Utilizing the waste heat recovered from the coke dry quenching process for power generation is an important energy-saving measure for steel enterprises. However, the discontinuous nature of the coke discharging operation leads to instability in waste heat supply and constant load fluctuations in the process of recovering the sensible heat from red hot coke, thereby affecting the power generation. Consequently, short-term prediction of power generation in coke dry quenching process is essential to provide guidance for the power scheduling of steel companies. In this paper, a prediction model is developed based on the Crossformer network. Experimental results based on real production data show that the proposed prediction model can relatively accurately predict power generation in the coke dry quenching process, which provides guidance for the short-term power scheduling of the steel company.

Severe drill-string vibration is a significant source of drilling problems. Most existing methods for vibration recognition rely on downhole data, facing great limitations in practice. In addition, complex and changeable formations result in limited ability of single-scale features to characterize the vibration. To resolve these issues, an end-to-end vibration recognition model using only surface drilling data is proposed based on multiscale features extraction. A multiscale temporal convolutional network is developed to extract multiscale temporal features of multisensor data, enhancing the vibration representation capability to adapt to complex formations. To further improve recognition capability, the bidirectional long short-term memory network is utilized to obtain contextual linkage of multiscale features. Experiments conducted with field data have verified the efficiency of proposed method. It has achieved 97% accuracy, and outperforms existing methods. Furthermore, compared with the recognition result based on single-scale features, it has improved by 6% in accuracy. The proposed method provides automated diagnostics of drill-string vibration.

Carbon consumption dynamic modeling is essential for energy saving, emission reduction, and green manufacturing of iron ore sintering process. This article proposes a novel adaptive weighted broad echo state learning system (AWBESLS) for carbon consumption dynamic prediction in the sintering process by integrating adaptive weights and a reservoir with echo state characteristics. Different from previous studies, the AWBESLS adaptively matches a weight to each production data to overcome the effects of anomalous data in production data and utilizes an echo state network (ESN) for catching the dynamic state in sintering process. Carbon consumption experiments using actual production data reveal the effectiveness of the AWBESLS and compare it with some state-of-the-art methods. The results show that the AWBESLS is superior to other methods in improving the prediction performance with lowest prediction error. In summary, the AWBESLS is an effective and applicable technique for dynamic modeling of the sintering process that is easily applicable for the modeling of other manufacturing processes.

It becomes apparent that time-series dynamic prediction for carbon consumption in sintering production process holds immense significance in the steel industry, as it plays a pivotal role in determining the efficiency and environmental impact of the operation. Given the complexities of the sintering process, encompassing multiple operating conditions, numerous parameters, nonlinearities, etc., this article proposes a time-series dynamic modeling method for carbon consumption based on an improved just-in-time learning (JITL) and a gated recurrent unit-based temporal cascade broad learning system (GRU-TCBLS). First, the data correlation analysis method is employed to determine the process parameters affecting carbon consumption. Further, an improved JITL method incorporating moving window and JITL is developed to obtain relevant training data in real-time for model training. Finally, based on these relevant training data, the GRU-TCBLS is formulated to construct a carbon consumption prediction model. Experiments based on actual production data demonstrate the superiority of the proposed method with respect to some state-of-the-art modeling methods.

Burn-through point (BTP) is an essential thermal state parameter in a sintering process, which is a direct reflection of the stability of this process. However, it cannot be measured online. Soft-sensing technology offers a reliable method for estimating unmeasurable variables in industrial processes. Here, a soft-sensing model for BTP based on weighted kernel just-in-time learning (WKJITL) and fuzzy broad-learning system (FBLS) is built. First, an abnormal production data detection and correction strategy is employed to process the production data, and the mechanism analysis and mutual information analysis are utilized to specify the detectable process variables that are directly related to BTP. Then, the WKJITL method is proposed to obtain historical production data similar to the query data of BTP for local learning modeling, and the FBLS is utilized as an efficient modeling method for the soft-sensing prediction of BTP. Finally, the results of simulation experiments based on actual sintering production data reveal that the developed soft-sensing model of BTP exhibits better prediction accuracy and efficiency compared with some advanced modeling methods. Furthermore, the proposed method is of general nature and can also be easily applied to other industrial processes.

The temperature of the cooling stave (TCS) is an important state parameter to indicate the states of the slag crust during the blast furnace ironmaking process. The state of the slag crust affects the quality and production of pig iron, and the gas flow distribution in the blast furnace. Thus, it is necessary to recognize the states of the slag crust. This article proposes a condition recognition strategy based on fuzzy clustering endowed with a novel distance with information granulation for recognizing the states of the slag crust. First, the raw TCS time-series data are split into segments according to the appropriate segmentation length, and the segments are represented in a granular form by the information granulation method. Then, information granules are clustered using fuzzy clustering endowed with a novel distance. After completing the data representation, each information granule is compounded of a lower bound and an upper bound that indicate the dynamic characteristics of the corresponding segments. In the fuzzy clustering, information granulation distance, a new distance, is established to measure the similarity between two information granules. Finally, the data experiments using the datasets from the UCR time-series database and actual industrial data from the blast furnace demonstrate the effectiveness and superiority of the proposed condition recognition strategy.

