Journal of Process Control最新文献

{"title":"Safe control strategy for energy storage cluster assisted load frequency control based on reinforcement learning","authors":"Lei Xu ,&nbsp;Jinxing Lin ,&nbsp;Xiang Wu ,&nbsp;Rong Fu","doi":"10.1016/j.jprocont.2025.103537","DOIUrl":"10.1016/j.jprocont.2025.103537","url":null,"abstract":"<div><div>The large-scale integration of renewable energy into the power grid introduces strong stochastic disturbances, posing new challenges to the safety of load frequency control (LFC). To deal with this issue, a safety control strategy is proposed for lithium-ion energy storage cluster into LFC. First, to achieve efficient frequency control with the energy storage cluster, a command allocation strategy for energy storage cluster and a control strategy for units are proposed, with comprehensive consideration of the state of charge, state of health and the real-time grid frequency deviation. Next, both the maximum frequency deviation (MFD) and the rate of change of frequency (RoCoF) are picked as dynamic response performance indexes to ensure frequency safety. Then, a novel LFC controller based on Safety Enhanced Deep Deterministic Policy Gradient (SE-DDPG) reinforcement learning algorithm is designed. The safety model of SE-DDPG which integrated with safety prediction network and intrinsic curiosity module (ICM) can enhance the exploratory capability while improving the safety and reliability of the policy. Finally, the effectiveness of the proposed safe LFC strategy is validated by numerical simulation. Compare with conventional proportional integral control, the proposed strategy reduces the MFD and the root mean square frequency deviation by 41.38 % and 22.74 % in the random noise scene. In the step load scene, MFD and the max RoCoF are reduced by 46.88 % and 48.15 %.</div></div>","PeriodicalId":50079,"journal":{"name":"Journal of Process Control","volume":"154 ","pages":"Article 103537"},"PeriodicalIF":3.9,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144925456","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":"Non-linear control of a fuel gas blending benchmark problem with added consumer dynamics","authors":"M.D. Sibiya,&nbsp;A.J. Wiid,&nbsp;J.D. le Roux,&nbsp;I.K. Craig","doi":"10.1016/j.jprocont.2025.103527","DOIUrl":"10.1016/j.jprocont.2025.103527","url":null,"abstract":"<div><div>This paper contributes to existing literature on fuel gas control by providing a feasible control solution with improved economic performance for an existing fuel gas control benchmark problem. Improved economic performance is achieved by implementing a non-linear model predictive controller (NMPC) that uses state estimates provided by a moving horizon estimator (MHE) and extended Kalman filter (EKF) for the fuel gas composition and flame speed index (FSI) to provide continuous inputs for the controller. Furthermore, the original fuel gas benchmark model is expanded to include consumer dynamics affecting fuel gas demand due to changes in the fuel gas heating value, making the model more representative of real industrial plants. The behaviour of an NMPC that neglects consumer dynamics (NMPC<span><math><msub><mrow></mrow><mrow><mn>1</mn></mrow></msub></math></span>) was compared against an NMPC that includes consumer dynamics (NMPC<span><math><msub><mrow></mrow><mrow><mn>2</mn></mrow></msub></math></span>).</div><div>The aim of the benchmark problem is to reduce the time-weighted average cost of fuel gas for three 46-hour cases, accounting for purchase costs and penalties for fuel gas specification violations. An optimal cost for each case is determined assuming ideal conditions and perfect control. The benchmark controller is a conventional multi-loop feedforward/feedback system and has an average cost for the three cases which is 38.5% higher than the optimal cost. The NMPC<span><math><msub><mrow></mrow><mrow><mn>1</mn></mrow></msub></math></span> controller has an average cost which is 33.9% higher than the optimal cost and better than the benchmark controller.</div><div>A new benchmark scenario was developed which includes the consumer dynamics. For the new scenario, NMPC<span><math><msub><mrow></mrow><mrow><mn>1</mn></mrow></msub></math></span> could not find a feasible solution, resulting in oscillations and specification violations. The oscillations would result in site-wide instabilities for all equipment using fuel gas. NMPC<span><math><msub><mrow></mrow><mrow><mn>2</mn></mrow></msub></math></span> was able to keep the process stable during these scenarios and maintain all specifications. This shows the necessity to include consumer dynamics for effective fuel gas blending control.