real integrated steel mill, which also provided the reference results for related researches. At last, some concepts, computational issues and limi- tations of the proposed model were discussed.展开更多
Online monitoring and diagnosis of production processes face great challenges due to the nonlinearity and multivariate of complex industrial processes.Traditional process monitoring methods employ kernel function or m...Online monitoring and diagnosis of production processes face great challenges due to the nonlinearity and multivariate of complex industrial processes.Traditional process monitoring methods employ kernel function or multilayer neural networks to solve the nonlinear mapping problem of data.However,the above methods increase the model complexity and are not interpretable,leading to difficulties in subsequent fault recognition/diagnosis/location.A process monitoring and diagnosis method based on the free energy of Gaussian-Bernoulli restricted Boltzmann machine(GBRBM-FE)was proposed.Firstly,a GBRBM network was established to make the probability distribution of the reconstructed data as close as possible to the probability distribution of the raw data.On this basis,the weights and biases in GBRBM network were used to construct F statistics,which represents the free energy of the sample.The smaller the energy of the sample is,the more normal the sample is.Therefore,F statistics can be used to monitor the production process.To diagnose fault variables,the F statistic for each sample was decomposed to obtain the Fv statistic for each variable.By analyzing the deviation degree between the corresponding variables of abnormal samples and normal samples,the cause of process abnormalities can be accurately located.The application of converter steelmaking process demonstrates that the proposed method outperforms the traditional methods,in terms of fault monitoring and diagnosis performance.展开更多
Aiming at the characteristics of the practical steelmaking process, a hybrid model based on ladle heat sta- tus and artificial neural network has been proposed to predict molten steel temperature. The hybrid model cou...Aiming at the characteristics of the practical steelmaking process, a hybrid model based on ladle heat sta- tus and artificial neural network has been proposed to predict molten steel temperature. The hybrid model could over- come the difficulty of accurate prediction using a single mathematical model, and solve the problem of lacking the consideration of the influence of ladle heat status on the steel temperature in an intelligent model. By using the hybrid model method, forward and backward prediction models for molten steel temperature in steelmaking process are es- tablished and are used in a steelmaking plant. The forward model, starting from the end-point of BOF, predicts the temperature in argon-blowing station, starting temperature in LF, end temperature in LF and tundish temperature forwards, with the production process evolving. The backward model, starting from the required tundish tempera- ture, calculates target end temperature in LF, target starting temperature in LF, target temperature in argon-blo- wiag station and target BOF end-point temperature backwards. Actual application results show that the models have better prediction accuracy and are satisfying for the process of practical production.展开更多
In this study,the formation mechanism and removal efficiency of polychlorinated dibenzo-p-dioxins, dibenzofurans（PCDD/Fs） achieved with bag filter（BF） in an electric arc furnace（EAF） in Taiwan is evaluated via i...In this study,the formation mechanism and removal efficiency of polychlorinated dibenzo-p-dioxins, dibenzofurans（PCDD/Fs） achieved with bag filter（BF） in an electric arc furnace（EAF） in Taiwan is evaluated via intensive stack sampling and analysis.The results indicate that the PCDD/F concentration measured in the stack gas of the EAF investigated was 0.16 ng I-TEQ/m1,which was significantly lower than the PCDD/F emission limit（0.5 ng I-TEQ/m;） set for existing EAFs in Taiwan.Due to the low operating temperature（<60℃） of the BF adopted by the EAF investigated,76%of total toxicity PCDD/Fs could be removed from the flue gas stream. In addition,the partitioning of PCDD/Fs between vapor and solid phase at different locations in EAF did not change significantly,while a reduction of solid-phase PCDD/Fs was observed at the outlet of BF.As the chlorination level of PCDD/Fs congener increases,the vapor pressure of PCDD/F congener decreases,resulting in the increase of PCDD/Fs existing in solid phase.Hence,the removal efficiencies of highly chlorinated congeners were significantly higher than that of lowly chlorinated congeners.PCDFs accounted for more than 85%of the TEQ in flue gas of the EAF investigated,among them 2,3,4,7,8-PeCDF（> 43%） was of the highest contribution.Overall,2,3,7,8-TCDD,2,3,7,8-TCDF and 2,3,4,7,8-PeCDF can serve as the unique congeners in the flue gas of the stainless steel EAF process.展开更多
基金supported by the National Natural Science Foundation of China(No.U1960202)。
