Search

Blast furnace ironmaking process monitoring is an important and challenging task. Due to the influence of hot blast stove switching and large fluctuations in the quality of raw materials, the measurements of ironmaking processes show obvious non‐stationary characteristics, and in addition, the observed data are also characterized by time‐series dynamic and non‐Gaussian characteristics. In this paper, a dynamic stationary subspace analysis method based on the Gaussian mixture model (DSSA–GMM) is proposed to address the difficulties in blast furnace ironmaking process monitoring. The time‐series dynamic relationship of the data is conducted by introducing a sliding time window. The Gaussian mixture model (GMM) is used to deal with the non‐Gaussian characteristics of the data, and the parameters of the GMMs are estimated using the expectation–maximization algorithm. The stationary projection matrix is obtained by optimizing the Kullback–Leibler (K–L) divergence between GMMs of different periods to realize the stationary subspace separation. Finally, the convex hull of the stationary subspace is established for fault detection, thus realizing the monitoring for non‐stationary and non‐Gaussian dynamic processes. The effectiveness of the DSSA–GMM method is verified by a numerical simulation and a dataset collected from an actual blast furnace ironmaking process. [ABSTRACT FROM AUTHOR]

Blast furnaces are the most crucial equipment in ironmaking processes. Stable operation of the blast furnace is a prerequisite for personnel safety and production efficiency. Therefore, early detection of abnormalities in blast furnaces is an important task for ironmaking processes. However, owing to the large fluctuations in the quality of raw materials, dynamic operating conditions, as well as the impact of the hot blast stoves switches, the measurements of blast furnace show severe nonstationary characteristics. All these factors make monitoring the blast furnace a challenging task. In this article, a nonstationary process monitoring method called consistent trend feature analysis (CTFA) is proposed, which can extract the trend-related features and discard perturbations in process data. The directions and amplitudes of the extracted trends are used for abnormality detection, and a local-learning-based method is proposed for determining a time-varying control limit. The detection performance of the proposed method is analyzed, with a sufficient condition and a necessary condition for the detectability given. The effectiveness of the proposed method is validated by the practical data collected from a large-scale blast furnace located in Liuzhou, China. [ABSTRACT FROM AUTHOR]

Remaining useful life (RUL) prediction is critical for health management of industrial equipment. It has been widely noted that degradation modeling is a core step for RUL prediction where the Brownian motion (BM)-based models attract much attention. However, the existing BM-based degradation models still have some impractical assumptions, where the increments of a BM are independent and stationary. To extend the application of the degradation models, a bifractional Brownian motion (biFBM)-based degradation model is developed in this article. The biFBM is a process with dependent and nonstationary increments, which includes the BM and fractional Brownian motion (FBM) as special cases. For the proposed degradation model, the estimation of parameters and degradation states as well as the prediction of RUL is further considered. To address the non-Markovian degradation processes, an improved particle filter is designed for degradation state estimation and RUL prediction. The proposed degradation model and RUL prediction method are validated by case studies of turbine engines and a blast furnace wall. [ABSTRACT FROM AUTHOR]

For some practical industrial systems or components, such as blast furnaces and Li-ion batteries, there are two important factors to model the degradation processes. One is the long-range dependence, which can reflect the non-Markovian nature of the degradation processes. The other factor is the existence of multiple modes, because the operating conditions and external environments inevitably change during the whole lifetime of these systems. In this paper, we first propose a fractional Brownian motion (FBM) based degradation model with long-range dependence and multiple modes, and then consider the prediction of remaining useful life. To identify the multiple modes in the degradation process, we propose a two-step method, including change-points detection and linear segments clustering. In each degradation mode, the degradation rate is assumed to be normally distributed. The means and variances of these distributions can be obtained by the maximum likelihood estimation. To describe the switching between different modes, the continuous-time Markov chain is applied, and its transition rate matrix can be estimated by the historical switching time. An approximation of the first passage time with a predefined threshold can be obtained by a weak convergence theorem and a time-space transformation. A numerical simulation and a practical case of a blast furnace wall are provided to demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

