The process industry is a continuous manufacturing system that comprises intricate physical and chemical reactions. Given the increasing constraints on resources and energy, it is urgent to optimize process indicators by maintaining an efficient reaction atmosphere. Reinforcement learning (RL), using trial and error to learn control strategies, has become a topic of interest in the control community. However, practical implementation reveals that the mapping between observed state variables and the reaction atmosphere is subject to uncertain disturbances, which seriously affect the reliability of process indicator control. To address these issues, a robust RL (RRL) control method based on knowledge-constrained adversarial perturbation is proposed. It applies the adversary to perturb the observed state to characterize the uncertain disturbance. First, the insight of composite modeling for the process industry is presented to factorize the inherent and external uncertainties. Based on this insight, a reaction atmosphere indicator surrogate model is built to quantify the inherent uncertainty. Second, by leveraging the variation boundary information of the surrogate model, a dynamic state perturbation set and its update policy are proposed to ensure the rationality of the state perturbation. Last, an external uncertain time series generation method with continuity constraints is proposed to incorporate reasonable external uncertainty in the training process. Case validation in zinc electrowinning demonstrates that the proposed method effectively enhances control performance in uncertain scenarios.

Precise control of roasting temperature is paramount for optimizing production efficiency in the zinc smelting process. However, existing research mainly focuses on average temperature control, and there is little research on temperature distribution control. To achieve this, a roasting temperature distribution model is first established based on the principles of heat transfer. Second, accounting for modeling errors and environmental disturbances, a discrete sliding mode control with disturbance compensation is proposed. Besides, continuous reaching law is implemented to address issues related to chattering, so as to ensure stable roasting temperature. Finally, the quasi-sliding-mode domain of the proposed method is obtained by boundary analysis. The simulation results of roasting temperature distribution control substantiate the efficacy of the proposed approach.Note to Practitioners—Roasting temperature is the most critical temperature that directly determines product quality and stable production during the roasting process. Currently popular schemes all use average temperature as the control target. However, the average temperature does not represent the actual temperature inside the roaster. This paper aims to achieve the temperature distribution of the roaster, thereby ultimately improving product quality and ensuring safe production. This paper proposes a roasting temperature control scheme based on discrete sliding mode control. During the implementation of this method, the current temperature error distribution is used as input to adjust the zinc concentrate feeding rate in real time. Experimental simulations verified the feasibility of this method, but it has not yet been applied in actual production.

The multiple-reactor cascade operation is a distinctive characteristic in the process industry. However, it is difficult to establish an accurate and global model for multi-reactor cascading processes. Moreover, the intricate and dynamic operating state of the reactor, coupled with rear reactors, poses significant challenges to the fine control of the entire process. Therefore, this paper proposes a distributed intelligent control method based on state self-learning. Initially, the time-varying dynamic model of each reactor unit is established by learning the parameters of the regression model at each state point, achieving a nonlinear description of the reactor under complex conditions. Subsequently, leveraging the dynamic model and the material conservation principle between reactors, multi-step collaborative prediction is conducted along the reactor cascade direction. Thirdly, distributed model predictive control based on error self-correction is adopted to realize distributed intelligent control of the cascade reactor. This method is verified in a zinc smelting leaching process. The results indicate its superiority over common methods, offering higher prediction accuracy for the cascade process and enabling more effective control of individual reactors through distributed intelligence, which provides a novel and promising control paradigm for the cascade process.Note to Practitioners—The future heralds an era of the Internet of Everything, and this transformation extends to the production processes of the process industry. Presently, decentralized control methods are prevalent in the process industry. However, these methods lack communication between controllers and autonomy in learning. While decentralized control methods can effectively regulate most industrial processes, they struggle to achieve optimal control in scenarios with cascading reactors, where the reactors are interdependent. Distributed control methods offer promise to address these limitations. Regrettably, research and application of distributed control in process industry engineering remain limited, lacking suitable methods for controller autonomy and information exchange among different controllers. Consequently, this paper presents a novel control approach for stable and efficient regulation of multi-reactor cascades in the process industry, offering promising avenues for widespread application.