</div></div>","PeriodicalId":50079,"journal":{"name":"Journal of Process Control","volume":"154 ","pages":"Article 103527"},"PeriodicalIF":3.9,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144864509","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":"Drill bit failure detection for drilling processes based on global–local feature extraction and multi-stage incremental learning","authors":"Peng Zhang,&nbsp;Wenkai Hu,&nbsp;Yupeng Li,&nbsp;Weihua Cao","doi":"10.1016/j.jprocont.2025.103532","DOIUrl":"10.1016/j.jprocont.2025.103532","url":null,"abstract":"<div><div>In drilling processes, real-time detection of drill bit failure states is essential to mitigate operational risks, reduce downtime, and enhance drilling precision. However, drilling signals often exhibit both long-term degradation and local subtle changes. This coexistence poses great challenges to the accurate detection of drill bit failures. Moreover, models trained on historical data often exhibit significant performance degradation when deployed to new drilling depths. This is because the distributions of drilling process data diverge at these new depths due to lithological heterogeneity. To overcome such limitations, this paper proposes a new drill bit failure detection method for drilling processes by integrating Transformer-Convolutional Selective Fusion Network (TCSFN) with multi-stage incremental learning. The main contributions are twofold: 1) A feature extraction method based on TCSFN is proposed to capture global long-term trend features and local transient fluctuation features; 2) a multi-stage incremental learning strategy is designed for different stages of the drilling processes, and composite losses are devised for these stages separately. Case studies involving real-world data are utilized to demonstrate the effectiveness and superiority of the proposed method.</div></div>","PeriodicalId":50079,"journal":{"name":"Journal of Process Control","volume":"154 ","pages":"Article 103532"},"PeriodicalIF":3.9,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144917020","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 dynamic programming-inspired approach for Mixed Integer Optimal Control Problems with dwell time constraints","authors":"Ramin Abbasi-Esfeden ,&nbsp;Christoph Plate ,&nbsp;Sebastian Sager ,&nbsp;Jan Swevers","doi":"10.1016/j.jprocont.2025.103522","DOIUrl":"10.1016/j.jprocont.2025.103522","url":null,"abstract":"<div><div>This paper introduces a dynamic programming-inspired approach for solving the Combinatorial Integral Approximation (CIA) problem within the CIA decomposition approach for Mixed-Integer Optimal Control Problems (MIOCPs). Additionally, we incorporate general dwell time constraints into this framework. The proposed method is tested on four MIOCPs with a minimum dwell time constraint, and its performance is compared to that of the state-of-the-art general purpose solver GuRoBi (MILP) and to the tailored branch-and-bound (BnB) solver from the pycombina package. The results show that the proposed approach is more computationally efficient, and its flexible cost-to-go function formulation makes it suitable for handling cases where simple approximations of the relaxed solution are insufficient.</div></div>","PeriodicalId":50079,"journal":{"name":"Journal of Process Control","volume":"154 ","pages":"Article 103522"},"PeriodicalIF":3.9,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144810228","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":"Identification of feasible regions using R-functions","authors":"S. Kucherenko ,&nbsp;N. Shah ,&nbsp;O.V. Klymenko","doi":"10.1016/j.jprocont.2025.103539","DOIUrl":"10.1016/j.jprocont.2025.103539","url":null,"abstract":"<div><div>The primary objective of feasibility analysis is to identify and define the feasibility region, which represents the range of operational conditions (e.g., variations in process parameters) that ensure safe, reliable, and feasible process performance. This work introduces a novel feasibility analysis method that requires only that model constraints (e.g., defining product Critical Quality Attributes or process Key Performance Indicators) be explicitly provided or approximated by a closed-form function, such as a multivariate polynomial model. The method is based on V.L. Rvachev's R-functions, enabling an explicit analytical representation of the feasibility region without relying on complex optimization-based approaches. R-functions offer a framework for describing intricate geometric shapes and performing operations on them using implicit functions and inequality constraints. The theory of R-functions facilitates the identification of feasibility regions through algebraic manipulation, making it a more practical alternative to traditional optimization-based methods. The effectiveness of the proposed approach is demonstrated using a suite of well-known test cases from the literature.