In multistage press hardening processes, where a sheet material undergoes rapid austenitization, tempering, stretch-forming (SF), and die bending (DB), the resulting product properties are influenced by the thermo-mechanical history. This work aims at controlling the product properties of the formed blanks by making use of the model-based estimation and control of the spatial-temporal temperature distribution in the sheet. A data-driven dynamical model is constructed using dynamic mode decomposition (DMD) based on finite element (FE) simulation data. This model is further extended by means of parametric DMD to accommodate changes in process parameters like stroke rate, blank holder force, and austenitization temperature. The approach is validated and the model accuracy is improved through experimental analysis. The dynamics of the available temperature sensors are identified, whereupon a Kalman filter is developed based on the parametric DMD model to estimate the spatial-temporal temperature distribution. A time-varying, stage-dependent output matrix is employed to account for different numbers and locations of thermocouples in the three stages. Additionally, an optimal control strategy is implemented to achieve desired temperature trajectories, allowing targeted manipulation of the blank’s geometry and properties. Experimental validation of this system design and control strategy is carried out under real-time constraints.
{"title":"Data-Based Estimation and Control of a Multistage Press Hardening Process","authors":"Malte Wrobel;Juri Martschin;Henry Baumann;A. Erman Tekkaya;Thomas Meurer","doi":"10.1109/TCST.2025.3589411","DOIUrl":"https://doi.org/10.1109/TCST.2025.3589411","url":null,"abstract":"In multistage press hardening processes, where a sheet material undergoes rapid austenitization, tempering, stretch-forming (SF), and die bending (DB), the resulting product properties are influenced by the thermo-mechanical history. This work aims at controlling the product properties of the formed blanks by making use of the model-based estimation and control of the spatial-temporal temperature distribution in the sheet. A data-driven dynamical model is constructed using dynamic mode decomposition (DMD) based on finite element (FE) simulation data. This model is further extended by means of parametric DMD to accommodate changes in process parameters like stroke rate, blank holder force, and austenitization temperature. The approach is validated and the model accuracy is improved through experimental analysis. The dynamics of the available temperature sensors are identified, whereupon a Kalman filter is developed based on the parametric DMD model to estimate the spatial-temporal temperature distribution. A time-varying, stage-dependent output matrix is employed to account for different numbers and locations of thermocouples in the three stages. Additionally, an optimal control strategy is implemented to achieve desired temperature trajectories, allowing targeted manipulation of the blank’s geometry and properties. Experimental validation of this system design and control strategy is carried out under real-time constraints.","PeriodicalId":13103,"journal":{"name":"IEEE Transactions on Control Systems Technology","volume":"33 6","pages":"2426-2438"},"PeriodicalIF":3.9,"publicationDate":"2025-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145341069","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}
Pub Date : 2025-08-27DOI: 10.1109/TCST.2025.3583100
Daniel Zamudio;Alessandro Falsone;Federico Bianchi;Maria Prandini
Aggregators of energy resources providing balancing services to the grid face the twofold challenge of assessing the overall amount of flexibility of the pool of energy resources, and mapping the power requests by the grid back to each single energy resource during the service window. In this article, we propose a framework that allows to jointly optimize the power flexibility limits and determines the disaggregation policy for managing the pool of energy resources, thus avoiding an additional computational step in the operational service phase. The baseline power exchange profiles of the energy resources can also be optimized to enhance the pool flexibility, and practical constraints such as time availability of the resources within the service window or network congestion constraints can be included. Notably, the resulting optimization problem is amenable for privacy-preserving and scalable distributed resolution schemes.