Complex industrial processes exhibit typical nonstationarity due to frequently fluctuating material flows and complex control loops. This poses three challenges for trustworthy process monitoring, including data drift, coordination of old and new knowledge, and interpretability. In this study, the adaptive and interpretable process monitoring problem is formulated as an online updating strategy and the spatial topology structure representation learning process monitoring problem. An incremental variational graph attention autoencoder with probabilistic inference framework is proposed, which aims to effectively learn continuously from dynamically changing industrial data to make interpretable monitoring results. First, an incremental learning strategy based on the Bayesian regularized self-organizing map is presented, which can distinguish between real faults and time-varying changes. Once normal samples are encountered, the itself and downstream model are elegantly updated with a dynamic down-sampling replay strategy without leading to catastrophic forgetting. Subsequently, a variational graph attention autoencoder with probabilistic inference is proposed, which endows interpretable spatial structural relationships through priors and effectively captures the variability of spatial latent representations suitable for nonstationary processes. Then, an incremental variational Bayesian inference is introduced to calculate the adaptive thresholds to adapt the system. In addition, an anomaly-aware graph attention localization mechanism is provided to localize fault root causes and propagation paths. Finally, the effectiveness of the proposed method is validated through two industrial applications. The results demonstrate that the proposed method can significantly enhance the performance of process monitoring, especially for reducing the false alarm rate (FAR) in process monitoring schemes. Moreover, it offers interpretable causal relationships among faults.

Granger causality has emerged as a valuable tool in comprehending industrial processes and facilitating data modeling by unveiling the inherent relationships within the data. However, the characteristic of causality discovery tasks results in a lack of validation sets, making hyperparameter tuning reliant solely on intuition. Existing nonlinear Granger causality discovery methods suffer from insufficient accuracy and robustness due to the significant impact of hyperparameters. Hence, this article proposes an attribution-aided nonlinear Granger causality discovery method (Attri-NGC) for accurate and robust inference of Granger causality. Attri-NGC comprises two stages. In the first stage, a novel strategy is proposed to assess whether the variability in contributions, derived from deep learning-based attribution, can significantly reflect causality. This assessment transfers the robust advantage of attribution to causal discovery. In the second stage, a sparsity-inducing penalty targeted at ambiguous causality is defined to fine-tune the deep networks used for attribution. Our method transforms the paradigm of Granger causality discovery from a challenging deep networks training problem to a fine-tuning problem, leading to a substantial enhancement in the accuracy and robustness of causality discovery. The effectiveness of the proposed method is comprehensively validated on four public datasets and two industrial process datasets. The superior performance of Attri-NGC in supporting industrial process modeling effectively promotes the integration of Granger causality into practical applications for more interpretable process modeling.

The process industry is a key manufacturing process that consumes a vast amount of energy consumption. On the premise of ensuring process stability, controlling process variables to operate the process close to the optimal working condition plays a critical role in reducing energy consumption. Reinforcement learning (RL), using trial and error to learn control strategies, has received much attention. However, the substantial fluctuations of process variables and the switching delay gap of the process industry result in a high-dimension state-action space, making it difficult to learn control strategies efficiently, and there is no guarantee of control stability. To get around these issues, first, a generic knowledge-extracted method for process industry RL control is proposed. It does not require laborious expert knowledge acquisition processes. Second, to improve learning efficiency, the implicit knowledge is extracted using decision trees from operation trajectory data and embedded into agent controllers. Third, an explicit knowledge-oriented reward constructing method is designed to guarantee control stability. A case of the zinc electrowinning process is provided to validate its superiority. The result shows that it can reduce power consumption while stabilizing process variables within the spec limits, without a laborious expert knowledge acquisition process.

Abstract Online and accurate estimation of key performance indicators (KPI) is the foundation for operational optimization of a chemical process. However, a chemical process usually consists of multiple reactors, and the factors influencing KPI are spatially distributed in the long process flow. In addition, due to the distinct time lags between KPI and each reactor, temporal relationships among KPI and its influence factors are a mixture of short‐term and long‐term relationships. In this regard, a deep distributed KPI estimator with a self‐attention mechanism is proposed in this paper. First, considering the process topology, a cascaded long short‐term memory network is developed to simulate the process topology and capture the short‐term effects. Then, to extract the long‐term dependencies, a de‐noise self‐attention layer is employed to model interactions of all the influence factors explicitly and dynamically. Lastly, the proposed method is compared with typical state‐of‐the‐art methods using real industrial data. The comparison results illustrate the performance and effectiveness of the proposed KPI estimation method.

