Prediction Of Silicon Content Research Articles

Deep learning for robust forecasting of hot metal silicon content in a blast furnace

AbstractThe hot metal silicon content is a key indicator of the thermal state in the blast furnace and it needs to be kept within a pre-defined range in order to ensure efficient operations. Effective monitoring of silicon content is challenging due to the harsh environment in the furnace and irregularly sampled measurements. Data-driven approaches have been proposed in the literature to predict silicon content using process data and overcome the sparsity of silicon content measurements. However, these approaches rely on the selection of hand-crafted features and ad hoc interpolation methods to deal with irregular sampling of the process variables, adding complexity to model training and optimisation, and requiring significant effort when tuning the model over time to keep it to the required level of accuracy. This paper proposes an improved framework for the prediction of silicon content using a novel deep learning approach based on Phased LSTM. The model has been trained using 3 years of data and validated over a 1-year period using a robust walk-forward validation method, therefore providing confidence in the model performance over time. The Phased LSTM model outperforms competing approaches due to its in-built ability to learn from event-based sequences and scalability for real-world deployments. This is the first time that Phased LSTM has been applied to real-world datasets and results suggest that the ability to learn from event-based data can be beneficial for the process industry where event-driven signals from multiple sensors are common.

Deep Learning of Partially Labeled Data for Quality Prediction Based on Stacked Target-Related Laplacian Autoencoder.

Partially labeled data, which is common in industrial processes due to the low sampling rate of quality variables, remains an important challenge in soft sensor applications. In order to exploit the information from partially labeled data, a target-related Laplacian autoencoder (TLapAE) is proposed in this work. In TLapAE, a novel target-related Laplacian regularizer is developed, which aims to extract structure-preserving and quality-related features by preserving the feature-target mapping according to the local geometrical structure of the data. In addition, stacked TLapAE (STLapAE) is further constructed to extract deep feature representations of the data by hierarchically stacking TLapAE blocks. For model training, backward propagation equations are derived based on matrix calculus techniques to update the model parameters of the proposed TLapAE. The effectiveness of the proposed STLapAE is evaluated using the butane content prediction case in a debutanizer column, the silicon content prediction case in a blast furnace (BF) ironmaking process, and the ethane concentration prediction case in an ethylene fractionator. The results show that the proposed TLapAE model has significantly improved prediction accuracy compared to soft sensors using only labeled data and other partially labeled data modeling methods.

Prediction of Silicon Content of Hot Metal in Blast Furnace Based on Optuna-GBDT

Prediction of Silicon Content of Hot Metal in Blast Furnace Based on Optuna-GBDT

Improved algorithm of extreme gradient boosting for predicting silicon content in large proportion pellet smelting process

The silicon content in molten iron is an important indicator, to characterize the temperature of blast furnace (BF) and the quality of molten iron, which is of great significance to the stable operation of large proportion pellets in the BF smelting. Aiming at the problem of poor prediction performance and insufficient accuracy of silicon content, a prediction model of silicon content in molten iron was established based on KMeans++ and improved XGBoost algorithm to divide the information from different BF conditions in the smelting process, The genetic algorithm(GA) was adopted to optimize the model iteratively, which improved the accuracy of the results and reduced the training time for optimal results. The experimental result showed that the prediction hit of the model was improved by clustering the data and reached above 90% on average, and the accurate prediction of silicon content in molten iron in case of a large proportion of pellets of BF smelting was realized.

Grated Recurrent Unit Network Quantile Regression for Silicon Content Prediction in Blast Furnace

Grated Recurrent Unit Network Quantile Regression for Silicon Content Prediction in Blast Furnace

Prediction of Silicon Content in the Hot Metal of a Blast Furnace Based on FPA-BP Model

In the process of blast furnace smelting, the stability of the hearth thermal state is essential. According to the analysis of silicon content in hot metal and its change trend, the operation status of the blast furnace can be judged to ensure the stable and smooth operation of the blast furnace. Based on the error back-propagation neural network (BP), the flower pollination algorithm (FPA) is used to optimize the weight and threshold of the BP neural network, and the prediction model of silicon content is established. At the same time, the principal component analysis method is used to reduce the dimension of the input sequence to obtain relevant indicators. The relevant indicators are used as the input, and silicon content in the hot metal is used as the output, which is substituted into the model for training and utilizes the trained model to predict. The results show that the hit rate of the prediction model is 16% higher than the non-optimized BP prediction model. At the same time, the evaluation indicators and operation speed of the model are improved compared with the BP prediction model, which can be more accurately applied to predict the silicon content of the hot metal.