{"title":"Balancing Services Provision via Optimization and Management of the Flexibility Offered by a Pool of Energy Resources","authors":"Daniel Zamudio;Alessandro Falsone;Federico Bianchi;Maria Prandini","doi":"10.1109/TCST.2025.3583100","DOIUrl":"https://doi.org/10.1109/TCST.2025.3583100","url":null,"abstract":"Aggregators of energy resources providing balancing services to the grid face the twofold challenge of assessing the overall amount of flexibility of the pool of energy resources, and mapping the power requests by the grid back to each single energy resource during the service window. In this article, we propose a framework that allows to jointly optimize the power flexibility limits and determines the disaggregation policy for managing the pool of energy resources, thus avoiding an additional computational step in the operational service phase. The baseline power exchange profiles of the energy resources can also be optimized to enhance the pool flexibility, and practical constraints such as time availability of the resources within the service window or network congestion constraints can be included. Notably, the resulting optimization problem is amenable for privacy-preserving and scalable distributed resolution schemes.","PeriodicalId":13103,"journal":{"name":"IEEE Transactions on Control Systems Technology","volume":"33 6","pages":"2304-2319"},"PeriodicalIF":3.9,"publicationDate":"2025-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145339694","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}
Pub Date : 2025-08-15DOI: 10.1109/TCST.2025.3597431
Kaixuan Huo;Ran Chen;Yuzhe Li
Aero-engines directly affect the aircraft’s performance, safety, and reliability, thus playing a vital role in aviation. However, aero-engines operate in tough environments with high temperatures and pressures, which impose system constraints and uncertainties due to environmental variations, such as inlet wind speed, ambient temperature, barometric pressure, and inlet conditions leading to challenging control problems for aero-engines. To cope with these challenges, we develop a robust model-predictive control (RMPC) approach for aero-engine applications in this brief. Furthermore, to address the computational burden in traditional RMPC methods caused by the iterative nature of solving optimization problems, we propose a method called efficient tube MPC (ET-MPC), which ensures efficient computation and reduced computational time through the design of an event-triggered mechanism while guaranteeing adherence to tracking control constraints. The stability and feasibility of the proposed method are demonstrated. Finally, we conducted simulation experiments on a real world aero-engine model to validate the efficiency and safety of the ET-MPC algorithm.
{"title":"Efficient Tube Model Predictive Control for Safety-Aware Operation of Aero-Engine","authors":"Kaixuan Huo;Ran Chen;Yuzhe Li","doi":"10.1109/TCST.2025.3597431","DOIUrl":"https://doi.org/10.1109/TCST.2025.3597431","url":null,"abstract":"Aero-engines directly affect the aircraft’s performance, safety, and reliability, thus playing a vital role in aviation. However, aero-engines operate in tough environments with high temperatures and pressures, which impose system constraints and uncertainties due to environmental variations, such as inlet wind speed, ambient temperature, barometric pressure, and inlet conditions leading to challenging control problems for aero-engines. To cope with these challenges, we develop a robust model-predictive control (RMPC) approach for aero-engine applications in this brief. Furthermore, to address the computational burden in traditional RMPC methods caused by the iterative nature of solving optimization problems, we propose a method called efficient tube MPC (ET-MPC), which ensures efficient computation and reduced computational time through the design of an event-triggered mechanism while guaranteeing adherence to tracking control constraints. The stability and feasibility of the proposed method are demonstrated. Finally, we conducted simulation experiments on a real world aero-engine model to validate the efficiency and safety of the ET-MPC algorithm.","PeriodicalId":13103,"journal":{"name":"IEEE Transactions on Control Systems Technology","volume":"34 1","pages":"517-523"},"PeriodicalIF":3.9,"publicationDate":"2025-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145915585","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}
Pub Date : 2025-08-04DOI: 10.1109/TCST.2025.3593243
Tong Yang;Changda Fan;Yongchun Fang;Ning Sun
This article proposes a comprehensive adaptive learning controller to improve control accuracy and resist disturbance force/torques for pneumatic artificial muscles (PAMs), as well as handling various problems, e.g., time-varying parameters, unmodeled dynamics, and input delay. The modified neural networks compensate for nonlinear structures with time-varying parameters, where it is unnecessary to repeatedly calculate activation functions, reducing computational efforts. Specifically, a new integral term of input delay errors and closed-loop filter-based auxiliary signals is introduced into the designed update law and controller. When PAMs suffer from external disturbance force/torques during man–machine interaction, a modified force/torque observer is designed independently of measurement values and model knowledge (e.g., measured outputs and dynamic matrices in PAMs), to avoid sampling errors, noise, and so on. As far as we know, this is the first solution to handle time-varying parameters, resist input delay, and compensate for force/torque impacts together for PAMs, without any structure limits or a prior model information. It is proven that the tracking errors exponentially converge to zero; moreover, all closed-loop signals are uniformly ultimately bounded (UUB) when disturbance forces/torques are injected into the system. Some experimental verification is also conducted on a self-built platform.