In this study, we propose a dynamics-learning multirate estimation approach to perceive the quality-related indices (QRIs) of the feeding solution of a unit process. A quality-related index for estimation is an intermediate technical indicator between a unit process and a proceeding unit process; hence, the estimation problem is formulated as a two-stage estimation problem utilizing the production data of both unit processes. Dynamics-learning bidirectional long short-term memory (BiLSTM) with different inputs for the forward and backward layers is proposed to manage the input data from the different unit processes. In the dynamics-learning BiLSTM, a cycle control gate is added in the memory cell to learn the dynamics of the QRIs, thereby enabling a high-rate estimation under multirate conditions. A Bayesian estimation model is then combined with the dynamics-learning BiLSTM model to manage the process delay. Ablation and comparative experiments are conducted to evaluate the feasibility and effectiveness of the proposed estimation approach. The experimental results illustrate the performance and high-rate estimation ability of the proposed approach.

The roasting temperature is critical for enhancing product quality, reducing air pollution, and ensuring the long term operation of the zinc roasting process. However, optimizing the roasting temperature is challenging due to complex reaction mechanisms, feed composition fluctuations, and the coupling relationship with downstream processes. In this paper, a two level decision-making system for co-optimization of the roasting temperature is proposed. In the first level, a fuzzy synthetic evaluation model with variable-weight degradation degree is established to accurately evaluate the operating performance of the zinc roasting process. The evaluation results are used to design the basic setting rules that provide the basic temperature setting values. In the second level, a concept of temperature adjustable margin is introduced via sensitivity analysis of the pro cess model to evaluate the optimality of two roasters in the zinc roasting process. Based on the temperature-adjustable margin, the collaborative setting rules are designed to reasonably allocate the basic setting value to the two zinc roasters for optimizing the operating performance of the zinc roasting process. Finally, an industrial case study is presented to demonstrate the effectiveness of the proposed two-level decision-making system.

Process industry indicator describes the production status and is crucial to the stable process operation. Its low sampling frequency makes it difficult to meet the indicator perception needs for real-time process control. Indicator estimation is a promising alternative to improve its obtaining frequency. However, the low sampling frequency of indicators leads to observation scarcity, discouraging shortening the estimation period. Moreover, fluctuations in working conditions (WCs) result in difficulty in reliable estimation. Therefore, a variable-period estimation method is proposed to change the estimation period reliably in the absence of observations. First, the WCs are identified by extracting semantic information from logs. Second, the network group is proposed, which achieves variable-period estimation by adjusting the number of subnetworks. Moreover, two mechanism constraints and a continuous accumulation mapping are proposed to ensure the estimation credibility. A case study of the zinc electrowinning process is provided to validate the method.

Modern industrial processes generate many interassociated variables, which are more likely to implicit associations knowledge for describing irregular changes at different times to accurately describe behavior changes. Motivated by this issue, a novel knowledge-data-based synchronization states analysis method is proposed in this article for process monitoring. Its advantage mainly refers to integrating physical–chemical mechanism knowledge to handle the representation of associated relationships between numerous monitor variables. Furthermore, this method utilizes the trend distributions of variable changes to observe the differences between operation states and their parents online, which can maintain the simple, practical, and efficient advantage of data-driven process monitoring. Specifically, global process monitoring can be achieved by the synchronization status exceeding its corresponding threshold ($\chi ^{2}$distribution). At the same time, the local cause of backtracking can also be identified by whether the weighting of eigenvector components of each variable exceeds their corresponding thresholds ($\chi ^{2}$distribution). This novel proposed process monitoring method is applied to one practical hydrometallurgical zinc purification process consisting of copper and cobalt removal processes. The application's comparable performance shows the applicability and effectiveness of this proposed method.

Industrial system monitoring includes fault diagnosis and anomaly detection, which have received extensive attention, since they can recognize the fault types and detect unknown anomalies. However, a separate fault diagnosis method or anomaly detection method cannot identify unknown faults and distinguish between different fault types simultaneously; thus, it is difficult to meet the increasing demand for safety and reliability of industrial systems. Besides, the actual system often operates in varying working conditions and is disturbed by the noise, which results in the intraclass variance of the raw data and degrades the performance of industrial system monitoring. To solve these problems, a metric learning-based fault diagnosis and anomaly detection method is proposed. Fault diagnosis and anomaly detection are adaptively fused in the proposed end-to-end model, where anomaly detection can prevent the model from misjudging the unknown anomaly as the known type, while fault diagnosis can identify the specific type of system fault. In addition, a novel multicenter loss is introduced to restrain the intraclass variance. Compared with manual feature extraction that can only extract suboptimal features, it can learn discriminant features automatically for both fault diagnosis and anomaly detection tasks. Experiments on three-phase flow (TPF) facility and Case Western Reserve University (CWRU) bearing have demonstrated that the proposed method can avoid the interference of intraclass variances and learn features that are effective for identifying tasks. Moreover, it achieves the best performance in both fault diagnosis and anomaly detection.

