Journal of Process Control最新文献

{"title":"Causal-geometry joint dictionary embedding learning for distributed monitoring and root cause analysis","authors":"Xue Xu ,&nbsp;Chaomin Luo ,&nbsp;Yuanjian Fu","doi":"10.1016/j.jprocont.2025.103566","DOIUrl":"10.1016/j.jprocont.2025.103566","url":null,"abstract":"<div><div>Interactions across process variables are complicated in large-scale industrial processes characterized with multiple operating units, posing significant challenges for fault detection and root cause analysis. In this work, a distributed modeling approach termed causal-geometry joint dictionary embedding learning (CGDE) is proposed to monitor large-scale industrial processes and identify the root cause. An information decomposition based block division algorithm is proposed to divide the entire process into blocks that account for unique, redundant, and synergistic information among variables. Meanwhile, a geometry similarity matrix derived by the minimum spanning tree is constructed to exploit the underlying structure of data. Furthermore, a causal consistency matrix is developed to characterize the causality among variables such that the intrinsic and stable information of industrial processes can be effectively captured. The CGDE approach provides an in-depth and faithful process analysis with consideration of causalities and geometry similarity of data, enhancing the distributed monitoring and root cause analysis performance. The effectiveness of CGDE is illustrated through a simulated platform and a real fluid catalytic cracking application.</div></div>","PeriodicalId":50079,"journal":{"name":"Journal of Process Control","volume":"155 ","pages":"Article 103566"},"PeriodicalIF":3.9,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145268431","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Sensor fault characteristics and fault detection in wastewater treatment plants: Current status and trend analysis","authors":"Shanshan Chen ,&nbsp;Xiaodong Wang ,&nbsp;Xuejun Bi ,&nbsp;Zakhar Maletskyi","doi":"10.1016/j.jprocont.2025.103574","DOIUrl":"10.1016/j.jprocont.2025.103574","url":null,"abstract":"<div><div>Sensor faults in wastewater treatment plants (WWTPs) significantly impact the data quality of online monitoring and further affect process operation. The reliability of online sensor data remains the key barrier which obstacles the digitalization of the water sector. Advances in machine learning (ML) and artificial intelligence (AI) offer new opportunities to improve fault detection and diagnosis. Using CiteSpace, this review analyzes literature from 2008 to 2024, highlighting the increasing adoption of hybrid fault detection models that integrate statistical, model-based, and data-driven methods. It categorizes sensor faults, examines their impact on WWTP monitoring, and evaluates mathematical approaches used for fault detection. While AI-driven models enhance detection accuracy, challenges persist in real-time implementation and adaptability to dynamic WWTP conditions. The review further explores strategies for enhancing fault resilience, emphasizing hybrid models, soft sensors, and advanced sensor networks as effective solutions for maintaining system functionality and ensuring continuous monitoring.</div></div>","PeriodicalId":50079,"journal":{"name":"Journal of Process Control","volume":"155 ","pages":"Article 103574"},"PeriodicalIF":3.9,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145416725","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A multi-model fault-tolerant control method for concurrent faults in wastewater treatment processes based on semi-supervised learning and physical constraints","authors":"Huan Luo,&nbsp;Ying Tian","doi":"10.1016/j.jprocont.2025.103560","DOIUrl":"10.1016/j.jprocont.2025.103560","url":null,"abstract":"<div><div>Wastewater treatment processes (WWTP) is one of the most essential means to achieve water resource protection and sustainable utilization, with dissolved oxygen and nitrate serving as main factors limiting effluent quality through their direct involvement in carbon consumption, nitrification, and denitrification processes. Existing fault-tolerant control strategies primarily focus on single sensor anomalies, while practical operations frequently encounter concurrent faults across multiple measurement channels. Moreover, the scarcity of labeled operational data in industrial settings poses significant challenges for developing reliable fault-tolerant control systems. This paper presents a passive fault-tolerant control approach using an innovative semi-supervised deep learning framework to address simultaneous failures in critical dissolved oxygen and nitrate sensors. The proposed methodology features four key innovations: (1) A novel SAE-MNN architecture that integrates stacked autoencoders with multi-output neural networks for simultaneous multi-parameter prediction through hierarchical feature extraction. (2) A confidence-based