</div></div>","PeriodicalId":50079,"journal":{"name":"Journal of Process Control","volume":"154 ","pages":"Article 103539"},"PeriodicalIF":3.9,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144932536","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":"Dual hormone predictive control for a fully automated intraperitoneal artificial pancreas: Preclinical evaluation in pigs","authors":"Karim Davari Benam ,&nbsp;Hasti Khoshamadi ,&nbsp;Marte Kierulf Å m ,&nbsp;Sverre Chr. Christiansen ,&nbsp;Patrick Christian Bösch ,&nbsp;Dag Roar Hjelme ,&nbsp;Øyvind Stavdahl ,&nbsp;Sven Magnus Carlsen ,&nbsp;Sébastien Gros ,&nbsp;Anders Lyngvi Fougner","doi":"10.1016/j.jprocont.2025.103499","DOIUrl":"10.1016/j.jprocont.2025.103499","url":null,"abstract":"<div><div>Fully automated regulation of blood glucose levels (BGL) has been the ultimate goal in the treatment of type 1 diabetes (T1D). In this context, full automation refers to a system that operates without requiring any patient interaction, such as meal or exercise announcements or manual insulin adjustments. However, achieving BGL control without such inputs remains a significant challenge for artificial pancreas (AP) systems, primarily due to the unfavorable mismatch between the time constants of meal absorption and the slower absorption kinetics of subcutaneously administered insulin. In this paper, we propose and test a dual-hormone intraperitoneal (IP) artificial pancreas system — delivering both insulin and glucagon — to explore the challenges and feasibility of achieving fully automated glucose regulation. To this, a predictive control approach was developed and tested in animal experiments. Experiments were conducted in six anesthetized pigs for 12–24 h and in an awake (unanaesthetized) pig for five days. The proposed method achieved a time-in-range (TIR, 3.9–10 mmol/L) of 73.1–94.2%, exceeding the average TIR reported for commercially available hybrid closed-loop systems. For comparison, the Medtronic MiniMed 670G reports a TIR of 70%, the Tandem t:slim X2 with Control-IQ achieves 72%, the Omnipod 5 with Horizon reports 70%, and the Diabeloop G7 achieves 74% TIR. The findings demonstrate that the full automation of BGL control using dual-hormone AP with IP injections is feasible. The paper also discusses the challenges and complexities associated with implementing the dual-hormone IP artificial pancreas system from the ground up. These challenges include addressing BGL measurement, estimation, prediction, and surgical considerations.</div></div>","PeriodicalId":50079,"journal":{"name":"Journal of Process Control","volume":"154 ","pages":"Article 103499"},"PeriodicalIF":3.9,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144889317","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 new method for monitoring flotation performance using a performance-guided autoencoder with the self-attention mechanism","authors":"Hao Yan ,&nbsp;Haoyu Shang ,&nbsp;Guangyu Zhu ,&nbsp;Fuli Wang","doi":"10.1016/j.jprocont.2025.103530","DOIUrl":"10.1016/j.jprocont.2025.103530","url":null,"abstract":"<div><div>Performance indicators are the core for reflecting the quality of flotation products. Real-time monitoring of these indicators is of great significance to enterprises. It remains challenging to capture the multimodal dynamic characteristics of time series data of the flotation process and effectively apply them to the monitoring task. To address this issue, this paper proposes a flotation performance monitoring method based on a performance-guided autoencoder that merges the self-attention mechanism. Firstly, the autoencoder takes the long short-term memory and the one-dimensional convolutional layer as its internal structure to parallel extract the long-term and local features of the time series data. Then, the self-attention mechanism is utilized to dynamically allocate weights to the fused features. The performance-guided autoencoder is based on unsupervised and supervised learning. The prediction error is incorporated into the loss function of the autoencoder to enhance the extracted features, making them more relevant to the monitoring task. Finally, the features extracted by the autoencoder are sent to the predictor module for real-time monitoring of performance indicators. The MAE, RMSE, and R² of the proposed method on the zinc flotation test data are 0.2475, 0.3433, and 0.7643, respectively, outperforming other existing advanced techniques. The experimental results verify the effectiveness and superiority of this method.</div></div>","PeriodicalId":50079,"journal":{"name":"Journal of Process Control","volume":"154 ","pages":"Article 103530"},"PeriodicalIF":3.9,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144879537","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":"Fault diagnosis of air handling units based on an MCNN-Transformer ensemble learning","authors":"Yin Xia,&nbsp;Danhong Zhang,&nbsp;Chenyu Liu,&nbsp;Zhiqiang Cao,&nbsp;Yixin Su,&nbsp;Yuhang Chen","doi":"10.1016/j.jprocont.2025.103526","DOIUrl":"10.1016/j.jprocont.2025.103526","url":null,"abstract":"<div><div>The Air Handling Units (AHU) in Heating Ventilation and Air Conditioning (HVAC) systems regulates air temperature and humidity to ensure indoor air quality and thermal comfort. Fault diagnosis of AHU is critical for reducing energy consumption and maintaining system performance. However, data noise and missing values introduce considerable uncertainty into AHU fault diagnosis, while most existing methods do not utilize time-series models and thus neglect the extraction of temporal features and the modeling of long-range dependencies. This limitation hinders the effective capture of fault evolution and long-term correlations, making it difficult to meet dynamic real-time requirements under complex operating conditions. To address these