文摘With the development of automation and informatization in the steelmaking industry,the human brain gradually fails to cope with an increasing amount of data generated during the steelmaking process.Machine learning technology provides a new method other than production experience and metallurgical principles in dealing with large amounts of data.The application of machine learning in the steelmaking process has become a research hotspot in recent years.This paper provides an overview of the applications of machine learning in the steelmaking process modeling involving hot metal pretreatment,primary steelmaking,secondary refining,and some other aspects.The three most frequently used machine learning algorithms in steelmaking process modeling are the artificial neural network,support vector machine,and case-based reasoning,demonstrating proportions of 56%,14%,and 10%,respectively.Collected data in the steelmaking plants are frequently faulty.Thus,data processing,especially data cleaning,is crucially important to the performance of machine learning models.The detection of variable importance can be used to optimize the process parameters and guide production.Machine learning is used in hot metal pretreatment modeling mainly for endpoint S content prediction.The predictions of the endpoints of element compositions and the process parameters are widely investigated in primary steelmaking.Machine learning is used in secondary refining modeling mainly for ladle furnaces,Ruhrstahl–Heraeus,vacuum degassing,argon oxygen decarburization,and vacuum oxygen decarburization processes.Further development of machine learning in the steelmaking process modeling can be realized through additional efforts in the construction of the data platform,the industrial transformation of the research achievements to the practical steelmaking process,and the improvement of the universality of the machine learning models.
基金Item Sponsored by National Natural Science Foundation of China (71171038,71021061 )Fundamental Research Funds for Central Universities of China (N100504001)
文摘The steelmaking process scheduling problem by considering variable electricity price （SMSPVEP） was in- vestigated. A decomposition approach was proposed for the SMSPVEP. At the first stage, mathematical program-ming was utilized to minimize the maximum completion time for each cast without considering variable electricity price. At the second stage, based on obtained relative schedules of all casts, a mathematical model was formulated with an objective of minimizing the energy cost for all casts scheduling problem. The two-stage models were tested on randomly generated instances based on the practical process in a Chinese steelmaking plant. Computational results demonstrate the effectiveness of the proposed approach.
基金Item Sponsored by National Natural Science Foundation of China(61174187,71021061,60974091,61104174)Startup Fund of Northeastern University of China(29321006)Basic Scientific Research Foundation of Northeast University of China(N110208001)
文摘The models, algorithms and implementation results of a computerized scheduling system were introduced for the steelmaking-continuous casting process （SCCP） of a steel plant in China. The scheduling of SCCP in this plant required that each cast plan should be processed on time, the charges in the same cast should be processed con- tinuously on the same caster, and the waiting time of the charges which are in front of each caster cannot exceed the given threshold. At the same time, the processing time of charges cannot be conflicted mutually in the same convert- ers or refining furnaces. Based on the research background, a hybrid optimal scheduling approach and its application were discussed. Aiming at the main equipment scheduling, an optimal scheduling method was proposed which con- sisted of equipment assignment algorithm based on dynamic program （DP） technique and conflict elimination algo rithm based on linear program （LP） technique. The approach guarantees that the charges are continuously processed on the same caster. Meanwhile, the requirement for high temperature ladle can also be satisfied due to the ladle matching function. Numerical results demonstrate solution quality, computational efficiency, and values of the mod els and algorithm.
文摘To promote sustainability, it has become increasingly vital to properly account material and energy flows in industrial production processes. Therefore, a generic process-level input-output （IO） model was developed to provide an integrated energy （material） accounting and analysis approach for industrial production processes. By extending the existing processlevel IO models, the production, usage, export and loss of by-products were explicitly considered in the proposed IO model. Moreover, the by-products allocation procedures were incorporated into the proposed IO model to reflect individual contributions of products to energy consumption. Finally, the proposed model enabled calculating embodied energy of main products and total energy consumption under hierarchical accounting scope. Plant managers, energy management consultants, governmental officials and academic researchers could use this input-output model to account material and energy flows, thus calculating energy consumption indicators of a production process with their specific system boundary requirements. The accounting results could be further used for energy labeling, identifying bottlenecks of production activities, evaluating industrial symbiosis effects, improving materials and energy utilization efficiency, etc. The model could also be used as a planning tool to determine the effect that a particular change of technology and supply chains may have on the industrial production processes. The proposed model was tested and applied in a real integrated steel mill, which also provided the reference results for related researches. At last, some concepts, computational issues and limi- tations of the proposed model were discussed.