pseudo-labeling semi-supervised co-training mechanism that effectively leverages limited labeled data and abundant unlabeled operational data under data scarcity conditions. (3) Physics-constrained learning that enforces biochemical principles and mass conservation laws to ensure physically plausible predictions. (4) A multi-sensor passive fault-tolerant control strategy that handles simultaneous failures across multiple critical measurement channels without hardware redundancy or controller reconfiguration. This integrated framework enables robust operation during concurrent sensor failures, where predicted values seamlessly replace multiple faulty sensor measurements while maintaining stable control performance. The effectiveness is validated using the Benchmark Simulation Model No. 1 (BSM1), demonstrating superior system performance during multi-sensor fault scenarios compared to conventional methods, thereby enhancing the reliability and resilience of wastewater treatment systems.</div></div>","PeriodicalId":50079,"journal":{"name":"Journal of Process Control","volume":"155 ","pages":"Article 103560"},"PeriodicalIF":3.9,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145221455","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Align knowledge with time-series: Cross-modal domain knowledge activation for LLM-enabled zero-shot fault diagnosis","authors":"Jiancheng Zhao ,&nbsp;Chunhui Zhao ,&nbsp;Jiaqi Yue","doi":"10.1016/j.jprocont.2025.103534","DOIUrl":"10.1016/j.jprocont.2025.103534","url":null,"abstract":"<div><div>The existing zero-shot fault diagnosis methods identify unseen categories that have no training samples by predicting fault attributes from samples. Although these methods alleviated the data scarcity issue, they raised a new challenge, i.e., professional fault attributes annotation. We recognize that the essence of fault attributes lies in describing the connections and differences between categories. Therefore, we propose a novel LLM-enabled zero-shot fault diagnosis paradigm, and the large language models (LLMs) fine-tuned based on domain-specific knowledge can capture similar information to replace manual annotation. It unlocks the potential of LLMs to handle zero-shot tasks related to industrial time-series data in a cross-modal manner. It aims to address the burden of semantic knowledge annotation posed by the existing attribute-enabled paradigm. Moreover, the domain shift problem (DSP) arising from the shortage of training samples for unseen faults is also tackled by leveraging the cross-modal activation of relevant knowledge that has been learned from domain-specific documents. Firstly, to address the issue that LLMs pretrained on general knowledge are lacking in the knowledge of the industrial field, we design prompts for industrial faults and fine-tune the LLM with domain knowledge from diagnosis reports. Subsequently, considering that LLMs lack the ability to process time-series data, we design a cross-modal transformation module to align the time-series modality with the text modality. Moreover, we propose a knowledge distillation strategy to further align these two modalities, so the unseen fault text descriptions can serve as substitutes for the unavailable samples to address the DSP. We conduct experiments on a real thermal power plant, and the proposed method achieves an average improvement of 9.83% in terms of the diagnosis accuracy of unseen faults.</div></div>","PeriodicalId":50079,"journal":{"name":"Journal of Process Control","volume":"155 ","pages":"Article 103534"},"PeriodicalIF":3.9,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145021013","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Maximum extrem biogas yield prediction based tracking control for two-stage anaerobic digestion using CKF robust observer feedback","authors":"Hongxuan Li ,&nbsp;Haoping Wang ,&nbsp;Yang Tian ,&nbsp;Nicolai Christov","doi":"10.1016/j.jprocont.2025.103558","DOIUrl":"10.1016/j.jprocont.2025.103558","url":null,"abstract":"<div><div>Two-stage anaerobic digestion process, recognized as a promising microbiological technology, can effectively converts organic pollutants into renewable energy gases. However, practical implementation faces two fundamental challenges: the critical process states (for example, concentrations of anaerobic microorganisms) are not directly measurable through conventional sensors, and the gas production efficiency remains suboptimal under current operational paradigms. To address these challenges, this study proposed a robust observer-based biogas yield extremum prediction tracking controller (RO-EPTC). The proposed RO-EPTC controller integrates a cubature Kalman filter robust observer and an artificial neural network-based prediction tracking controller. The RO-EPTC enables dynamic extremum prediction of biogas yield while ensuring real-time convergence of actual gas production to the identified optimal trajectory. Additionally, the proposed scheme provides accurate estimation of unmeasurable system states. Finally, through simulation comparison experiments, the effects of proposed RO-EPTC method were verified.