The electrowinning process is a critical operation in nonferrous hydrometallurgy and consumes large quantities of power consumption. Current efficiency is an important process index related to power consumption, and it is vital to operate the electrolyte temperature close to the optimum point to ensure high current efficiency. However, the optimal control of electrolyte temperature faces the following challenges. First, the temporal causal relationship between process variables and current efficiency makes it difficult to estimate the current efficiency accurately and set the optimal electrolyte temperature. Second, the substantial fluctuation of influencing variables of electrolyte temperature leads to difficulty in maintaining the electrolyte temperature close to the optimum point. Third, due to the complex mechanism, building a dynamic electrowinning process model is intractable. Hence, it is a problem of index optimal control in the multivariable fluctuation scenario without process modeling. To get around this issue, an integrated optimal control method based on temporal causal network and reinforcement learning (RL) is proposed. First, the working conditions are divided and the temporal causal network is used to estimate current efficiency accurately to solve the optimal electrolyte temperature under multiple working conditions. Then, an RL controller is established under each working condition, and the optimal electrolyte temperature is placed into the controller's reward function to assist in control strategy learning. An experiment case study of the zinc electrowinning process is provided to verify the effectiveness of the proposed method and to show that it can stabilize the electrolyte temperature within the optimal range without modeling.

Modern industry processes are typically composed of multiple operating units with reaction interaction and energy-mass coupling, which result in a mixed time-varying and spatial-temporal coupling of process variables. It is challenging to develop a comprehensive and precise fault detection model for the multiple interconnected units by simple superposition of the individual unit models. In this study, the fault detection problem is formulated as a spatial-temporal fault detection problem utilizing process data of multiple interconnected unit processes. A spatial-temporal variational graph attention autoencoder (STVGATE) using interactive information is proposed for fault detection, which aims to effectively capture the spatial and temporal features of the interconnected unit processes. First, slow feature analysis (SFA) is implemented to extract temporal information that reveals the dynamic relevance of the process data. Then, an integration method of metric learning and prior knowledge is proposed to construct coupled spatial relationships based on temporal information. In addition, a variational graph attention autoencoder (VGATE) is suggested to extract temporal and spatial information for fault detection, which incorporates the dominances of variational inference and graph attention mechanisms. The proposed method can automatically extract and deeply mine spatial-temporal interactive feature information to boost detection performance. Finally, three industrial process experiments are performed to verify the feasibility and effectiveness of the proposed method. The results demonstrate that the proposed method dramatically increases the fault detection rate (FDR) and reduces the false alarm rate (FAR).

In real industrial processes, factors, such as the change in manufacturing strategy and production technology lead to the creation of multimode industrial processes and the continuous emergence of new modes. Although the industrial SCADA system has accumulated a large amount of historical data, which can be used for modeling and monitoring multimode processes to a certain extent, it is difficult for the model learned from historical data to adapt to emerging modes, resulting in the model mismatch. On the other hand, updating the model with data from new modes allows the model to continuously match the new modes, but it may cause the model to lose the ability to represent the historical modes, resulting in "catastrophic forgetting." To address these problems, this article proposed a jointly mode-matching and similarity-preserving dictionary learning (JMSDL) method, which updated the model by learning the data of new modes, so that the model can adaptively match the newly emerged modes. At the same time, a similarity metric was put forward to guarantee the representation ability of the proposed method for historical data. A numerical simulation experiment, the CSTH process experiment, and an industrial roasting process experiment indicated that the proposed JMSDL method can match new modes while maintaining its performance on the historical modes accurately. In addition, the proposed method significantly outperforms the state-of-the-art methods in terms of fault detection and false alarm rate.