{"title":"Adaptive Learning Control for Time-Varying Parameter Pneumatic Artificial Muscle Robots With Force/Torque Perturbation and Input Delays","authors":"Tong Yang;Changda Fan;Yongchun Fang;Ning Sun","doi":"10.1109/TCST.2025.3593243","DOIUrl":"https://doi.org/10.1109/TCST.2025.3593243","url":null,"abstract":"This article proposes a comprehensive adaptive learning controller to improve control accuracy and resist disturbance force/torques for pneumatic artificial muscles (PAMs), as well as handling various problems, e.g., time-varying parameters, unmodeled dynamics, and input delay. The modified neural networks compensate for nonlinear structures with time-varying parameters, where it is unnecessary to repeatedly calculate activation functions, reducing computational efforts. Specifically, a new integral term of input delay errors and closed-loop filter-based auxiliary signals is introduced into the designed update law and controller. When PAMs suffer from external disturbance force/torques during man–machine interaction, a modified force/torque observer is designed independently of measurement values and model knowledge (e.g., measured outputs and dynamic matrices in PAMs), to avoid sampling errors, noise, and so on. As far as we know, this is the first solution to handle time-varying parameters, resist input delay, and compensate for force/torque impacts together for PAMs, without any structure limits or a prior model information. It is proven that the tracking errors exponentially converge to zero; moreover, all closed-loop signals are uniformly ultimately bounded (UUB) when disturbance forces/torques are injected into the system. Some experimental verification is also conducted on a self-built platform.","PeriodicalId":13103,"journal":{"name":"IEEE Transactions on Control Systems Technology","volume":"33 6","pages":"2366-2377"},"PeriodicalIF":3.9,"publicationDate":"2025-08-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145339685","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}
Pub Date : 2025-08-04DOI: 10.1109/TCST.2025.3589462
Yuqiang Jin;Hu Sun;Wen-An Zhang
Collaborative localization (CL) in multiagent systems has become an increasingly prominent research area, particularly under an uncertain and partially connected dynamic communication environment. This article presents a CL framework based on a resilient sequential fusion approach that guarantees consistency in matrix Lie groups. The proposed method is inherited from the pipeline of the distribution Kalman filter, which utilizes invariant error defined on a manifold to establish the global state propagation and update process for estimating the pose of all agents in the predefined reference frame. Furthermore, the communication update process is treated separately by generalizing the covariance intersection (CI) fusion into the designed geometric group structure, enabling flexible updates while maintaining the consistency of estimates and ensuring the independence of the filter update process. Specifically, to address the potential issues in agents’ communication, a weighted fusion criterion with an analytical form is proposed, allowing communication fusion to be performed on the manifold with arbitrary information fusion order and structure. Extensive validation through simulations and real-world experiments demonstrates that the proposed method is resilient to varying communication conditions and achieves superior performance compared with state-of-the-art methods.