</div></div>","PeriodicalId":50079,"journal":{"name":"Journal of Process Control","volume":"155 ","pages":"Article 103558"},"PeriodicalIF":3.9,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145097510","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Spatiotemporal integrated control for ballast water heat treatment via the kernel learning and model predictive path integral","authors":"Guoqing Zhang ,&nbsp;Yang Xu ,&nbsp;Jiqiang Li ,&nbsp;Zehua Jia ,&nbsp;Weidong Zhang","doi":"10.1016/j.jprocont.2025.103564","DOIUrl":"10.1016/j.jprocont.2025.103564","url":null,"abstract":"<div><div>In this article, a spatiotemporal integrated control scheme for ballast water heat treatment is proposed that utilizes an improved nonlinear predictive control algorithm relying on a kernel-learning-based model to lower the concentration of microorganisms by manipulating the temperature of heated water indirectly. Firstly, multiple heat exchangers treating process is simplified into a plug flow reactor model with the properties of distributed parameter systems (DPSs). Based on the simplified model, the kernel-learning-based model is derived by using kernel principal component analysis (KPCA) and kernel extreme learning machine (KELM) for modeling the spatiotemporal temperature data. Further, the hyperparameters of the KELM involved therein are determined by a numerical optimization approach. The superiority of this design is to accurately explore the nonlinear dynamics and uncertainties of the actual system. Associated with the modeling method, the nonlinear predictive control strategy is designed to control and maintain the heating temperature. The remarkable trait is that a model predictive path integral (MPPI) is introduced to avoid the problem of “sinking into the local optimal solution”, which often emerges searching for the optimal control sequence. Finally, the stability analysis and numerical experiments support the effectiveness of the proposed scheme.</div></div>","PeriodicalId":50079,"journal":{"name":"Journal of Process Control","volume":"155 ","pages":"Article 103564"},"PeriodicalIF":3.9,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145324808","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Boiler operation predictions by integrating thermo-fluid principles within an artificial neural network framework","authors":"C. Bisset ,&nbsp;R. Coetzer ,&nbsp;PVZ. Venter","doi":"10.1016/j.jprocont.2025.103568","DOIUrl":"10.1016/j.jprocont.2025.103568","url":null,"abstract":"<div><div>Optimising boiler operations is challenging due to fluctuating conditions in complex thermo-fluid systems. This study introduces a novel approach to improve efficiency in coal-fired boilers by developing and validating an artificial neural network (ANN) model that provides both statistically accurate and scientifically feasible predictions. Three multi-layer perceptron (MLP) feedforward ANN models were developed, with variable selection supported by principal component analysis (PCA) and hyperparameter optimisation performed using Latin hypercube sampling (LHS). The best ANN achieved test root mean square errors (RMSEs) of 2.11 t/h for steam flow, 2.11 t/h for blowdown, 4.98 °C for superheated steam temperature, 0.69 bar for steam pressure, and 0.86 % for efficiency. The mean absolute percentage error (MAPE) for efficiency remained below 1.25 %, with deviations constrained within ±4.25 %. Statistical and thermodynamic validations were applied, including bootstrap aggregation of prediction variance and mass and energy balance checks. Results showed that 96.76 % of samples achieved water mass balance deviations of less than 0.01 %. Furthermore, 100 % of predictions for efficiency and energy output fell within a 5 % absolute error range. The novelty of this work lies in integrating ANN predictions with thermo-fluid validation. Theoretically, it advances current literature by bridging the gap between statistical accuracy and physical feasibility. Practically, it provides a reliable framework for evaluating efficiency in operational settings and lays the foundation for a machine learning (ML)–aided decision-support framework (DSF) for energy efficiency optimisation in coal-fired boilers.