Cascade model for continuous prediction of silicon content of molten iron with coupled state variable nodes

Cascade model for continuous prediction of silicon content of molten iron with coupled state variable nodes

Collaborative Extraction of Intervariable Coupling Relationships and Dynamics for Prediction of Silicon Content in Blast Furnaces

Soft sensor of silicon content in hot metal is challenging due to the nonlinearity, dynamics, and non-Euclidean coupling relationship between process variables. The inter-variable coupling relationship widely exists in the ironmaking process. Nevertheless, they are generally overlooked in modeling. To solve this issue, this paper proposes a novel temporal graph convolutional network with supervised graphs (TGCN-S). Different from traditional soft sensor methods, TGCN-S explicitly models the inherent non-Euclidean and irregular coupling relationships in the ironmaking process. First, a graph structure learning module is designed, which can adaptively learn potential inter-variable relationships from data. This module is embedded in TGCN-S to learn the supervised graph structures in an end-to-end manner. Second, a novel methodological framework is proposed based on the supervised graph structures, which can collaboratively extract the inter-variable coupling relationships and intra-variable dynamics. In this structure, information between process variables is selectively aggregated while taking into account the dynamics within variables. Finally, experiments based on the real ironmaking process demonstrate the effectiveness of the proposed method.

A Context-Aware Enhanced GRU Network With Feature-Temporal Attention for Prediction of Silicon Content in Hot Metal

Blast furnace ironmaking is one of the most complicated industrial processes. As an essential reference index for blast furnace operation, the prediction of silicon content in hot metal is important. Most previous works focus on the dynamics and nonlinearity of the process without comprehensively considering the correlation between the process variables and the silicon content. Besides, the timing mismatch of input and output variables caused by the inertia of the process still cannot be handled effectively. To solve these problems, in this article, the proposed model puts extra attention on input variables to strengthen the information of key variables, and introduces causal convolution-based self-attention to incorporate local context into attention mechanism in the temporal dimension, which realizes local awareness enhancement and variables soft alignment. With a series of theoretical and practical verification, the hybrid model shows significant improvement at hit rate and mean-square error.

A Multiobjective Evolutionary Nonlinear Ensemble Learning With Evolutionary Feature Selection for Silicon Prediction in Blast Furnace.

In the blast furnace ironmaking process, accurate prediction of silicon content in molten iron is of great significance for maintaining stable furnace conditions, improving hot metal quality, and reducing energy consumption. However, most of the current research works employ linear correlation coefficient methods to select input features in modeling, which may not fully take the nonlinear and coupling relationships between features into account. Therefore, this article considers the input feature selection issue of silicon content prediction model from a new perspective and proposes a multiobjective evolutionary nonlinear ensemble learning model with evolutionary feature selection mechanism (MOENE-EFS), in which extreme learning machine is adopted as the base learner. MOENE-EFS takes the input feature scheme of each base learner as well as their network structure and parameters as decision variables and proposes a modified nondominated sorting differential evolution algorithm to optimize two conflicting objectives, i.e., accuracy and diversity of base learners, simultaneously. Through the optimization, a set of Pareto optimal base learners with high accuracy and strong diversity can be obtained. Moreover, different from the linear ensemble methods commonly used in classical evolutionary ensemble learning, this article proposes a nonlinear ensemble method to combine the obtained base learners based on differential evolution. Experimental results indicate that the two proposed strategies, i.e., evolutionary feature selection and nonlinear ensemble, are very effective in improving the accuracy and stability of the prediction model. MOENE-EFS also outperforms the other prediction models in both benchmark data and practical industrial data. Furthermore, analysis on the input features of all Pareto optimal base learners shows that the evolutionary feature selection is capable of selecting essential features and is consistent with human experience, which indicates it is a promising method to deal with the input feature selection issue in silicon content prediction.

Temporal Hypergraph Attention Network for Silicon Content Prediction in Blast Furnace

Online dynamic prediction of the hot metal silicon content in the blast furnace ironmaking process is crucial for stabilizing the furnace condition and improving the molten iron quality. However, due to the complex nonlinear correlations and time-varying time lags between silicon content and process variables, silicon content prediction is a challenging task. To tackle the problem, we propose a novel silicon content prediction method, called temporal hypergraph attention network (T-HyperGAT), which is combined the hypergraph attention network (HyperGAT) and the gated recurrent unit (GRU) network. Specifically, the Hyper-GAT is used to capture the high-order correlations of input features and perform equal-dimensional feature transformation to maintain the temporality of input features, and the GRU network is used to capture the time-series characteristics of transformed input features to overcome the time-varying time lags of process variables. Then the T-HyperGAT model can capture high-order correlations and time-series characteristics from complex industrial data. The effectiveness of the proposed T-HyperGAT method is verified by actual blast furnace ironmaking process data from a blast furnace in China.