{"title":"Resilient Sequential Fusion on Lie Groups for Consistent Collaborative Localization","authors":"Yuqiang Jin;Hu Sun;Wen-An Zhang","doi":"10.1109/TCST.2025.3589462","DOIUrl":"https://doi.org/10.1109/TCST.2025.3589462","url":null,"abstract":"Collaborative localization (CL) in multiagent systems has become an increasingly prominent research area, particularly under an uncertain and partially connected dynamic communication environment. This article presents a CL framework based on a resilient sequential fusion approach that guarantees consistency in matrix Lie groups. The proposed method is inherited from the pipeline of the distribution Kalman filter, which utilizes invariant error defined on a manifold to establish the global state propagation and update process for estimating the pose of all agents in the predefined reference frame. Furthermore, the communication update process is treated separately by generalizing the covariance intersection (CI) fusion into the designed geometric group structure, enabling flexible updates while maintaining the consistency of estimates and ensuring the independence of the filter update process. Specifically, to address the potential issues in agents’ communication, a weighted fusion criterion with an analytical form is proposed, allowing communication fusion to be performed on the manifold with arbitrary information fusion order and structure. Extensive validation through simulations and real-world experiments demonstrates that the proposed method is resilient to varying communication conditions and achieves superior performance compared with state-of-the-art methods.","PeriodicalId":13103,"journal":{"name":"IEEE Transactions on Control Systems Technology","volume":"33 6","pages":"2320-2333"},"PeriodicalIF":3.9,"publicationDate":"2025-08-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145339710","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}
In this article, we propose a time-varying model predictive control (MPC)-based scheme to enhance the dynamic performance of dc–dc converters. The proposed approach employs MPC as a reference governor (RG), addressing industrial certification constraints that may limit modifications to the low-level controller. To accommodate the computational limitations of conventional control boards, we introduce a highly efficient real-time optimization algorithm for solving equality-constrained quadratic programming (QP) problems. The algorithm is based on a tailored QR factorization that outperforms well-known linear algebra libraries, and it is shown to be superior to condensing with state elimination. Furthermore, we implement an efficient recursive least-squares (RLS) method to provide a linear-time varying model for the adaptive MPC-based RG. No information regarding the topology of the converter nor the structure of the low-level controller is required for such adaptation, making the proposed method self-tuning and eliminating the need for prior identification steps. The proposed control scheme has been tested on various simulated and real dc–dc converters, demonstrating its computational and memory efficiency, as well as its versatility across different converter topologies.
{"title":"Adaptive Reference Governor for DC–DC Converters Based on Model Predictive Control","authors":"Gionata Cimini;Riccardo Felicetti;Francesco Ferracuti;Luca Cavanini;Andrea Monteriù","doi":"10.1109/TCST.2025.3587117","DOIUrl":"https://doi.org/10.1109/TCST.2025.3587117","url":null,"abstract":"In this article, we propose a time-varying model predictive control (MPC)-based scheme to enhance the dynamic performance of dc–dc converters. The proposed approach employs MPC as a reference governor (RG), addressing industrial certification constraints that may limit modifications to the low-level controller. To accommodate the computational limitations of conventional control boards, we introduce a highly efficient real-time optimization algorithm for solving equality-constrained quadratic programming (QP) problems. The algorithm is based on a tailored QR factorization that outperforms well-known linear algebra libraries, and it is shown to be superior to condensing with state elimination. Furthermore, we implement an efficient recursive least-squares (RLS) method to provide a linear-time varying model for the adaptive MPC-based RG. No information regarding the topology of the converter nor the structure of the low-level controller is required for such adaptation, making the proposed method self-tuning and eliminating the need for prior identification steps. The proposed control scheme has been tested on various simulated and real dc–dc converters, demonstrating its computational and memory efficiency, as well as its versatility across different converter topologies.","PeriodicalId":13103,"journal":{"name":"IEEE Transactions on Control Systems Technology","volume":"33 6","pages":"2350-2365"},"PeriodicalIF":3.9,"publicationDate":"2025-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145339700","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}
Pub Date : 2025-07-23DOI: 10.1109/TCST.2025.3587842
Hoang Anh Tran;Nikolai Lauvås;Tor Arne Johansen;Rudy R. Negenborn
This article focuses on the problem of collaborative collision avoidance (CCAS) for autonomous inland ships. Two solutions are provided to solve the problem in a distributed manner. We first present a distributed model predictive control (MPC) algorithm that allows ships to directly negotiate their intention to avoid collision in a synchronous communication framework. Moreover, we introduce a new approach to shape the ship’s behavior to follow the waterway traffic regulations. The conditional convergence toward a stationary solution of this algorithm is guaranteed by the theory of the alternating direction method of multipliers (ADMM). To overcome the problem of asynchronous communication between ships, we adopt a new asynchronous nonlinear ADMM (Async-NADMM) and present an asynchronous distributed MPC algorithm based on it. Several simulations and field experiments show that the proposed algorithms can guarantee a safe distance between ships in complex scenarios while following the traffic regulations. Furthermore, the asynchronous algorithm has an efficient computational time and satisfies the real-time computing requirements of ships in field experiments.