</div></div>","PeriodicalId":50079,"journal":{"name":"Journal of Process Control","volume":"155 ","pages":"Article 103568"},"PeriodicalIF":3.9,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145324809","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhancing hierarchical learning of real-time optimization and model predictive control for operational performance","authors":"Rui Ren,&nbsp;Shaoyuan Li","doi":"10.1016/j.jprocont.2025.103559","DOIUrl":"10.1016/j.jprocont.2025.103559","url":null,"abstract":"<div><div>In process control, the integration of Real-Time Optimization (RTO) and Model Predictive Control (MPC) enables the system to achieve optimal control over both long-term and short-term horizons, thereby enhancing operational efficiency and economic performance. However, this integration still faces several challenges. In the two-layer structure, the upper layer RTO involves solving nonlinear programming problems with significant computational complexity, making it difficult to obtain feasible solutions in real-time within the limited optimization horizon. Simultaneously, the lower layer MPC must solve rolling optimization problems within a constrained time frame, placing higher demands on real-time performance. Additionally, uncertainties in the system affect both optimization and control performance. To address these issues, this paper proposes a noval hierarchical learning approach for RTO and MPC controller using reinforcement learning. This method learns the optimal strategies for RTO and MPC across different time scales, effectively mitigating the high computational costs associated with online computations. Through reward design and experience replay during the hierarchical learning process, efficient training of the upper and lower layer strategies is achieved. Offline training under various uncertainty scenarios, combined with online learning, effectively reduces performance degradation due to model uncertainties. The proposed approach demonstrates excellent performance in two representative chemical engineering case studies.</div></div>","PeriodicalId":50079,"journal":{"name":"Journal of Process Control","volume":"155 ","pages":"Article 103559"},"PeriodicalIF":3.9,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145221413","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Data-driven Koopman MPC using mixed stochastic–deterministic tubes","authors":"Zhengang Zhong,&nbsp;Ehecatl Antonio del Rio-Chanona,&nbsp;Panagiotis Petsagkourakis","doi":"10.1016/j.jprocont.2025.103533","DOIUrl":"10.1016/j.jprocont.2025.103533","url":null,"abstract":"<div><div>This paper presents a novel data-driven stochastic MPC design for discrete-time nonlinear systems with additive disturbances by leveraging the Koopman operator and a distributionally robust optimization (DRO) framework. By lifting the dynamical system into a linear space, we achieve a finite-dimensional approximation of the Koopman operator. We explicitly account for the modeling approximation and additive disturbance error by a mixed stochastic–deterministic tube for the lifted linear model. This ensures the regulation of the original nonlinear system while complying with the prespecified constraints. Stochastic and deterministic tubes are constructed using a DRO and a hyper-cube hull, respectively. We provide finite sample error bounds for both types of tubes. The effectiveness of the proposed approach is demonstrated through numerical simulations.</div></div>","PeriodicalId":50079,"journal":{"name":"Journal of Process Control","volume":"155 ","pages":"Article 103533"},"PeriodicalIF":3.9,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145362337","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-objective optimal control of biochemical processes based on reinforcement learning","authors":"Chongyang Liu ,&nbsp;Jinxu Cui ,&nbsp;Jianzhi Wu ,&nbsp;Zhaohua Gong","doi":"10.1016/j.jprocont.2025.103572","DOIUrl":"10.1016/j.jprocont.2025.103572","url":null,"abstract":"<div><div>Optimal control of biochemical processes remains an open research and industrial challenge due to intrinsic system nonlinearity, unsteady dynamics and stringent operation constraints. Although reinforcement learning has recently gained attention, its direct application in biochemical process control has been hindered by the presence of multiple conflicting control objectives. To address this, we formulate a multi-objective optimal control problem in biochemical processes with both control inputs and terminal time as decision variables and subject to path and terminal inequality constraints. For this problem, a time-scaling transformation and an exact penalty method are exploited to convert it into the one with fixed terminal time and simple box constraints. Furthermore, the problem is transformed to a set of single-objective problems by using the scalarization techniques of weighted sum and normalized norm constraint. Then, based on an improved proximal policy optimization algorithm with dynamic clipping threshold, we develop a reinforcement learning algorithm to solve the resulting problems. Finally, two case studies on glucose batch fermentation and lysine fed-batch fermentation show that the proposed reinforcement algorithm can achieve more uniform distribution of optimal solution sets and faster convergence.</div></div>","PeriodicalId":50079,"journal":{"name":"Journal of Process Control","volume":"155 ","pages":"Article 103572"},"PeriodicalIF":3.9,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145362339","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}