challenges, this paper proposes an ensemble learning framework that integrates Dempster–Shafer (DS) theory with a Multi-Channel Convolutional Neural Network and Transformer (MCNN-Transformer) model, aiming to enhance generalization and improve diagnostic performance. The DS theory combines the strengths of Random Forest, Pearson Correlation, and Mutual Information, effectively mitigating uncertainty and noise in fault feature data by fusing multi-source information. The MCNN-Transformer integrates multi-scale convolutional layers with a self-attention mechanism, enabling effective extraction of features across multiple temporal scales and modeling of long-range dependencies. Experimental results show that the proposed MCNN-Transformer framework achieves high efficiency and strong generalization capability, reaching a fault diagnosis accuracy of 99.2%, a precision of 0.992, a recall of 0.992, and an F1 score of 0.991, significantly outperforming traditional models. Moreover, the improved stability of the model’s accuracy curve further demonstrates its robustness.</div></div>","PeriodicalId":50079,"journal":{"name":"Journal of Process Control","volume":"154 ","pages":"Article 103526"},"PeriodicalIF":3.9,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144852365","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":"Closed-loop control framework for optimal startup of cryogenic air separation units","authors":"Anthony W.K. Quarshie ,&nbsp;Jose Matias ,&nbsp;Christopher L.E. Swartz ,&nbsp;Yanan Cao ,&nbsp;Yajun Wang ,&nbsp;Jesus Flores-Cerrillo","doi":"10.1016/j.jprocont.2025.103525","DOIUrl":"10.1016/j.jprocont.2025.103525","url":null,"abstract":"<div><div>Current volatile electricity market conditions incentivize the adaptation of the operation, including the startup, of cryogenic air separation units (ASUs) which are large consumers of electricity. Improvement in ASU startups using earlier proposed open-loop control strategies may not be fully realized in the presence of uncertainties and disturbances. This paper assesses the potential benefit of using a proposed closed-loop control framework to address uncertainty and disturbances. A rolling-horizon economic nonlinear model predictive control (ENMPC) approach is considered, for which strategies are proposed to improve computation time. Online parameter estimation is performed using a computationally efficient method that is easy to implement. Through the case studies presented, it is shown that the proposed framework outperforms the use of offline pre-computed optimal inputs in response to the disturbance and uncertainty considered.</div></div>","PeriodicalId":50079,"journal":{"name":"Journal of Process Control","volume":"154 ","pages":"Article 103525"},"PeriodicalIF":3.9,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144889318","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":"Obscured by terminology: Hidden parallels in direct methods for open-loop optimal control","authors":"Susanne Sass ,&nbsp;Alexander Mitsos","doi":"10.1016/j.jprocont.2025.103513","DOIUrl":"10.1016/j.jprocont.2025.103513","url":null,"abstract":"<div><div>Active research on optimal control methods comprises the developments of research groups from various fields, including control, mathematics, and process systems engineering. Although there is a consensus on the classification of the main solution methods, different terms are often used for the same method. For example, solving optimal control problems with control discretization and embedded state integration may be called sequential method or direct single shooting. Equally severely, the same term may be used ambiguously: Is control vector parameterization a synonym for control discretization or for direct single shooting? Both misleading distinctions and ambiguity complicate the scientific discourse. Thus, we delineate standard terms from open-loop optimal control in this tutorial. More precisely, we formulate and challenge hypotheses on the terminology of direct methods, i.e., solution methods using control discretization combined with state integration and/or state discretization. In particular, we point out the parallel of the embedded state integration with a numerical integration scheme and the reduced-space formulation of approaches using state discretization. Taking a step further towards integrated scheduling and control problems, we additionally investigate the similarities and differences between the discrete-time solution of optimal control problems and optimal quasi-steady operation. In this context, we also hint on the discrete-time representation in scheduling which refers to the handling of controls rather than the handling of process dynamics. Rather than concluding with the “correct” term to use, this tutorial concludes with recommendations on how to avoid misunderstandings in the versatile research community.</div></div>","PeriodicalId":50079,"journal":{"name":"Journal of Process Control","volume":"154 ","pages":"Article 103513"},"PeriodicalIF":3.9,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144893776","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}