{"title":"Asynchronous Distributed Collision Avoidance With Intention Consensus for Inland Autonomous Ships","authors":"Hoang Anh Tran;Nikolai Lauvås;Tor Arne Johansen;Rudy R. Negenborn","doi":"10.1109/TCST.2025.3587842","DOIUrl":"https://doi.org/10.1109/TCST.2025.3587842","url":null,"abstract":"This article focuses on the problem of collaborative collision avoidance (CCAS) for autonomous inland ships. Two solutions are provided to solve the problem in a distributed manner. We first present a distributed model predictive control (MPC) algorithm that allows ships to directly negotiate their intention to avoid collision in a synchronous communication framework. Moreover, we introduce a new approach to shape the ship’s behavior to follow the waterway traffic regulations. The conditional convergence toward a stationary solution of this algorithm is guaranteed by the theory of the alternating direction method of multipliers (ADMM). To overcome the problem of asynchronous communication between ships, we adopt a new asynchronous nonlinear ADMM (Async-NADMM) and present an asynchronous distributed MPC algorithm based on it. Several simulations and field experiments show that the proposed algorithms can guarantee a safe distance between ships in complex scenarios while following the traffic regulations. Furthermore, the asynchronous algorithm has an efficient computational time and satisfies the real-time computing requirements of ships in field experiments.","PeriodicalId":13103,"journal":{"name":"IEEE Transactions on Control Systems Technology","volume":"33 6","pages":"2410-2425"},"PeriodicalIF":3.9,"publicationDate":"2025-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145339691","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}
Pub Date : 2025-07-23DOI: 10.1109/TCST.2025.3589409
Casian Iacob;Hany Abdulsamad;Simo Särkkä
Model predictive control (MPC) is a powerful framework for optimal control of dynamical systems. However, MPC solvers suffer from a high computational burden that restricts their application to systems with low sampling frequencies. This issue is further amplified in nonlinear and constrained systems that require nesting MPC solvers within iterative procedures. In this brief, we address these issues by developing parallel-in-time algorithms for constrained nonlinear optimization problems that take advantage of massively parallel hardware to achieve logarithmic computational time scaling over the planning horizon. We develop time-parallel second-order solvers based on interior point (IP) methods and the alternating direction method of multipliers (ADMM), leveraging fast convergence and lower computational cost per iteration. The parallelization is based on a reformulation of the subproblems in terms of associative operations that can be parallelized using the associative scan algorithm. We validate our approach on numerical examples of nonlinear and constrained dynamical systems.
{"title":"A Parallel-in-Time Newton’s Method for Nonlinear Model Predictive Control","authors":"Casian Iacob;Hany Abdulsamad;Simo Särkkä","doi":"10.1109/TCST.2025.3589409","DOIUrl":"https://doi.org/10.1109/TCST.2025.3589409","url":null,"abstract":"Model predictive control (MPC) is a powerful framework for optimal control of dynamical systems. However, MPC solvers suffer from a high computational burden that restricts their application to systems with low sampling frequencies. This issue is further amplified in nonlinear and constrained systems that require nesting MPC solvers within iterative procedures. In this brief, we address these issues by developing parallel-in-time algorithms for constrained nonlinear optimization problems that take advantage of massively parallel hardware to achieve logarithmic computational time scaling over the planning horizon. We develop time-parallel second-order solvers based on interior point (IP) methods and the alternating direction method of multipliers (ADMM), leveraging fast convergence and lower computational cost per iteration. The parallelization is based on a reformulation of the subproblems in terms of associative operations that can be parallelized using the associative scan algorithm. We validate our approach on numerical examples of nonlinear and constrained dynamical systems.","PeriodicalId":13103,"journal":{"name":"IEEE Transactions on Control Systems Technology","volume":"34 1","pages":"509-516"},"PeriodicalIF":3.9,"publicationDate":"2025-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11091448","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145915548","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This contribution presents a continuous integral sliding mode methodology for disturbed linear time-invariant systems with input and state asymmetric constraints affected by matched disturbances. This methodology preserves the nominal behavior of a system generated by a saturated nominal control law in finite time. It generates a bounded continuous control signal that satisfies the asymmetric constraints and compensates for the effects of matched bounded Lipschitz asymmetric disturbances in finite time. The proposed approach ensures that the state fulfils the constraints from the initial time. It is shown that when the initial conditions are outside the state constraints, the system solution converges to the constrained set. Simulations and experimental validation illustrate the applicability of the proposed approach.