The detection of small infrared targets with a low signal-to-noise ratios and contrasts in noisy and cluttered backgrounds is challenging and therefore a domain of active research. Traditional methods result in a large number of false alarms and missed detections. In the case of convolutional neural network-based methods, it may not be possible to identify deep small targets, or the details of the target’s edge contours may not be appropriately considered. Therefore, this paper proposes MSAFFNet to perform infrared small target detection based on an encoder-decoder framework. In the encoder stage, small target features are extracted using a resnet-20 backbone network, and the global contextual features of small targets are extracted using an atrous spatial pyramid pooling module. In the decoding stage, a dual-attention module is used to selectively enhance the spatial details of the target at the shallow level and representative features of the semantic information at the deep level. Multi-scale feature maps are then concatenated to achieve superior feature fusion. Additionally, multi-scale labels are constructed to focus on the details of the target contour and internal features based on edge information and an internal feature aggregation module. Experiments conducted on the NUAA-SIRST, NUDT-SIRST and XDU-SIRST datasets revealed that the proposed approach outperforms the representative methods and achieves an improved detection performance.

Infrared (IR) ship tracking is becoming increasingly important in various applications. However, it remains a challenging task as the information that can be obtained from infrared images is limited. Aiming at enhancing IR ship tracking accuracy, we propose an innovative approach by presenting feature integration module (FIM) and backup matching module (BMM). FIM takes appearance feature, complete intersection over union (CIoU), and motion direction metrics into account. Regarding appearance feature extraction, an end-to-end characteristic learning strategy with a cross-guided multi-granularity fusion network is proposed to obtain more integral appearance features and enhance re-identification accuracy, which helps to distinguish individual IR ship targets better. Besides, a backup matching strategy is then used to match the unmatched tracks and detections after cascaded matching. Virtual trajectories are generated for the matched tracks to optimize parameters by parameter optimization module (POM). The accumulation of errors caused by the lack of observations in the Kalman filter is reduced. Thus, the position of IR ships can be estimated more accurately, and more robust IR ship tracking can be achieved. In addition, we present a sequential frame IR ship tracking dataset, providing the first public benchmark for testing IR ship tracking performance. Experimental results indicate that the MOTA, MOTP and IDs of the proposed method are 73.441, 80.826, and 32, respectively, outperforming other state-of-the-art methods. This demonstrates the superior robustness of the proposed method, particularly when the IR ships are occluded or the target texture information is lacking. Our dataset is available at https://github.com/echo-sky/SFIST.

The inability of conventional algorithms to detect infrared (IR) ship targets in complex scenes led to the development of detection methods based on convolutional neural networks (CNNs). In this study, we propose a CNN-based stacked residual coordinate attention network (SRCANet) for detecting IR ship targets. Three-directional stacked interaction modules and a full-scale skip connection feature fusion scheme are introduced. The proposed network maintains and integrates sufficient contextual information of IR ship targets and obtains clear target boundary information. A cascaded residual coordinate attention module (CRCAM) is designed as the basic node in the SRCANet. Additionally, a residual coordinate attention module (RCAM) is introduced, which combines a two-dimensional convolution layer with batch normalisation and rectified linear unit (CBR), a coordination attention module, and a residual connection. The RCAM enhances the input feature map and improves the representability of objects of interest. The CRCAM comprises several cascading RCAMs that deepen the feature extraction layers. Furthermore, because there is no publicly available IR ship target dataset for segmentation, pixel-level annotations are performed on a set of IR ship target images and released as a single-frame IR ship detection (SISD) dataset. Extensive experiments were conducted on the SISD dataset and the widely used single-frame IR small target dataset to demonstrate the superiority of the proposed method. The results indicate that the SRCANet outperforms the state-of-the-art models, and it is more robust when target texture information is lacking. The SISD dataset is available at https://github.com/echo-sky/SISD.

Population-based intelligent optimization algorithms are popular due to their global optimality. However, it will be time-consuming if they are directly applied to nonlinear dynamical systems. In this article, an efficient computational cost reduction strategy is presented for the population-based intelligent optimization algorithms when they are employed to search for the global optimum of nonlinear systems in dynamic equilibrium. Specifically, a novel constrained optimization problem is formulated according to the demand of dynamic equilibrium in industry, and the reason why population-based methods take much more computing time is analyzed from the perspective of solving nonlinear equations. Since the computational complexity of solving nonlinear equations is sensitive to their initial values, a sensitivity transfer condition is provided to explain the significance of candidate testing order. Finally, an optimal evaluation sequence, which minimizes the Euclidean distance of adjacent candidates, is designed for the population-based intelligent optimization algorithms. Simulations show the effectiveness.