Data-driven fault diagnosis methods have been widely applied at present. In actual processes, there usually exist multiple failure modes; the data frequency spectrum varies in different failure modes, which would bring challenges for feature extraction and subsequent fault diagnosis. In this article, a decision fusion scheme based on the mode decomposition and evidence theory is proposed for fault diagnosis during drilling. The raw data are decomposed into multiple series with different center frequencies, the decomposed series are reconstructed to several groups. For each group, the local diagnosis model is established, thus, several local diagnostic results are obtained. Then, all local diagnostic results are fed into the evidence theory-based decision fusion model. Meanwhile, a confidence matrices-based weight adjustment method is designed to enhance the reliability of fused results. An industrial case study based on the actual drilling data verifies that the proposed method is beneficial to improve the diagnostic effect during drilling.

A key energy consumption in steel metallurgy comes from an iron ore sintering process. Enhancing carbon utilization in this process is important for green manufacturing and energy saving and its prerequisite is a time-series prediction of carbon efficiency. The existing carbon efficiency models usually have a complex structure, leading to a time-consuming training process. In addition, a complete retraining process will be encountered if the models are inaccurate or data change. Analyzing the complex characteristics of the sintering process, we develop an original prediction framework, that is, a weighted kernel-based fuzzy C-means (WKFCM)-based broad learning model (BLM), to achieve fast and effective carbon efficiency modeling. First, sintering parameters affecting carbon efficiency are determined, following the sintering process mechanism. Next, WKFCM clustering is first presented for the identification of multiple operating conditions to better reflect the system dynamics of this process. Then, the BLM is built under each operating condition. Finally, a nearest neighbor criterion is used to determine which BLM is invoked for the time-series prediction of carbon efficiency. Experimental results using actual run data exhibit that, compared with other prediction models, the developed model can more accurately and efficiently achieve the time-series prediction of carbon efficiency. Furthermore, the developed model can also be used for the efficient and effective modeling of other industrial processes due to its flexible structure.

To reduce energy consumption and harmful emission, it is of great significance to improve carbon efficiency in sintering process, which is able to be achieved if the carbon efficiency can be accurately predicted. In this article, the ratio of CO and CO2(CO/CO2) is taken as a measurement of the carbon efficiency. As CO/CO2is hard to measure, and there exist multiple working conditions, multiple variables, and nonlinearity, a hybrid CO/CO2prediction model is devised based on the aforementioned characteristics. First, the sintering process is analyzed, and the key characteristics to predict the CO/CO2are extracted. Next, the configuration of the prediction model is given based on the analysis. The model consists by two submodels, one is to predict the state variables by an improved just-in-time learning model, combining three neural network (NN) models. The other is to predict CO/CO2with semisupervised algorithm, based on deep belief network with a combination of the three NN regression methods. Then, the configurations of the two submodels are introduced in detail. The test results based on actual running data exhibit the good performance of the model.

Sinter ore is the main raw material for ironmaking, and burn-through point (BTP) is one of the significant factors to measure the stability of the sintering process. In this article, through the feature extraction of time-series trend, a fuzzy control strategy is presented for the BTP. First, the Hurst exponent of the time series for the BTP is calculated by resorting to the rescaled range analysis method, by which the trend feature is analyzed. Then, by using the Mann-Kendall test, both global and local trend feature variable of the time series for the BTP are extracted and regarded as the inputs of the fuzzy controller. Next, a fuzzy controller for the BTP is designed to produce the control quantity of the strand velocity. Finally, based on a semiphysical simulation system and the raw data collected from an iron and steel plant, an experiment is carried out to demonstrate the effectiveness of the proposed control strategy.

The accuracy and integrity of the actual production data influence the reliability and stability of sintering process in steel industry. However, the actual production data may encounter various outliers due to noise, sensor failure, and operator negligence existing in this process. To tackle this issue, this article develops an original framework for the detection and correction of abnormal production data in the sintering process. First, an improved kernel-based Fuzzy C-Means algorithm is developed to effectively divide normal production data under multiple operating conditions. Then, different one-class support vector machine (SVM) classifiers are constructed for different operating conditions. According to which operating condition the actual production data belongs to, the one-class SVM under this operating condition is called to accurately detect abnormal production data. Finally, the most similar normal historical data in the operating condition is obtained to correct the abnormal data by using k nearest neighbor algorithm based on the Mahalanobis distance. Simulation results involving actual production data illustrate the effectiveness of the proposed method. By taking two existing models of the sintering process as examples, their prediction performance becomes improved after detecting and correcting the abnormal production data, so that the proposed framework has important engineering application impact.