{"title":"Continuous Integral Sliding Mode Control for Systems With State and Input Constraints","authors":"Rosalba Galván-Guerra;Juan Eduardo Velázquez-Velázquez;Leonid Fridman;Emanuel Ortiz-Ortiz","doi":"10.1109/TCST.2025.3588039","DOIUrl":"https://doi.org/10.1109/TCST.2025.3588039","url":null,"abstract":"This contribution presents a continuous integral sliding mode methodology for disturbed linear time-invariant systems with input and state asymmetric constraints affected by matched disturbances. This methodology preserves the nominal behavior of a system generated by a saturated nominal control law in finite time. It generates a bounded continuous control signal that satisfies the asymmetric constraints and compensates for the effects of matched bounded Lipschitz asymmetric disturbances in finite time. The proposed approach ensures that the state fulfils the constraints from the initial time. It is shown that when the initial conditions are outside the state constraints, the system solution converges to the constrained set. Simulations and experimental validation illustrate the applicability of the proposed approach.","PeriodicalId":13103,"journal":{"name":"IEEE Transactions on Control Systems Technology","volume":"34 1","pages":"502-508"},"PeriodicalIF":3.9,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145915591","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}
This work develops novel machine learning modeling and predictive control techniques for nonlinear chemical systems that experience asynchronous and delayed measurements that result in missingness in both offline and online data collections. Specifically, a phased long short-term memory (PLSTM) network is used to learn the process dynamics amidst the missingness in the data measurements, during the offline training process. The generalization performance of PLSTM is theoretically studied on the basis of statistical machine learning theory to better understand the capabilities of PLSTM models. The PLSTM model is subsequently employed to forecast the evolution of states for a Lyapunov-based model predictive control (LMPC). The proposed PLSTM-based LMPC is designed to account for data loss and delays in real-time implementation, and guarantees the closed-loop stability of nonlinear systems subjected to missing real-time data, provided that there is an upper bound on the number of consecutively missing real-time data. Finally, two chemical processes including an extractive dividing wall column (EDWC) and a continuous stirred tank reactor (CSTR) are used to demonstrate the effectiveness of PLSTM modeling and predictive control methods.
{"title":"Phased Long Short-Term Memory-Based Predictive Control of Chemical Processes With Asynchronous and Delayed Measurements","authors":"Wanlu Wu;Yujia Wang;Haohao Zhang;Ming-Qing Zhang;Min-Sen Chiu;Zhe Wu","doi":"10.1109/TCST.2025.3587908","DOIUrl":"https://doi.org/10.1109/TCST.2025.3587908","url":null,"abstract":"This work develops novel machine learning modeling and predictive control techniques for nonlinear chemical systems that experience asynchronous and delayed measurements that result in missingness in both offline and online data collections. Specifically, a phased long short-term memory (PLSTM) network is used to learn the process dynamics amidst the missingness in the data measurements, during the offline training process. The generalization performance of PLSTM is theoretically studied on the basis of statistical machine learning theory to better understand the capabilities of PLSTM models. The PLSTM model is subsequently employed to forecast the evolution of states for a Lyapunov-based model predictive control (LMPC). The proposed PLSTM-based LMPC is designed to account for data loss and delays in real-time implementation, and guarantees the closed-loop stability of nonlinear systems subjected to missing real-time data, provided that there is an upper bound on the number of consecutively missing real-time data. Finally, two chemical processes including an extractive dividing wall column (EDWC) and a continuous stirred tank reactor (CSTR) are used to demonstrate the effectiveness of PLSTM modeling and predictive control methods.","PeriodicalId":13103,"journal":{"name":"IEEE Transactions on Control Systems Technology","volume":"33 6","pages":"2439-2454"},"PeriodicalIF":3.9,"publicationDate":"2025-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11084944","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145339686","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：