Artificial Neural Network for Predicting Silicon Content in the Hot Metal Produced in a Blast Furnace Fueled by Metallurgical Coke

The main production route for cast iron and steel is through the blast furnace. The silicon content in cast iron is an important indicator of the thermal condition of a blast furnace. High silicon contents indicate an increase in the furnace's thermal input and, in some cases, may indicate an excess of coke in the reactor. As coke costs predominate in the production of cast iron, tighter control of the silicon content therefore has economic advantages. The main objective of this article was to design an artificial neural network to predict the silicon content in hot metal, varying the number of neurons in the hidden layer by 10, 20, 25, 30, 40, 50, 75, 100, 125 , 150, 170 and 200 neurons. In general, all neural networks showed excellent results, with the network with 30 neurons showing the best results among the 12 modeled networks. The validation of the models was confirmed using the Mean Square Error (MSE) and Pearson's correlation coefficient. The cross-validation technique was used to re-evaluate the performance of neural networks. In short, neural networks can be used in practical operations due to the excellent correlations between the real values ​​and those calculated by the neural network.

Predictive modeling of the hot metal silicon content in blast furnace based on ensemble method

The silicon content of the hot metal is not only an important indicator of the quality of the hot metal and blast furnace (BF) operation but also reflects the level of energy utilization and the thermal state within a BF. It is important to develop an accurate prediction model for hot metal silicon content. In present study, two models for predicting hot metal silicon content are developed based on two ensemble learning methods, random forest regression (RFR) and extreme gradient boosting (XGBoost). First, Box plot was used to visualize the collected data and determine extreme outliers in the raw data. Extreme outliers are replaced with null value, and all null value are filled by linear interpolation. Secondly, feature selection is performed using recursive feature elimination. Cross-validation is performed to optimize machine learning hyperparameters and having a robust accuracy measure. Based on this, two hot metal silicon content prediction models are developed. Finally, the prediction results of the two models are compared and evaluated. The results show that both ensemble learning models show good prediction performance in predicting hot metal silicon content, but the prediction performance of the RFR model is better than that of the XGBoost model and reaching 98.77%.

Prediction of silicon content in the hot metal using Bayesian networks and probabilistic reasoning

The blast furnace is the principal method of producing cast iron. In the production of cast iron, the control of silicon is vital because this impurity is harmful to almost all steels. Artificial neural networks with Bayesian regularization are more robust than traditional back-propagation networks and can reduce or eliminate the need for tedious cross-validation. Bayesian regularization is a mathematical process that converts a nonlinear regression into a "well-posed" statistical problem in the manner of ridge regression. The main objective of this work was to develop an artificial neural network to predict silicon content in hot metal by varying the number of neurons in the hidden layer by 10, 20, 25, 30, 40, 50, 75, and 100 neurons. The results show that all neural networks converged and presented reliable results, neural networks with 20, 25, and 30 neurons showed the best overall results. However, In short, Bayesian neural networks can be used in practice because the actual values correlate excellently with the values calculated by the neural network.

Prediction Model of Hot Metal Silicon Content Based on Improved GA-BPNN.

The inconsistency of the detection period of blast furnace data and the large time delay of key parameters make the prediction of the hot metal silicon content face huge challenges. Aiming at the problem that the hot metal silicon content is not consistent with the detection period of time series of multiple control parameters, the cubic spline interpolation fitting model was used to realize the data integration of multiple detection periods. The large time delay of the blast furnace iron making process was analyzed. Moreover, Spearman analysis was combined with the weighted moving average method to optimize the data set of silicon content prediction. Aiming at the problem of low prediction accuracy of the ordinary neural network model, genetic algorithm was used to optimize parameters on the BP neural network model to improve the convergence speed of the model to achieve global optimization. Combined with the autocorrelation analysis of the hot metal silicon content, a modified model for the prediction of hot metal silicon content based on error analysis was proposed to further improve the accuracy of the prediction. The model comprehensively considers problems such as data detection inconsistency, large time delay, and inaccuracy of prediction results. Its average absolute error is 0.05009, which can be used in actual production.