基金the financial support from the National Key R&D Program of China(Grant No.2020YFA0405700).
文摘Online monitoring and diagnosis of production processes face great challenges due to the nonlinearity and multivariate of complex industrial processes.Traditional process monitoring methods employ kernel function or multilayer neural networks to solve the nonlinear mapping problem of data.However,the above methods increase the model complexity and are not interpretable,leading to difficulties in subsequent fault recognition/diagnosis/location.A process monitoring and diagnosis method based on the free energy of Gaussian-Bernoulli restricted Boltzmann machine(GBRBM-FE)was proposed.Firstly,a GBRBM network was established to make the probability distribution of the reconstructed data as close as possible to the probability distribution of the raw data.On this basis,the weights and biases in GBRBM network were used to construct F statistics,which represents the free energy of the sample.The smaller the energy of the sample is,the more normal the sample is.Therefore,F statistics can be used to monitor the production process.To diagnose fault variables,the F statistic for each sample was decomposed to obtain the Fv statistic for each variable.By analyzing the deviation degree between the corresponding variables of abnormal samples and normal samples,the cause of process abnormalities can be accurately located.The application of converter steelmaking process demonstrates that the proposed method outperforms the traditional methods,in terms of fault monitoring and diagnosis performance.
基金Item Sponsored by Fundamental Research Funds for Central Universities of China(FRF-BR-10-027B)
文摘Aiming at the characteristics of the practical steelmaking process, a hybrid model based on ladle heat sta- tus and artificial neural network has been proposed to predict molten steel temperature. The hybrid model could over- come the difficulty of accurate prediction using a single mathematical model, and solve the problem of lacking the consideration of the influence of ladle heat status on the steel temperature in an intelligent model. By using the hybrid model method, forward and backward prediction models for molten steel temperature in steelmaking process are es- tablished and are used in a steelmaking plant. The forward model, starting from the end-point of BOF, predicts the temperature in argon-blowing station, starting temperature in LF, end temperature in LF and tundish temperature forwards, with the production process evolving. The backward model, starting from the required tundish tempera- ture, calculates target end temperature in LF, target starting temperature in LF, target temperature in argon-blo- wiag station and target BOF end-point temperature backwards. Actual application results show that the models have better prediction accuracy and are satisfying for the process of practical production.
文摘In this study,the formation mechanism and removal efficiency of polychlorinated dibenzo-p-dioxins, dibenzofurans（PCDD/Fs） achieved with bag filter（BF） in an electric arc furnace（EAF） in Taiwan is evaluated via intensive stack sampling and analysis.The results indicate that the PCDD/F concentration measured in the stack gas of the EAF investigated was 0.16 ng I-TEQ/m1,which was significantly lower than the PCDD/F emission limit（0.5 ng I-TEQ/m;） set for existing EAFs in Taiwan.Due to the low operating temperature（<60℃） of the BF adopted by the EAF investigated,76%of total toxicity PCDD/Fs could be removed from the flue gas stream. In addition,the partitioning of PCDD/Fs between vapor and solid phase at different locations in EAF did not change significantly,while a reduction of solid-phase PCDD/Fs was observed at the outlet of BF.As the chlorination level of PCDD/Fs congener increases,the vapor pressure of PCDD/F congener decreases,resulting in the increase of PCDD/Fs existing in solid phase.Hence,the removal efficiencies of highly chlorinated congeners were significantly higher than that of lowly chlorinated congeners.PCDFs accounted for more than 85%of the TEQ in flue gas of the EAF investigated,among them 2,3,4,7,8-PeCDF（> 43%） was of the highest contribution.Overall,2,3,7,8-TCDD,2,3,7,8-TCDF and 2,3,4,7,8-PeCDF can serve as the unique congeners in the flue gas of the stainless steel EAF process.