A prerequisite for the existing remaining useful life prediction methods based on stochastic processes is the assumption of independent increments. However, this is in sharp contrast to some practical systems including batteries and blast furnace walls, in which the degradation processes have the property of long-range dependence. Based on the fractional Brownian motion, we adopt a degradation process with long-range dependence to predict the remaining useful life of the above systems. Because the degradation process with long-range dependence is neither a Markovian process nor a semimartingale, the exact analytical first passage time is difficult to derive directly. To address this problem, a weak convergence theorem is first adopted to approximately transform a fractional Brownian motion-based degradation process into a Brownian motion-based one with a time-varying coefficient. Then, with a space-time transformation, the first passage time of the degradation process with long-range dependence can be obtained in a closed form. Unknown parameters in the degradation model can be identified using discrete dyadic wavelet transform and maximum likelihood estimation. Numerical simulations and a practical example of a blast furnace wall are given to verify the effectiveness of the proposed method. [ABSTRACT FROM PUBLISHER]

The blast furnace reactor has a large volume, and smelting occurs at high temperatures and high-pressure environments. The blast furnace reactor has a large volume, and smelting occurs at high temperatures and high-pressure environments. Considering the small samples of fault data and non-stationarity of the production process with a considerable time lag, it is difficult to diagnose and eliminate the faults timely. This paper proposes for the first time a data and knowledge collaborative-driven framework to identify fault and perform self-healing control. Following the adaptive two-scale sequence division method, the prior knowledge matrix refined K2 algorithm is used to construct Bayesian networks to model the control process. When a new abnormal condition arises, the most similar historical sequence will be found, and inference on the corresponding Bayesian network can lead to fault identification results and self-healing control action. The proposed framework has achieved satisfactory results in simulation experiments and real data experiments. In the simulation data, the Bayesian network (BN) modeling achieved an accuracy of 98.75%, and the time required for structure learning of the BN was reduced by over 75.69%. In real data, the accuracy of identifying blast furnace faults reached 97.1%, and the results of self-healing control reasoning surpassed those of traditional BN reasoning methods in terms of correctness, order, completeness, and necessity. • A self-healing control framework for blast furnace is proposed. • Introducing prior knowledge into industry can efficiently benefit process modeling. • Put forward an adaptive sequence division method for non-stationary data. • The method is suitable for non-stationary and high safety requirements situations. [ABSTRACT FROM AUTHOR]

Detecting early abnormalities in blast furnaces is important for the smooth operation of the iron-making process. In this paper, recursive transformed component statistical analysis (RTCSA)-based algorithms are developed to monitor the iron-making process with the task of early abnormality detection. The increments of variables instead of the absolute measurements are used for RTCSA, in order to decrease the effect of the time-varying nature of the process. Owing to the peak-like disturbances caused by the switching of hot blast stoves, an online identification algorithm is designed to locate the disturbance intervals. Then an index-switching scheme is used for monitoring the process. The effectiveness of the proposed method is verified using the real data of two blast furnaces. Compared with the conventional methods such as the two-stage principal component analysis, the increment-based RTCSA can effectively detect early abnormalities in the iron-making process. [ABSTRACT FROM AUTHOR]

Within the crucial domain of blast furnace ironmaking and sintering, the quality of sinter ore and molten iron holds supreme importance, with direct implications for downstream processes. However, the complexities of utilizing operational experience, understanding mechanisms, leveraging extensive data for precise modeling, and optimizing multiple objectives have persistently posed challenges for engineers. In this research, we propose an novel data/mechanism hybrid-driven modeling and global sequential optimization framework, with three core contributions: (1) Synthesizing field operation insights and mechanistic principles to construct models for molten iron production and energy consumption in ironmaking. (2) Crafting the broad learning approximate-aided subspace identification method (BLASIM), encapsulating the system's dynamic and nonlinear characteristics. This method pioneers a parametric modeling strategy predicated on correlation error for dynamic nonlinear system identification, with its feasibility robustly underpinned by theoretical verification. (3) Streamlining the optimization process by applying expert knowledge to deconstruct a complex multi-objective optimization problem into manageable single-objective tasks. These tasks are addressed sequentially, reflecting operational chronology, and are adeptly resolved using gray wolf optimization algorithm with a sequence relaxant factor. To conclude, the proposed methods are thoroughly validated using real-world blast furnace smelting data, affirming the feasibility and efficiency of modeling accuracy and optimization performance. • Hybrid Modeling: Introduces a hybrid approach combining data and mechanisms. • BLASIM: Crafts a data-centric model, enabling nonlinear state space identification. • Optimization Strategy: Features a global sequential optimization using gray wolf algorithm. • Real-World Validation: Validates methods using real blast furnace data, proving efficacy. [ABSTRACT FROM AUTHOR]