Double-hyperplane fuzzy classifier design for tendency prediction of silicon content in molten iron

It is very difficult to control a blast furnace efficiently because of the characteristics of the closed smelting and the system's large delay. It is very helpful for blast furnace control to establish a model that can accurately predict the tendency of silicon content in molten iron. In this article, a nonparallel hyperplane–based fuzzy classifier is proposed reconciling the model accuracy with the model interpretability. Each hyperplane is as close as possible to one class of samples and as far as possible from the other class of samples. This leads to two small quadratic programming problems instead of a large one as a support vector machine–based fuzzy classifier. This makes our classifier faster than a support vector machine–based fuzzy classifier. We evaluate the proposed method on blast furnace data and synthetic data, which verifies the effectiveness.

Domain knowledge based explainable feature construction method and its application in ironmaking process

Data driven industrial modeling has been comprehensively studied for its high modeling accuracy. However, the unexplainable characteristic of data-driven modeling hinders workers from understanding the model and controlling process, and further holds back its application in industrial process. In order to solve this problem, we propose a genetic algorithm based method to construct interpretable features for industrial modeling in this paper. This model adopts the framework similar to genetic algorithm, but redefines the populations as features to adapt to the task of feature construction. The populations are evaluated by fitness function with punishment term to ensure the constructed features are concise. By using different genetic mutation and crossover operators, the proposed framework has the ability to combine domain knowledge to handle the characteristics of data, such as nonlinear, dynamic and time lag. The proposed method is experimented on the silicon content prediction task in ironmaking process, which is a classical process industry, achieving high accuracy.

Silicon content prediction of hot metal in blast furnace based on attention mechanism and CNN-IndRNN model

The stability of blast furnace temperature is an important condition to ensure the efficient production of hot metal. Accurate prediction of silicon content in hot metal is of great significance to the control of blast furnace temperature in iron and steel plants. At present, the accuracy of most silicon prediction models can only be guaranteed when the furnace condition is stable. However, due to many factors affecting the silicon content in hot metal of blast furnace, and there are large noises, large delays and large fluctuations in the data, the previous prediction results are of limited guiding significance to the actual production. In this paper, combined with the actual situation, the convolution neural network is used to extract the furnace condition characteristics, and then combined with the attention mechanism and the IndRNN model to get the prediction results, so that the prediction can better adapt to the fluctuating data set. The experimental results show that the prediction error of this model is lower than that of other models, which provides a new solution for the research of silicon content in hot metal of blast furnace.

Classification of blast furnace internal state based on FLS and its application in furnace temperature prediction

The real-time and accurate prediction of the molten iron silicon content of the blast furnace plays an important role in regulating the temperature of the blast furnace and stabilizing the furnace condition. When the time is large, the accuracy and credibility of the forecast results decrease rapidly, which is not conducive to on-site operators to carry out production operations according to the forecast results. To this end, this paper adds a state variable to each piece of data through the flexible least square parameter estimation method, and selects the training set in a state similar to the test sample. This makes the selection of training data more accurate and reliable. Application examples show that the method proposed in this paper improves the accuracy of silicon content prediction results and has good guiding significance for actual production operations.

Soft Sensors Based on Adaptive Stacked Polymorphic Model for Silicon Content Prediction in Ironmaking Process

The silicon content of molten iron is a key index to reveal the thermal state in the blast furnace (BF). To improve the quality of molten iron, it is important to obtain silicon content online accurately. However, owing to the harsh environment in BF hearth, it is difficult to realize the online measurement of the silicon content of molten iron. To handle this problem, a novel soft sensoring method based on adaptive stacked polymorphic model is proposed. First, an adaptive uncertain fuzzy clustering algorithm (AUFC) is utilized to obtain the distribution characteristics and reduce the complexity of the process data, thereby the training samples are divided into several sub-classes. Second, the adaptive weight echo state network (AW-ESN) and time difference AW-ESN (TD-AWESN) are proposed to establish polymorphic models. Inspired by the stacking strategy, sub-models based on stacked polymorphic model are designed, which can effectively solve the time-varying and the strong nonlinear problems in the BF ironmaking process. Third, to avoid model degradation caused by the changes of working conditions, an ensemble strategy is adopted to combine the predicted results of multiple sub-models. The experimental results in an industrial BF show that the soft sensor based on an adaptive polymorphic model can reach the required prediction accuracy and provide reliable silicon content.